1 INFO-VAX	Tue, 23 Sep 2003	Volume 2003 : Issue 527       Contents:0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way?0 Re: A flood of spams - another virus on the way? Alpha team bringing "3" cores!" Re: Alpha team bringing "3" cores!" Re: Alpha team bringing "3" cores! Re: Can VMS ZIP encrypt?$ Re: Creating a wide area VMS Cluster$ Re: Creating a wide area VMS Cluster$ Re: Creating a wide area VMS Cluster" Re: HELP: Using CLI with C program Re: How to order VMS licenses? Re: merging queue databases  Re: merging queue databases  Re: merging queue databases  More on Microsoft, Sun and SCO Re: Need quotes for a system) Re: ODS-5 or not ODS-5 What is the answer ) Re: ODS-5 or not ODS-5 What is the answer ) Re: ODS-5 or not ODS-5 What is the answer ' Oracle 8.1.7.4 on VMS 7.2-2 Performance + Re: Oracle 8.1.7.4 on VMS 7.2-2 Performance + Re: Oracle 8.1.7.4 on VMS 7.2-2 Performance + Re: Oracle 8.1.7.4 on VMS 7.2-2 Performance + Re: Oracle 8.1.7.4 on VMS 7.2-2 Performance " OT: Verisign's contract with ICANN& Re: OT: Verisign's contract with ICANN& Re: OT: Verisign's contract with ICANN& Re: OT: Verisign's contract with ICANN Re: P.C. Market share  PCI Device Access ( Peculiar (header-free) junk e-mail.  Eh?, Re: Peculiar (header-free) junk e-mail.  Eh?, Re: Peculiar (header-free) junk e-mail.  Eh?' Possible e-mail scam allegedly from HP. + Re: Possible e-mail scam allegedly from HP. + Re: Possible e-mail scam allegedly from HP. D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?F Re: Search all Files for any Lines containing specific set of Strings." Re: source for 431 pin CPU socket?# Re: Strange problem with DLT drives # Re: Strange problem with DLT drives 6 Re: strangeness with temporary text files used by MAIL6 Re: strangeness with temporary text files used by MAIL6 Re: strangeness with temporary text files used by MAIL. Re: TCPIP:DNS Dual interfaces and domain names. Re: TCPIP:DNS Dual interfaces and domain namesK Re: There is no Joy in Sunville -- mighty Billy has struck out [on his own] * Re: Trying to figure out license transfers Virtul Tape Driver Re: Virtul Tape Driver Re: Virtul Tape Driver Re: Virtul Tape Driver Re: Virtul Tape Driver Re: Virtul Tape Driver Re: Virtul Tape Driver VMS mentioned in Linux article" Re: VMS mentioned in Linux article" Re: VMS mentioned in Linux article" Re: VMS mentioned in Linux article" Re: VMS mentioned in Linux article" Re: VMS mentioned in Linux article" Re: VMS mentioned in Linux article Re: VMS Upgrade Needed Re: VMS Upgrade Needed8 Re: We can stop these SPAMS by blocking all the SPAMBOTs  F ----------------------------------------------------------------------  % Date: Mon, 22 Sep 2003 11:26:20 -0700 % From: Dean Woodward <deanw@rdrop.com> 9 Subject: Re: A flood of spams - another virus on the way? ( Message-ID: <3F6F3ECC.7000205@rdrop.com>   Paul Sture wrote:   H > One claim I have seen is that these things come in with block capitals3 > in the headers, so filtering on "SUBJECT:" works.   3 Quoting the last handful to land in my spam folder:   Q > FROM: "Microsoft Corporation Customer Bulletin" <ohkofbzx-bgpqxfvm@support.net> ( > TO: "MS Partner" <partner@support.net>* > SUBJECT: Latest Internet Critical Update --- 8 > FROM: "Microsoft" <njswjzvwhnwiy-lufd@advisor.msn.com>B > TO: "MS Corporation Partner" <cmygqasx-otuczsme@advisor.msn.com>& > SUBJECT: Last Internet Security Pack --- 2 > FROM: "Microsoft" <urzbhi-koizwxm@news.msdn.com>7 > TO: "Microsoft Corporation User" <user@news.msdn.com>  > SUBJECT: Last Update --- H > FROM: "microsoft internet storage system" <emailengine@rocketmail.net>$ > TO: " " <recipient@homeserver.com>" > SUBJECT: Mail Returned To Sender --- K > FROM: "Microsoft Network Message Storage Service" <zmailform@america.com> - > TO: "Network User" <recipient@mxserver.com> + > SUBJECT: undelivered message user unknown    Paul: * > I'm not seeing that here, using TCPware.  ' So, is TCPware cleaning up the headers?   H The filter that I'm catching these with, as someone suggested (and I've F always done), is to reject anything that doesn't have my email in the  To: or Cc: fields.   ------------------------------   Date: 22 Sep 03 20:37:02 +0200) From: p_sture@elias.decus.ch (Paul Sture) 9 Subject: Re: A flood of spams - another virus on the way? ) Message-ID: <yc+glvbv3l$$@elias.decus.ch>   P In article <3F6F3ECC.7000205@rdrop.com>, Dean Woodward <deanw@rdrop.com> writes: > Paul Sture wrote:  > I >> One claim I have seen is that these things come in with block capitals 4 >> in the headers, so filtering on "SUBJECT:" works. > 5 > Quoting the last handful to land in my spam folder:  > R >> FROM: "Microsoft Corporation Customer Bulletin" <ohkofbzx-bgpqxfvm@support.net>) >> TO: "MS Partner" <partner@support.net> + >> SUBJECT: Latest Internet Critical Update  > --- 9 >> FROM: "Microsoft" <njswjzvwhnwiy-lufd@advisor.msn.com> C >> TO: "MS Corporation Partner" <cmygqasx-otuczsme@advisor.msn.com> ' >> SUBJECT: Last Internet Security Pack  > --- 3 >> FROM: "Microsoft" <urzbhi-koizwxm@news.msdn.com> 8 >> TO: "Microsoft Corporation User" <user@news.msdn.com> >> SUBJECT: Last Update  > --- I >> FROM: "microsoft internet storage system" <emailengine@rocketmail.net> % >> TO: " " <recipient@homeserver.com> # >> SUBJECT: Mail Returned To Sender  > --- L >> FROM: "Microsoft Network Message Storage Service" <zmailform@america.com>. >> TO: "Network User" <recipient@mxserver.com>, >> SUBJECT: undelivered message user unknown >  > Paul: + >> I'm not seeing that here, using TCPware.  > ) > So, is TCPware cleaning up the headers?   ? We are using MX 5.3 here. Yes, it does appear to be doing that.   D I've never used MX before, and this is a shared system, so I am wary0 of doing anything which will affect other users.   > J > The filter that I'm catching these with, as someone suggested (and I've H > always done), is to reject anything that doesn't have my email in the  > To: or Cc: fields. >   E Time to grab the manuals. As if I didn't have anything else to do :-(    ------------------------------  % Date: Mon, 22 Sep 2003 11:45:59 -0700 0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com>9 Subject: Re: A flood of spams - another virus on the way? ' Message-ID: <3f6ee0f8$1@cpns1.saic.com>    Don Sykes wrote: >  > JF Mezei wrote:  [...] , >>typically, a SMTP session would look like:
 >>-----------  >>HELO <chocolate.com>! >>MAIL FROM: <chef@chocolate.com>  >>RCPT TO: <elves@cookies.com> >  > < > Yes. Right here I want say NO SUCH USER and drop the link.G > And I don't care if chef@chocolate.com is valid or not. If he's valid G > he'll most likely get the NO SUCH USER response. If he's fictional, I H > don't want to waste another nanosecond on him. Even if he IS valid andG > doesn't get the message because his ISP is down or something, I don't G > want to bog down my SMTP service trying over and over to reach him to I > tell him he used a bad address. If chef really wants to find out if the 2 > elves got the message, he can request a receipt.  G The MAIL FROM command is followed by one or more RCPT TO commands, one  H for each recipient of the message.  In your scenario, if there were two C recipients and one was bad, the good recipient would never see the  G message.  This is the reason mailers do not currently work the way you   are wishing them to.  I Now, after the RCTP TO command(s) comes the DATA command.  At this point  G it would be perfectly legitimate to return a "554 no valid recipients"  E status and refuse to accept the message.  In no case, however, is it  < acceptable within the SMTP protocol to simply drop the link.  
 Mark Berryman    ------------------------------  % Date: Mon, 22 Sep 2003 11:50:04 -0700 0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com>9 Subject: Re: A flood of spams - another virus on the way? ' Message-ID: <3f6ee1ec$1@cpns1.saic.com>    Dean Woodward wrote: [...]   ) > So, is TCPware cleaning up the headers?  > J > The filter that I'm catching these with, as someone suggested (and I've H > always done), is to reject anything that doesn't have my email in the  > To: or Cc: fields. >   C I hope you don't subscribe to any mailing lists then.  Mail from a  I mailing list will typically have the name of the mailing list in its To:  1 field, not the name of each individual recipient.   
 Mark Berryman    ------------------------------  # Date: Mon, 22 Sep 2003 17:10:24 GMT ' From: Don Sykes <anonymous@pacbell.net> 9 Subject: Re: A flood of spams - another virus on the way? + Message-ID: <3F6F2D84.C71B058F@pacbell.net>    david20@alpha2.mdx.ac.uk wrote:  > W > In article <3F6CABF0.BAD2DD4C@pacbell.net>, Don Sykes <anonymous@pacbell.net> writes:  > > H > >As I've been complaining about recently, I can't even get the HP SMTPI > >service to check incoming messgaes for a valid user during the initial I > >connection, which IIRC could be done in the 2nd step of the connection  > >process!  > > D > What do you mean by a valid user ? How can you check it is valid ?  F My SMTP service like all others listen on port 25 for mail directed toF my domain. When I get connection request I must respond with READY (orF something), they then say who they are and who the mail is directed toF (MAIL FROM: RCPT TO: ). Since my server is responsible for my domain's7 mail it should have a list of the domain's valid users. E While this doesn't stop invalid mails until they get to my domain, at  least I can drop them there.  M > Even before verisign's idiocy checking that the sender's domain existed was Q > extremely error prone because of the large number of misconfigured mail systems  > and DNSs. K > The other check which people used was an ident check - but again this was L > pretty useless - either the information was incorrect or the sender wasn't > running an ident daemon. > I > >This most recent onslaught of crap is just another example of what the A > >problem REALLY is - i.e. no check points along the email path.  > J > You would need to totally redesign TCPIP and SMTP with security in mind.I > To make it really work would probably involve mandatory digital signing  > of all mail.  @ This is my point. Something drastic needs to be done because theG situation is only going to get worse until there is a serious "handle". G Mandatory digital signing MAY be part of the solution. It's a big issue G and I'm not going to try to suggest all the technical details here. But , the current protocols are not doing the job.   > 
 > > In theG > >current SMTP model the only one who even has an opportunity to block - > >spam & assc viruses is the end receipient.  > M > You can block viruses at the mailhub level the problem is what you then do. G > Do you send the sender a message to let them know they are infected ? < > Do you clean the message and deliver it to the recipient ? > Do you delete the message ?  > P > With the latest viruses which are prolific, contain nothing of value and forgeM > the from address I'd argue the only sensible thing to do is to delete them. 4 > However that may be illegal in some jurisdictions.  H It can't be illegal for me to delete mail that's addressed to my domain. I think I need to clarify : ? There are 2 possible points of checking even using the existing H protocols. When a user sends a mail to his outbound mail server and whenC the mail arrives at the inbound mail server. These points are where E checking needs to be done. The outbound mail server could insure that E any mail it passes on is from a valid user (i.e. anyone authorized to F use that domain & server) and, optionally, check some (suspect, randomH or other method) emails for viruses. The receiving server needs to checkG that the recipient is a vaild user in that domain, or group of domains,  which it is serving.   > L > Spam is a worse problem. Unfortunately one persons Spam is another persons > important mail message.  > 
 > For example  > , > Mailhub receives a mail message for a user > " > Message contains phrases such as >  > You have won >  > Million Pounds >  > Please respond > ) > Typical spam right ?  So you delete it.  > M > What you don't know is this user has been gambling on say lottery.co.uk and ) > has just won their Million pound prize.  > O > Are they going to be happy when they find out the mailhub deleted that mail - L > especially if there is a time limit as to when they can claim there prize. > K > The best a central mailhub can do is either deliver the mail marked up in O > someway to say it looks like spam or quarantine it and let the user know they K > can release it otherwise it will be automatically deleted after a certain 	 > period.  > Q > The end user is the only one who can determine that any particular mail is SPAM  > as far as they are concerned.   H Your example demonstrates just how feeble the current methods of dealingE with spam are. Filtering a subject line is a poor way of dealing with E spam, regardless of where in the path you apply it, as I argue below.    >  > >This means that even if youI > >have a "good filter" on your email reader and don't ever "see" the bad G > >emails, an enormous amount of bandwidth is taken up on the internet, H > >because each piece of crap sent out gets the same treatment all alongJ > >the way to the destination user. Only then do we get a chance to ignoreJ > >it. So who's at fault? Our own industry for embracing tcpip/smtp as theJ > >holly grail in its original form - i.e. no forced checkpoints. As I seeJ > >it any ISP that is authorized to hand out an IP address s/b responsibleI > >for its misuse. They should at minimum be required to check the source G > >of all email to be sure it's valid and has not been spoofed. Further E > >each of the ISP's customers should have to register an approximate I > >number of emails they will be sending out in any one day. Then if they F > >grossly exceeded that, the initial ISP router should reject furtherF > >emails and immediately inform their customer of the action. Any ISPH > >failing to do this should have their IP addresses revoked or put on aF > >lookaside list of all legit routers and not route emails from them. > > C > >Granted this is a half-baked idea at this point, but if WE, as a I > >community, are ever going to stop this madness, we're going to have to I > >come up with a technical solution at a fundamental, routing level; not E > >just add more and different filters for the end user to implement.  > >  > L > For this too work it would have to be applied to every source of smtp mailO > worldwide - it's just not going to happen. Despite blacklists there are still 4 > tons and tons of open-relays for a spammer to use.  D It's going to HAVE TO happen sooner or later. Consider the bandwidthE this crap is taking up, let alone the thousands of man hours it takes  just delete this stuff. E The answer may be to abandon Simple Mail Transfer Protocol for Secure F Mail Transfer Protocol. Same acronym - different methods. Part of thisC new protocol may be that you don't "relay" anything, unless it's to @ addresses in other domains that are within your area of control.   --     Have VMS, Will Travel  Wire paladin, San Francisco    (paladinATalphaseDOTcom)   <snip>   ------------------------------  % Date: Mon, 22 Sep 2003 12:53:16 -0700 % From: Dean Woodward <deanw@rdrop.com> 9 Subject: Re: A flood of spams - another virus on the way? ( Message-ID: <3F6F532C.6040601@rdrop.com>   Mark Berryman wrote:   > Dean Woodward wrote: > [...]  > F >> The filter that I'm catching these with, as someone suggested (and J >> I've always done), is to reject anything that doesn't have my email in  >> the To: or Cc: fields.  >>E > I hope you don't subscribe to any mailing lists then.  Mail from a  K > mailing list will typically have the name of the mailing list in its To:  3 > field, not the name of each individual recipient.   G What, like Info-VAX? I subscribe to a round dozen or so, each of which  A has it's own rule to sort that traffic into it's own folder. The  + spam-trap rule is the last one in the list.    ------------------------------   Date: 22 Sep 2003 20:01:02 GMT, From: bill@gw5.cs.uofs.edu (Bill Gunshannon)9 Subject: Re: A flood of spams - another virus on the way? 9 Message-ID: <bknkdt$3o2sn$1@ID-135708.news.uni-berlin.de>   + In article <3F6F2D84.C71B058F@pacbell.net>, * 	Don Sykes <anonymous@pacbell.net> writes: > B > This is my point. Something drastic needs to be done because theI > situation is only going to get worse until there is a serious "handle". I > Mandatory digital signing MAY be part of the solution. It's a big issue I > and I'm not going to try to suggest all the technical details here. But . > the current protocols are not doing the job.  G It has never and likely will never be possible to solve social problems H with technical solutions.  As long as the cost is minimal and punishment. pretty much non-existant SPAM will continue.     > J > It can't be illegal for me to delete mail that's addressed to my domain. > I think I need to clarify : A > There are 2 possible points of checking even using the existing J > protocols. When a user sends a mail to his outbound mail server and whenE > the mail arrives at the inbound mail server. These points are where G > checking needs to be done. The outbound mail server could insure that G > any mail it passes on is from a valid user (i.e. anyone authorized to  > use that domain & server)   G The senders of SPAM are frequently using machines specifically designed G to do precisely that.  How do you force them to verify users when their ( whole intent is to forge the originator?  I >                            and, optionally, check some (suspect, random ' > or other method) emails for viruses.    F Most viruses are not sent (originated) by servers, they are originatedE by individual machines, usually running MS Outhouse and the users are E not even aware they are infected.  When forced to go through a server F to send email they frequently are scanned, but many sites allow (as is: their priveledge) anyone to send outgoing data to port 25.  J >                                      The receiving server needs to checkI > that the recipient is a vaild user in that domain, or group of domains,  > which it is serving.  H In a system where all mail originally comes into one machine and is thenF parcelled out (I assume there still are very large sites that do this) that might well be impossible.  G Personally, I would like to see the INTERNET split into two parts.  One H like it is today where anything goes and another like it used to be withH extremely strong AUP's and no forgiveness.  Commercializing the INTERNETJ was the worst mistake made in the computer industry.  It took an extremelyI useful tool and made it more of a "vast wasteland" than even TV.  I think = it is clear which of the two above I would choose to work on.    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Mon, 22 Sep 2003 16:20:25 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>9 Subject: Re: A flood of spams - another virus on the way? ) Message-ID: <3F6F5985.9F933388@istop.com>    Don Sykes wrote:H > my domain. When I get connection request I must respond with READY (orH > something), they then say who they are and who the mail is directed toH > (MAIL FROM: RCPT TO: ). Since my server is responsible for my domain's9 > mail it should have a list of the domain's valid users.   N *should* is the keyword. The problem is that you then need a central directoryM of all valid userna,es in your domain, and such a directory must be available I to the SMTO server in charge of receiving all mail on behalf of a dokain.   L Consider a case of having 20 nodes in a domain and all mail arrives from theK outside through Node1. Node1 then distributes the messages to one of the 20 H nodes. But in the case of VMSmail, there is no central directory used byN VMSmail, each node has its own private vmsmail_profile.data, and even at that,= there are cases where there are valid users with no record in @ vmsmail_profile.data (if they have never received an email yet).  L Consider also the aspect of the SMTP receiver: it is used no only to receiveH mail from the outside world, but also from other nodes in your intranet,N including PC/MAC clients. And those emails are no necessarily destined for theG intranet, so the receiver must have smarts to apply the verification of 8 username to only destinations that belon to that domain.  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Month  x--   UNLIMITED DOWNLOAD   ------------------------------  % Date: Mon, 22 Sep 2003 16:06:59 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>9 Subject: Re: A flood of spams - another virus on the way? ) Message-ID: <3F6F5661.5EBDD33D@istop.com>   ! VAXman-, @SendSpamHere.ORG wrote: I > Oh!  I had a 12 incher with a magnifying cursor and my name engraved on J > its end frame plates.  I'll have to ask my wife if 12" turned her on. ;)  I Interestingly, I think it was when mobile phones were introduced that the J whole pragma of "bigger is better" changed and then males started to brag 7 about how small their equipment (phones, pdas etc) was.   2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Month  x--   UNLIMITED DOWNLOAD   ------------------------------  % Date: Mon, 22 Sep 2003 16:41:26 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>9 Subject: Re: A flood of spams - another virus on the way? ) Message-ID: <3F6F5E72.55DC209C@istop.com>    Dean Woodward wrote:5 > Quoting the last handful to land in my spam folder:  > S > > FROM: "Microsoft Corporation Customer Bulletin" <ohkofbzx-bgpqxfvm@support.net> * > > TO: "MS Partner" <partner@support.net>, > > SUBJECT: Latest Internet Critical Update  H In the 8 messages I have received since the start of the virus (all veryM early, none in recent days), only the SUBJECT was uppercase, the other fields I were normal. Perhaps there are 2 strains to the virus, the first of which  didn't survive/propate much ?   2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Month  x--   UNLIMITED DOWNLOAD   ------------------------------  % Date: Mon, 22 Sep 2003 13:40:45 -0700 % From: Dean Woodward <deanw@rdrop.com> 9 Subject: Re: A flood of spams - another virus on the way? ( Message-ID: <3F6F5E4D.5020901@rdrop.com>   JF Mezei wrote:  > Dean Woodward wrote: > 5 >>Quoting the last handful to land in my spam folder:  >>2 >>>FROM: "Microsoft Corporation Customer Bulletin") >>>TO: "MS Partner" <partner@support.net> + >>>SUBJECT: Latest Internet Critical Update  > J > In the 8 messages I have received since the start of the virus (all veryO > early, none in recent days), only the SUBJECT was uppercase, the other fields K > were normal. Perhaps there are 2 strains to the virus, the first of which  > didn't survive/propate much ?   ; Or some mail handler is "fixing" the From: and To: headers.    ------------------------------  # Date: Mon, 22 Sep 2003 20:49:13 GMT ' From: Don Sykes <anonymous@pacbell.net> 9 Subject: Re: A flood of spams - another virus on the way? + Message-ID: <3F6F60CD.E481B98F@pacbell.net>    Bill Gunshannon wrote: > - > In article <3F6F2D84.C71B058F@pacbell.net>, 3 >         Don Sykes <anonymous@pacbell.net> writes:  > > D > > This is my point. Something drastic needs to be done because theK > > situation is only going to get worse until there is a serious "handle". K > > Mandatory digital signing MAY be part of the solution. It's a big issue K > > and I'm not going to try to suggest all the technical details here. But 0 > > the current protocols are not doing the job. > I > It has never and likely will never be possible to solve social problems J > with technical solutions.  As long as the cost is minimal and punishment. > pretty much non-existant SPAM will continue.  E That may in fact be the main problem - the cost is virtually free! If E there were a cost to send each email - one which could be refunded by G the recipient if it was mail they wanted - I bet spams would nose dive. B Even at .01 / email I'm sure it wouldn't be worth it. I can't evenE figure how spam senders make any money as it is. Who responds to this  stuff?   >  > > L > > It can't be illegal for me to delete mail that's addressed to my domain. > > I think I need to clarify : C > > There are 2 possible points of checking even using the existing L > > protocols. When a user sends a mail to his outbound mail server and whenG > > the mail arrives at the inbound mail server. These points are where I > > checking needs to be done. The outbound mail server could insure that I > > any mail it passes on is from a valid user (i.e. anyone authorized to  > > use that domain & server)  > I > The senders of SPAM are frequently using machines specifically designed I > to do precisely that.  How do you force them to verify users when their * > whole intent is to forge the originator?  D I don't know. Maybe by a confirmation sent back by the outgoing mailH server asking if they just sent a message... it will certainly take some thought.   > K > >                            and, optionally, check some (suspect, random ( > > or other method) emails for viruses. > H > Most viruses are not sent (originated) by servers, they are originatedG > by individual machines, usually running MS Outhouse and the users are % > not even aware they are infected.     F This issue is really sinister and, I hope, illegal in most places. TheG most immediate idea that comes to mind is a class action suit againt MS ; for providing such holes. Maybe it's already being pursued.   $ > When forced to go through a serverH > to send email they frequently are scanned, but many sites allow (as is< > their priveledge) anyone to send outgoing data to port 25. > L > >                                      The receiving server needs to checkK > > that the recipient is a vaild user in that domain, or group of domains,  > > which it is serving. > J > In a system where all mail originally comes into one machine and is thenH > parcelled out (I assume there still are very large sites that do this)  > that might well be impossible. > I > Personally, I would like to see the INTERNET split into two parts.  One J > like it is today where anything goes and another like it used to be withJ > extremely strong AUP's and no forgiveness.  Commercializing the INTERNETL > was the worst mistake made in the computer industry.  It took an extremelyK > useful tool and made it more of a "vast wasteland" than even TV.  I think ? > it is clear which of the two above I would choose to work on.  >   F COUNT ME IN!!! I would even be willing to pay a little more for such a service - ISP's take note.   > bill >  > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   --     Have VMS, Will Travel  Wire paladin, San Francisco    (paladinATalphaseDOTcom)   ------------------------------  % Date: Mon, 22 Sep 2003 14:34:32 -0700 % From: Dean Woodward <deanw@rdrop.com> 9 Subject: Re: A flood of spams - another virus on the way? ( Message-ID: <3F6F6AE8.3090409@rdrop.com>   Don Sykes wrote:  D > Even at .01 / email I'm sure it wouldn't be worth it. I can't evenG > figure how spam senders make any money as it is. Who responds to this  > stuff?  I It's possible you're looking at it from the wrong angle. The money isn't  7 in selling things via spam, it's in selling spam tools.    ------------------------------  % Date: Mon, 22 Sep 2003 18:08:29 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>9 Subject: Re: A flood of spams - another virus on the way? ) Message-ID: <3F6F72D2.BC1DEDA9@istop.com>    Dean Woodward wrote:J > It's possible you're looking at it from the wrong angle. The money isn't9 > in selling things via spam, it's in selling spam tools.   K Or in generating visits to advertising included in the html sent to spammed K parties. When a microsoft use opens an email written in HTML, any <img tags I inside the HTML get resolved and result in the microsoft user fetching an H advertising. This generates revenus to the spammers who get paid for the; number of times the advertising is fetched by mail clients.   2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Month  x--   UNLIMITED DOWNLOAD   ------------------------------  ( Date: Mon, 22 Sep 2003 23:08 +0100 (BST)9 From: duncan@macdonald.compulink.co.uk (Duncan Macdonald) 9 Subject: Re: A flood of spams - another virus on the way? A Message-ID: <memo.20030922230837.1840A@macdonald.compulink.co.uk>   K As most SPAM messages seem to have a lot of recipients, perhaps one way to  P reduce the flood would be for mail servers to count the number of RCPT TO lines 9 and discard any messages for more than say 20 recipients. F (According to RFC 821 error code 421 "Service not available - closing N transmission channel" may be given to any command - even in the middle of the K RCPT TO list, and this would seem to be the right thing to do to spammers.)    ------------------------------   Date: 23 Sep 03 00:17:55 +0200) From: p_sture@elias.decus.ch (Paul Sture) 9 Subject: Re: A flood of spams - another virus on the way? ) Message-ID: <1olQBnmZ$+Df@elias.decus.ch>   U In article <3F6F60CD.E481B98F@pacbell.net>, Don Sykes <anonymous@pacbell.net> writes:  >  >  > Bill Gunshannon wrote: >>  . >> In article <3F6F2D84.C71B058F@pacbell.net>,4 >>         Don Sykes <anonymous@pacbell.net> writes: >> >E >> > This is my point. Something drastic needs to be done because the L >> > situation is only going to get worse until there is a serious "handle".L >> > Mandatory digital signing MAY be part of the solution. It's a big issueL >> > and I'm not going to try to suggest all the technical details here. But1 >> > the current protocols are not doing the job.  >>  J >> It has never and likely will never be possible to solve social problemsK >> with technical solutions.  As long as the cost is minimal and punishment / >> pretty much non-existant SPAM will continue.  > G > That may in fact be the main problem - the cost is virtually free! If G > there were a cost to send each email - one which could be refunded byuI > the recipient if it was mail they wanted - I bet spams would nose dive.eD > Even at .01 / email I'm sure it wouldn't be worth it. I can't evenG > figure how spam senders make any money as it is. Who responds to thisa > stuff? >   F But one of the problems is the other day I received a spam advertising5 28.5 million email addresses for something like $100.    ------------------------------   Date: 22 Sep 2003 22:49:24 GMT, From: bill@gw5.cs.uofs.edu (Bill Gunshannon)9 Subject: Re: A flood of spams - another virus on the way?f9 Message-ID: <bknu9j$3t1ju$1@ID-135708.news.uni-berlin.de>a  + In article <3F6F60CD.E481B98F@pacbell.net>,-* 	Don Sykes <anonymous@pacbell.net> writes: >  >  > Bill Gunshannon wrote: >>  J >> It has never and likely will never be possible to solve social problemsK >> with technical solutions.  As long as the cost is minimal and punishmento/ >> pretty much non-existant SPAM will continue.o > G > That may in fact be the main problem - the cost is virtually free! IfoG > there were a cost to send each email - one which could be refunded by I > the recipient if it was mail they wanted - I bet spams would nose dive.oD > Even at .01 / email I'm sure it wouldn't be worth it. I can't evenG > figure how spam senders make any money as it is. Who responds to thise > stuff?  H Sadly, if only one idiot responds then the return has exceeded the cost.J I have often wondered at the general intelligence of the SPAMers.  KnowingI that America is famous for being a mono-lingual culture, just who do they K think is going to read all that junk in Spanish, Portugese and worse still,eI in big5!!  I am multi-lingual and even I don't have a mail program (I uset9 Pine which makes me virus safe) that can comprehend big5.t   >> eJ >> The senders of SPAM are frequently using machines specifically designedJ >> to do precisely that.  How do you force them to verify users when their+ >> whole intent is to forge the originator?1 > F > I don't know. Maybe by a confirmation sent back by the outgoing mailJ > server asking if they just sent a message... it will certainly take some
 > thought.  G That would increase the amount of traffic on the net thus making one ofrF the facets of the problem worse.  And, considering that the purpose ofG SPAM is to send all this junk, what would keep server from just sendingm back an ACK.   >>  I >> Most viruses are not sent (originated) by servers, they are originatedgH >> by individual machines, usually running MS Outhouse and the users are& >> not even aware they are infected.   > E > This issue is really sinister and, I hope, illegal in most places. m  > OK, I'm baffled.  Under what possible law could it be illegal?  I >                                                                     ThetI > most immediate idea that comes to mind is a class action suit againt MS-= > for providing such holes. Maybe it's already being pursued.t  I No lawyer worth a damn would take the case, even if it had merit.  First,aG the Federal Govt. was unsuccessful at reigning in MS and second you areeH back to the illegal baseball bat law (just because you can break the law0 with something doesn't make the object illegal.)   > % >> When forced to go through a serverrI >> to send email they frequently are scanned, but many sites allow (as isg= >> their priveledge) anyone to send outgoing data to port 25.  >>  M >> >                                      The receiving server needs to checkgL >> > that the recipient is a vaild user in that domain, or group of domains, >> > which it is serving.N >> lK >> In a system where all mail originally comes into one machine and is then I >> parcelled out (I assume there still are very large sites that do this)i! >> that might well be impossible.r >>  J >> Personally, I would like to see the INTERNET split into two parts.  OneK >> like it is today where anything goes and another like it used to be witheK >> extremely strong AUP's and no forgiveness.  Commercializing the INTERNET M >> was the worst mistake made in the computer industry.  It took an extremelyeL >> useful tool and made it more of a "vast wasteland" than even TV.  I think@ >> it is clear which of the two above I would choose to work on. >> h > H > COUNT ME IN!!! I would even be willing to pay a little more for such a > service - ISP's take note.  1E Sadly, I think it unlikely to occur.  Hmmmm....  I wonder if we couldfC revive comp.mail.maps and bring USENET back using the INTERNET only0C for the transport medium.  And then just refuse email and news fromaL those that were not members.  Could give a whole new meaning to the UDP. :-)   bill   -- sJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   R   ------------------------------   Date: 22 Sep 2003 22:54:12 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)9 Subject: Re: A flood of spams - another virus on the way? 9 Message-ID: <bknuik$3t1ju$2@ID-135708.news.uni-berlin.de>c  A In article <memo.20030922230837.1840A@macdonald.compulink.co.uk>,e< 	duncan@macdonald.compulink.co.uk (Duncan Macdonald) writes:M > As most SPAM messages seem to have a lot of recipients, perhaps one way to nR > reduce the flood would be for mail servers to count the number of RCPT TO lines ; > and discard any messages for more than say 20 recipients. H > (According to RFC 821 error code 421 "Service not available - closing P > transmission channel" may be given to any command - even in the middle of the M > RCPT TO list, and this would seem to be the right thing to do to spammers.)s  H How long do you think it would take before they just stated breaking theK lists down into smaller chunks??  Cost is the same to them.  Example:  NewstF systems started limiting the number of allowed cross posts so the NewsG spammer started sending out their messages with one alt address and onetG Big8 address.  Got wide dissemination and beat the cross-posting limit.:  F Like I said, you can't solve social problems with technical solutions.G You need strong, enforced social rules to control anti-social behavior.:D Failure to obey needs to result in being cut off from the technology	 entirely.:   bill   -- pJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   P   ------------------------------    Date: 22 Sep 2003 17:43:20 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)<9 Subject: Re: A flood of spams - another virus on the way?o3 Message-ID: <R05no3UxjZkE@eisner.encompasserve.org>   } In article <memo.20030922230837.1840A@macdonald.compulink.co.uk>, duncan@macdonald.compulink.co.uk (Duncan Macdonald) writes:SM > As most SPAM messages seem to have a lot of recipients, perhaps one way to tR > reduce the flood would be for mail servers to count the number of RCPT TO lines ; > and discard any messages for more than say 20 recipients. H > (According to RFC 821 error code 421 "Service not available - closing P > transmission channel" may be given to any command - even in the middle of the M > RCPT TO list, and this would seem to be the right thing to do to spammers.)   G No, the right way to treat spammers is to keep them hanging on the line'. as long as possible, tying up their resources.   ------------------------------  # Date: Mon, 22 Sep 2003 22:49:34 GMTi' From: Don Sykes <anonymous@pacbell.net>t9 Subject: Re: A flood of spams - another virus on the way? + Message-ID: <3F6F7D04.AE8599B0@pacbell.net>e   Dean Woodward wrote: >  > Don Sykes wrote: > F > > Even at .01 / email I'm sure it wouldn't be worth it. I can't evenI > > figure how spam senders make any money as it is. Who responds to this 
 > > stuff? > J > It's possible you're looking at it from the wrong angle. The money isn't9 > in selling things via spam, it's in selling spam tools.a  A Sort of like MLM heh? You don't really make any money selling the : product, but rather by selling someone ELSE on selling the product....UGH!e  G I think I'll just cash in on part of that $28M of Nigerian money I keepS hearing about:):)s   -- n   Have VMS, Will Travela Wire paladin, San Francisco.   (paladinATalphaseDOTcom)   ------------------------------  % Date: Tue, 23 Sep 2003 01:33:22 -0400a* From: JF Mezei <jfmezei.spamnot@istop.com>9 Subject: Re: A flood of spams - another virus on the way? ) Message-ID: <3F6FDAFA.933A0F11@istop.com>    Don Sykes wrote:D > bat company. The situation with MS products is more like cigaretteJ > manufactures, which HAVE been sued successfully. MS is selling a productJ > that is, ostensibly, designed to provide some level of security. I thinkG > the argument could easily be made that they have been so negligent inuC > that area that, like cigarettes, even if their product is used asT > intended, it can cause harm. m  J Or we could compare MS to a butcher who knowingly sells meat that has beenM prepared in unsanitary conditions but is packaged in such flashy colours thate it sells meat by the tonne.   L When people get sick, MS simply accuse them of not cooking their meat enough  and declines any responsability.  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Montha x--   UNLIMITED DOWNLOAD   ------------------------------  # Date: Tue, 23 Sep 2003 05:05:40 GMTi' From: Don Sykes <anonymous@pacbell.net>e9 Subject: Re: A flood of spams - another virus on the way?h+ Message-ID: <3F6FD52F.946DF84C@pacbell.net>    Bill Gunshannon wrote: > - > In article <3F6F60CD.E481B98F@pacbell.net>,M3 >         Don Sykes <anonymous@pacbell.net> writes:m > >t > >i > > Bill Gunshannon wrote: > >>L > >> It has never and likely will never be possible to solve social problemsM > >> with technical solutions.  As long as the cost is minimal and punishmento1 > >> pretty much non-existant SPAM will continue.n > > I > > That may in fact be the main problem - the cost is virtually free! If@I > > there were a cost to send each email - one which could be refunded by K > > the recipient if it was mail they wanted - I bet spams would nose dive. F > > Even at .01 / email I'm sure it wouldn't be worth it. I can't evenI > > figure how spam senders make any money as it is. Who responds to this 
 > > stuff? > J > Sadly, if only one idiot responds then the return has exceeded the cost.L > I have often wondered at the general intelligence of the SPAMers.  KnowingK > that America is famous for being a mono-lingual culture, just who do theyrM > think is going to read all that junk in Spanish, Portugese and worse still,dK > in big5!!  I am multi-lingual and even I don't have a mail program (I useo; > Pine which makes me virus safe) that can comprehend big5.a  
 What's big 5?    >  > >>L > >> The senders of SPAM are frequently using machines specifically designedL > >> to do precisely that.  How do you force them to verify users when their- > >> whole intent is to forge the originator?s > >tH > > I don't know. Maybe by a confirmation sent back by the outgoing mailL > > server asking if they just sent a message... it will certainly take some > > thought. > I > That would increase the amount of traffic on the net thus making one of H > the facets of the problem worse.  And, considering that the purpose ofI > SPAM is to send all this junk, what would keep server from just sending  > back an ACK. >  > >>K > >> Most viruses are not sent (originated) by servers, they are originatedwJ > >> by individual machines, usually running MS Outhouse and the users are& > >> not even aware they are infected. > >bF > > This issue is really sinister and, I hope, illegal in most places. > @ > OK, I'm baffled.  Under what possible law could it be illegal?  H Breaking and entering maybe. This is no different than a thief who couldG figure out a way to steal my electric and have me pay for it. InfectingnD a system with software that steals my CPU cycles is exactly the sameB thing, as far as I'm concerned and s/b prosecuted in the same way.   > K > >                                                                     ThesK > > most immediate idea that comes to mind is a class action suit againt MSn? > > for providing such holes. Maybe it's already being pursued.h > K > No lawyer worth a damn would take the case, even if it had merit.  First,fI > the Federal Govt. was unsuccessful at reigning in MS and second you areoJ > back to the illegal baseball bat law (just because you can break the law2 > with something doesn't make the object illegal.)  H Not at all. Baseball bats are sold with the intention of the buyer usingG them to play baseball. If they are misused, that's not the fault of theaB bat company. The situation with MS products is more like cigaretteH manufactures, which HAVE been sued successfully. MS is selling a productH that is, ostensibly, designed to provide some level of security. I thinkE the argument could easily be made that they have been so negligent in,A that area that, like cigarettes, even if their product is used asoF intended, it can cause harm. The fact that they're not actually makingF the viruses is not significant, if they knowingly provided a bed where* such viruses and worms are known to breed.   >  > >o' > >> When forced to go through a serveraK > >> to send email they frequently are scanned, but many sites allow (as is ? > >> their priveledge) anyone to send outgoing data to port 25.  > >>O > >> >                                      The receiving server needs to checktN > >> > that the recipient is a vaild user in that domain, or group of domains, > >> > which it is serving.M > >>M > >> In a system where all mail originally comes into one machine and is then K > >> parcelled out (I assume there still are very large sites that do this)a# > >> that might well be impossible.a > >>L > >> Personally, I would like to see the INTERNET split into two parts.  OneM > >> like it is today where anything goes and another like it used to be withtM > >> extremely strong AUP's and no forgiveness.  Commercializing the INTERNET O > >> was the worst mistake made in the computer industry.  It took an extremely N > >> useful tool and made it more of a "vast wasteland" than even TV.  I thinkB > >> it is clear which of the two above I would choose to work on. > >> > > J > > COUNT ME IN!!! I would even be willing to pay a little more for such a > > service - ISP's take note. > G > Sadly, I think it unlikely to occur.  Hmmmm....  I wonder if we couldtE > revive comp.mail.maps and bring USENET back using the INTERNET onlyeE > for the transport medium.  And then just refuse email and news from N > those that were not members.  Could give a whole new meaning to the UDP. :-) >  > bill >  > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   -- h   Have VMS, Will Travele Wire paladin, San Francisco    (paladinATalphaseDOTcom)   ------------------------------  # Date: Tue, 23 Sep 2003 05:31:52 GMTo' From: Don Sykes <anonymous@pacbell.net>n9 Subject: Re: A flood of spams - another virus on the way?e+ Message-ID: <3F6FDB53.DCB7FA5D@pacbell.net>t   Bill Gunshannon wrote: > C > In article <memo.20030922230837.1840A@macdonald.compulink.co.uk>,.E >         duncan@macdonald.compulink.co.uk (Duncan Macdonald) writes:oN > > As most SPAM messages seem to have a lot of recipients, perhaps one way toS > > reduce the flood would be for mail servers to count the number of RCPT TO linesm= > > and discard any messages for more than say 20 recipients.eI > > (According to RFC 821 error code 421 "Service not available - closing Q > > transmission channel" may be given to any command - even in the middle of the O > > RCPT TO list, and this would seem to be the right thing to do to spammers.)  > J > How long do you think it would take before they just stated breaking theM > lists down into smaller chunks??  Cost is the same to them.  Example:  News H > systems started limiting the number of allowed cross posts so the NewsI > spammer started sending out their messages with one alt address and one I > Big8 address.  Got wide dissemination and beat the cross-posting limit.a > H > Like I said, you can't solve social problems with technical solutions.I > You need strong, enforced social rules to control anti-social behavior.dF > Failure to obey needs to result in being cut off from the technology > entirely.p  5 And how would you do that with no technical solution?i  A Technical solutions are like laws. They do not guarantee that the1D actions they protect against won't be done anyway. What they do is :F 1) make it clear to the citizenry, what actions are/are not acceptableD 2) prescribe appropriate punishments to those who ignore those rules  1 Technical solutions can and should provide this. mH Social rules aren't much good with those folks who don't give a rats assF about what you, or I, say about them. In the good old days most people; prized their reputations and avoided shame. Today you get alD multi-million dollar book deal for wrong doing. There seems to be noA distinction between the famous and the infamous... but I digress.e   -- A   Have VMS, Will Travel  Wire paladin, San Francisco    (paladinATalphaseDOTcom)   ------------------------------    Date: 22 Sep 2003 13:40:20 -0700( From: bob@instantwhip.com (Bob Ceculski)' Subject: Alpha team bringing "3" cores! = Message-ID: <d7791aa1.0309221240.767348e8@posting.google.com>o  ) http://www.theinquirer.net/?article=11661s   ------------------------------    Date: 22 Sep 2003 15:45:11 -0500+ From: young_r@encompasserve.org (Rob Young)<+ Subject: Re: Alpha team bringing "3" cores! 3 Message-ID: <m$xxPlfnzoxL@eisner.encompasserve.org>t  h In article <d7791aa1.0309221240.767348e8@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:+ > http://www.theinquirer.net/?article=11661   	 	8 - Bob:o  ) http://www.theinquirer.net/?article=11685e   				Rob    ------------------------------  % Date: Mon, 22 Sep 2003 17:00:32 -0400k$ From: "rob kas" <news@paychoice.com>+ Subject: Re: Alpha team bringing "3" cores! / Message-ID: <vmupdks66egsb5@corp.supernews.com>y  5 "Bob Ceculski" <bob@instantwhip.com> wrote in messaget7 news:d7791aa1.0309221240.767348e8@posting.google.com...d+ > http://www.theinquirer.net/?article=11661r           One will anyone buy them          Two can anyone afford them  1       Three will any VMS Applications run on theme  	       and   I      Four   If Yes to One ,Two and Three  will anyone know they exist andi run VMS ApplicationT    >                                                            Rob   ------------------------------  % Date: Mon, 22 Sep 2003 17:06:17 -0500r- From: Hunter Goatley <goathunter@goatley.com>m! Subject: Re: Can VMS ZIP encrypt?d8 Message-ID: <vmKbb.10698$jO.7951@bignews3.bellsouth.net>   David Gray wrote:eH > Yeah I got ZIPCLOAK but was hopping to do the ZIP/Encrypt  ZIP/Decrypt > all in one hit.  p > 8 You can.  But you need the ZCRYPT23 source module to add; on top of it.  I couldn't add it to my distribution becausen; doing so would have violated the U.S. export laws.  I think 6 those restrictions have expired now, but I'm not sure.  6 In any case, you can find a pointer to ZCRYPT from the, Info-ZIP web page:  http://www.info-zip.org/  : And my distribution, with the CLI version, can be found on3 ftp.process.com in [.VMS-FREEWARE.FILESERV]ZIP.ZIP.s   Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/ 6 goathunter@goatley.com, http://www.goatley.com/hunter/9 Check out PreciseMail Anti-Spam Gateway @ www.process.comh   ------------------------------  # Date: Mon, 22 Sep 2003 18:44:35 GMT)# From: "John Smith" <a@nonymous.com>e- Subject: Re: Creating a wide area VMS ClustertI Message-ID: <nqHbb.23103$Lnr1.16350@news01.bloor.is.net.cable.rogers.com>k   Paul Repacholi wrote:r; > moroney@world.std.spaamtrap.com (Michael Moroney) writes:N >N3 >> Lyndon Bartels <lbartels@pressenter.com> writes:r > 1 >>> I'm thinking of building a wide area cluster.o > E >>> My goal is to provide a disaster tolerant cluster for both OS anda	 >>> data.d >o/ >>> My current plan has three sites, A, B, & C.  > ? >> Your initial ideas look good, but you'll want to talk to thec? >> Disaster Tolerant Cluster people at HP.  There are a few bigrD >> "gotchas" you have to watch out for. One is the following: Site AC >> goes down.  Site B stays up for a while.  Site B then goes down. C >> Site A comes back up.  Guess what - Site A is running with _old_ % >> data and you've lost transactions!  >uB > True story. Site had at least 3 redundant fibres, with 2 telcos.E > Backhoe get one. Then a hill falls over, I ma not kidding, and thatiD > take out the fibre cable that has BOTH remaining circuits :( SeemsC > telcos have this sharing and caring thing. For themselves anyway.s >eA > Telcos seem to like money by the truckload to dig a second holed@ > and force Murphy and backhoes to work harder. LOTS more money.D > Droping in microwave links to link up all 3 sites that way as wellE > will allow it to all fail over to MSCP over the air when the yelloweD > horror out on the road digs up your fibres. Mind, he will possibly0 > take out you power as well. Life is like that. >s5 > Old 10Mb microwaves should be pretty cheap nowdays.t     www.proxim.com6 and look for the Tsunami product line of wireless kit.  H With the right set of antennae at each end, and sufficient tower height,H these can reach out 40 miles at 100Mbps. I know a company that does this! between islands in the Caribbean.t   ------------------------------  % Date: Mon, 22 Sep 2003 17:06:38 -0400M* From: JF Mezei <jfmezei.spamnot@istop.com>- Subject: Re: Creating a wide area VMS Clusterr) Message-ID: <3F6F6458.DCDF20F4@istop.com>c   > someone wrote:G > > will allow it to all fail over to MSCP over the air when the yellowh0 > > horror out on the road digs up your fibres.     M Since backhoes seems to be used solely to cut telecom wires, it is a surprisewN that the US government hasn't declared all such vehicles to be terrorist tools that should be banned ?0  J Remember when they got paranoid about cropduster planes ? Seems to be thatI backhoes far much more damage/headaches in the USA than cropdusterplanes.p  K Imagine a fleet of backhoes in manhattan, digging through a large number ofe" cables all at the same time... :-)  & Do backhoes have any legitimate uses ?   ------------------------------  # Date: Tue, 23 Sep 2003 03:51:04 GMTo# From: "John Smith" <a@nonymous.com>e- Subject: Re: Creating a wide area VMS ClustereH Message-ID: <IqPbb.25193$Lnr1.6194@news01.bloor.is.net.cable.rogers.com>   JF Mezei wrote:e >> someone wrote:dG >>> will allow it to all fail over to MSCP over the air when the yellow,/ >>> horror out on the road digs up your fibres./ >. >PF > Since backhoes seems to be used solely to cut telecom wires, it is aF > surprise that the US government hasn't declared all such vehicles to, > be terrorist tools that should be banned ?  I And the operators of said backhoes carted off incommunicado to Guantanamoo Bay as 'illegal combattants'.m    G > Remember when they got paranoid about cropduster planes ? Seems to be > > that backhoes far much more damage/headaches in the USA than > cropdusterplanes.  >MC > Imagine a fleet of backhoes in manhattan, digging through a larget. > number of cables all at the same time... :-) > ( > Do backhoes have any legitimate uses ?  L ....digging missile silos in contravention of the ABM Treaty, and bunkers atL undisclosed locations for Cheney to hole up in. All other uses are likely to be banned under Patriot Act II.2   ------------------------------  % Date: Mon, 22 Sep 2003 14:44:07 -0700 , From: Ken Fairfield <My.Full.Name@intel.com>+ Subject: Re: HELP: Using CLI with C programv) Message-ID: <3F6F6D27.5684A919@intel.com>r   Lee Clark wrote: >  > A few observations...N > ? > I think lib$get_command is for use with foreign commands; you0D > should not mix foreign cmommand-handling and CLI command-handling.  >     Lib$Get_Foreign works just fine with CLI-defined commands.A As I have posted "lo, these many years ago", you can design your >? program to work equally well when invoked as a foreign command,:@ through the MCR command and/or through a verb in the DCL tables.> Lib$Get_Foreign was my preferred way of retrieving the command line.   =     An alternative to Lib$Get_Foreign is to use Cli$Get_Value(< with $VERB and/or $LINE as the 1st argument.  Unfortunately,= $VERB is not reliable, e.g., ISTR that an MCR invocation left2< $VERB with the value from the last actual verb invoked.  But $LINE is still OK...  B >                                             Cli$dcl_parse can beD > made to parse the command line that was used to start the program,D > if that's your aim (it wasn't clear from your sample run that that > is your intentions, though).  C     In the case I outlines above, you do, indeed, use Cli$DCL_ParsenA to parse the command line in preparation for calls to the variouso? Cli$Present and Cli$Get_Value, etc., functions.  Depending uponn? how the command line was retrieved, you need to assure the VERBn7 is present on the string you hand off to Cli$DCL_Parse.a   [...]i 	-Keni --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfieldo" D1C Automation VMS System Support " who:   kenneth dot h dot fairfield where: intel dot com   ------------------------------  % Date: Mon, 22 Sep 2003 15:21:43 -0400i* From: JF Mezei <jfmezei.spamnot@istop.com>' Subject: Re: How to order VMS licenses?e) Message-ID: <3F6F4BC3.CF466ECB@istop.com>n   Bob Koehler wrote:J >    I haven't looked for a while, but IIRC you said you wanted 64, that's. >    probably one of the available selections.  N Just a quick note. At the end of the boot sequence, you may see a "InteractiveM login limit set to 64" message on the console. However, this does not reflectsL the actual licence limits loaded on the machine.  It is a different limit on number of logins allowed.o  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Monthc x--   UNLIMITED DOWNLOAD   ------------------------------  % Date: Mon, 22 Sep 2003 14:54:21 -0700n, From: Ken Fairfield <My.Full.Name@intel.com>$ Subject: Re: merging queue databases) Message-ID: <3F6F6F8D.63CD60FF@intel.com>l  / Phillip Helbig---remove CLOTHES to reply wrote:h   [...]  > A > On the ALPHA and the VAX which has been around for a while, thet! > following is in SYLOGICALS.COM:g > H > $  MOUNT/CLUSTER/NOASSIST DSA1235:/SHADOW=($1$DKA200:,$3$DKA500:) USER# > $  IF F$GETDVI("DISK$USER","MNT")i	 > $  THEN ? > $    FILE := DISK$USER:[SYSTEM.MANAGER]CLUSTER_WIDE_FILES.COMa/ > $    IF F$SEARCH(FILE) .NES. "" THEN @ 'FILE'i
 > $  ENDIF > < > (I know, I should take Ken's advice and use /INCLUDE here)  1     What's perhaps more important is /NOCOPY...  h  ?     In situations like this, especially where you're specifyingi? _both_ shadow members on the MOUNT, it's good defensive programs> to check that they exist and are available prior to attempting8 the mount.  You can loop (a limited number of times) on > F$GetDvi("$1$DKA200:","EXISTS") with a WAIT of 5 seconds or so= between checks.  Followed by the same for $3$DKA500:.  Follown: those checks with F$GetDvi("$1$DKA200:","AVL") and similar for $3$DKA500:, then MOUNT.   :     You'll want to handle the case where one of the shadow? members can't be mounted, or neither can.  Perhaps you're doingo> that through your test on whether the DSA device is mounted...  3     Your list of logicals looked pretty good to me.a   	-Keni --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfields" D1C Automation VMS System Support " who:   kenneth dot h dot fairfield where: intel dot com   ------------------------------  % Date: Mon, 22 Sep 2003 20:22:02 -0500l1 From: "David J. Dachtera" <djesys.nospam@fsi.net>N$ Subject: Re: merging queue databases' Message-ID: <3F6FA03A.17138598@fsi.net>h   JF Mezei wrote:b >  > "David J. Dachtera" wrote:9 > >> Well, understand: DSA devices canNOT be MSCP-served!  > O > Since when ?  Support for DSA sets consisting of MSCP served hard drives came N > with the first version of Host Base Volume Shadowing (then called Phase II). > Been there, done that.  F You answered your own question. "...DSA sets consisting of MSCP servedG hard drives...". That is, the drives themselves can be MSCP-served. The  _DSA_ (virtual) devices cannot.c  M > Cluster Quorum is still the big limiting factor anyways. If you lose enougheP > that none of the hard drives making up the showdow set are accessible, chancesO > are very good that you will have lost quorum and won't be allowed to even tryi > to look at the drives.  # Depends how the votes are arranged.-   -- - David J. Dachteraa dba DJE Systems0 http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/i   ------------------------------  % Date: Mon, 22 Sep 2003 20:23:51 -0500t1 From: "David J. Dachtera" <djesys.nospam@fsi.net> $ Subject: Re: merging queue databases' Message-ID: <3F6FA0A7.4B038318@fsi.net>r   JF Mezei wrote:  > 1 > Phillip Helbig---remove CLOTHES to reply wrote:rI > > Perhaps.  However, the docs talk about the queue MANAGER being set uppH > > to fall over to another node etc, so it sounds like I just need one.  > > See HELP START/QUE/MANA/ON . > M > It is my *understanding* that you still want to start/queue/manager on each  > node.    See HELP START/QUE/MANA/ON .   -- n David J. Dachtera  dba DJE Systems' http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 22 Sep 2003 11:21:03 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)s' Subject: More on Microsoft, Sun and SCOg< Message-ID: <cf15391e.0309221021.6deaf10@posting.google.com>  F A recent 10-Q filing sheds some interesting light on funding of SCO byg Microsoft and Sun.  http://www.sec.gov/Archives/edgar/data/1102542/000110465903020700/a03-3388_110q.htms  C "During the three months ended July 31, 2003, Microsoft CorporationrE ("Microsoft") accounted for approximately 25 percent of total revenue A and Sun Microsystems, Inc. (Sun") accounted for approximately 12  percent of total revenue."  C Total revenue for this 3-month period was $20.055 million, so Sun'sj@ contribution for the 3 months ending July 31 would be about $2.4, million, and Microsoft's about $5.0 million.  D "During the nine months ended July 31, 2003, Microsoft accounted for? approximately 16 percent of total revenue and Sun accounted for , approximately 12 percent, of total revenue."  C Total revenue for this 9-month period was $54.964 million, so Sun's @ contribution for the 9 months ending July 31 would be about $6.6, million, and Microsoft's about $8.8 million.  A Sun contributed 12% for the last 3 months, and 12% for the last 9hA months.  One wonders if that was 12% every quarter for 3 quarters C (meaning Sun started contributing earlier than was admitted before) ? or, more likely, that Sun instead contributed a full 24% of SCOg& revenues for the 1st 3 months of 2003.  C SCO expects more revenue from Microsoft as well: "On July 31, 2003,lD Microsoft exercised an option to acquire expanded licensing rights. C Upon delivery, we expect to recognize additional revenue related tob
 this option.":  F And although the 10-Q doesn't name Sun, it notes the 210K option grantD reported in other news stories to have been given to Sun, as well as@ an additional new 12,500-share grant in the most-recent quarter:  F "During the quarter ended April 30, 2003, the Company issued a warrantD to a SCOsource licensee.  The warrant allows the licensee to acquireD 210,000 shares of the Company's common stock at an exercise price ofA $1.83 per share for a term of five years from the date of grant. eD Because the warrant was issued for no consideration to the SCOsourceC licensee, the Company has recorded the fair value of the warrant ofsF $500,000, as determined using the Black-Scholes option-pricing model." ...dD "During the quarter ended July 31, 2003, the Company issued a secondD warrant to the above mentioned SCOsource licensee in connection with@ payment of amounts owed to the Company under the initial licenseD agreement.  The warrant allows the licensee to acquire 12,500 sharesE of the Company's common stock at an exercise price of $1.83 per sharedE for a term of five years from the date of the agreement.  Because theaF warrant was issued in connection with the advance payment, the CompanyE has recorded the fair value of the warrant of $150,000, as determined-. using the Black-Scholes option-pricing model."  E So Sun not only effectively helped bankroll the IBM suit, SCO appears@@ to be paying Sun back with low-strike-price stock option grants.   ------------------------------  # Date: Tue, 23 Sep 2003 00:10:13 GMT>2 From: rdeininger@mindspring.com (Robert Deininger)% Subject: Re: Need quotes for a systemnL Message-ID: <rdeininger-2209032019040001@user-uinj5tu.dialup.mindspring.com>  < In article <bkml2l$l38@newton.cc.rl.ac.uk>, "Richard Brodie" <R.Brodie@rl.ac.uk> wrote:  @ >"Robert Deininger" <rdeininger@mindspring.com> wrote in messageG >news:rdeininger-1909032104470001@user-105n86e.dialup.mindspring.com...t >t3 >> >Can you still get DS10/466 systems new from HP?n >aM >> This question has come up several times lately.  I just (re)checked the HPh5 >> Alphaserver web page, and the DS10 is still shown.e > L >Yes, but the question was about a 466. As I recall, HP retired the DS10/466/ >some time ago, and now only sell the DS10/600.  >hP >If you really want to buy a model that has dropped off the entry level end, you' >can't buy it new. What's the big deal?h  J Yes, this particular note was about the 466 MHz system.  I missed that theH first time through.  I've also seen reports of problems getting DS10 600 MHz systems.  C 466 MHz has been retired for quite a while, but should be availableh used/refurbished.   G 600 MHz DS10 is still available new and (according to information I gottC today) is due to be sold well into next year.  There _should_ be no I problems buying new DS10s at 600 MHz.  (I think the minimum memory is now 3 512 MB; the 256 MB system seems to have gone away.)    ------------------------------    Date: 22 Sep 2003 13:10:40 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)p2 Subject: Re: ODS-5 or not ODS-5 What is the answer3 Message-ID: <1GYeOqQs3gnq@eisner.encompasserve.org>h  [ In article <00A26479.4EC6910B.11@tachysoft.com>, Wayne Sewell <wayne@tachysoft.com> writes:>  N > My feeling is that there is no point in dealing with the addition hassles ofP > ods-5 unless there is a specific need for it.  I have *one* ods-5 volume, withP > a label of billydisk, and everything that insists on billy-style names goes on > it.h  F I have an ODS-5 disk and I use it to ensure that LJK/Security can dealD with ODS-5 disks properly.  I make the rash assumption there are notG any bugs that would occur only if the ODS-5 disk had a different volumet name :-)   ------------------------------  % Date: Mon, 22 Sep 2003 15:31:35 -0500 ( From: Wayne Sewell <wayne@tachysoft.com>2 Subject: Re: ODS-5 or not ODS-5 What is the answer/ Message-ID: <00A264A7.12511DDA.2@tachysoft.com>t   >h\ >In article <00A26479.4EC6910B.11@tachysoft.com>, Wayne Sewell <wayne@tachysoft.com> writes: >pO >> My feeling is that there is no point in dealing with the addition hassles of Q >> ods-5 unless there is a specific need for it.  I have *one* ods-5 volume, withsQ >> a label of billydisk, and everything that insists on billy-style names goes onl >> it. >yG >I have an ODS-5 disk and I use it to ensure that LJK/Security can dealrE >with ODS-5 disks properly.  I make the rash assumption there are not>H >any bugs that would occur only if the ODS-5 disk had a different volume	 >name :-)E  O Yes, testing is the other reason for having one ods-5 disk.  I deliberately putiM some weird file and subdirectory names in a crap directory tree on that disk,aJ containing all possible ascii characters.  This was to verify that tapesysM could back up the disk, and more importantly handle the strange file names in M the history.   Due to the radically different characteristics of ODS-5, I had N to completely rewrite the tapesys history system, and it would have been kinda. difficult to test that without an actual disk.  L Of course, as far as actually *using* the disk, I don't, except for netbeans= and a few other products insisting on billy-esque file names.o   WaynenO ===============================================================================bN Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html   vO ===============================================================================iN Butler:"Gentlemen!"  Curly(as he and other Stooges look around):"Who came in?"   ------------------------------  % Date: Mon, 22 Sep 2003 15:32:35 -0700 , From: Ken Fairfield <My.Full.Name@intel.com>2 Subject: Re: ODS-5 or not ODS-5 What is the answer) Message-ID: <3F6F7883.9FC60F18@intel.com>e   "Michael D. Ober" wrote: > B > We run ODS-5 on all our disks, including system disks.  You have@ > to convert the system disk from the console or during a VMS OS > upgrade, however.e  A     Quick question: did you enable hard links on the system disk?b  <     Reason I ask is that I went ahead and bit the bullet and= let the VMS 7.3-1 installation procedure make my hobbiest VMSr? system disk ODS-5.  But I was, ahem, hesitant to go all the way = and allow hard inks to be enabled since I could find precioush? little documentation telling me what impact this would have on i. the system root structure and VMS$COMMON, etc.  
     Comments?t   	-Kene --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfield " D1C Automation VMS System Support " who:   kenneth dot h dot fairfield where: intel dot com   ------------------------------    Date: 22 Sep 2003 09:07:36 -0700, From: vinit.adya@mizuhocbus.com (Vinit Adya)0 Subject: Oracle 8.1.7.4 on VMS 7.2-2 Performance= Message-ID: <eb8f4d7b.0309220807.3eee8fd3@posting.google.com>.  B We have noticed that a query like "select * from dual;" takes muchA longer on OpenVMS than on Windows based PC. (Did the opetation 1MhE times and compared the results...) It takes twice as long and it usesr? 100% of one on the cpu on alpha. on intel it is 3 times faster.|  = Any ideas what might be wrong (with VMS or this test itself?)n   Regards, Vinitn   ------------------------------  # Date: Mon, 22 Sep 2003 17:44:42 GMT-9 From: Hein van den Heuvel <hein_netscape@eps.zko.dec.com>g4 Subject: Re: Oracle 8.1.7.4 on VMS 7.2-2 Performance/ Message-ID: <3F6F3487.CFB32E4A@eps.zko.dec.com>g   Vinit Adya wrote:i  D > We have noticed that a query like "select * from dual;" takes much- > longer on OpenVMS than on Windows based PC.k   :k  ? > Any ideas what might be wrong (with VMS or this test itself?)a  < The TCP stack would be my guess. You can perhpas verify that; with statspack CPU counts or show system / moni mode to get 0 an impression where the system is psending time.  H I could try to be funny and say that VMS systems, perhpas unlike Windows systems,F do not expect to be asked to do frivolous works, and are not optimized	 for that.   E The VMS provided TCP stack may be little heavy (spinlocks and stuff)..F A test as described would  exacerbate that by making the net-to-oracle? proportion be for example 80:20 versus perhaps 10:90 for a more- typical business transaction.   . Alternative stacks may or might not do better.  E For SMP systems the next VMS version (opal?) with the next UCX (5.4?)tC will be significantly improved from a SCALING point of view, due tor reduced  IOLOCK8 usage.   fwiw,  Hein.v   ------------------------------  % Date: Mon, 22 Sep 2003 21:22:21 +0100e( From: "John Travell" <john@jomatech.com>4 Subject: Re: Oracle 8.1.7.4 on VMS 7.2-2 Performance9 Message-ID: <bknm15$3qqeh$1@ID-120847.news.uni-berlin.de>t  9 "Vinit Adya" <vinit.adya@mizuhocbus.com> wrote in message 7 news:eb8f4d7b.0309220807.3eee8fd3@posting.google.com...eD > We have noticed that a query like "select * from dual;" takes muchC > longer on OpenVMS than on Windows based PC. (Did the opetation 1MoG > times and compared the results...) It takes twice as long and it usesaA > 100% of one on the cpu on alpha. on intel it is 3 times faster.- >-? > Any ideas what might be wrong (with VMS or this test itself?)a  : What performance level are the systems you are comparing ?3 What versions of the respective Operating systems ?DI If you are, for example, comparing an 500MHz Alphaserver 1200 with a 3GHztB Intel machine you do not have a level playing field to start with.K There could be many reasons, but you have not given us anything like enough,9 background information to base any rational answers upon.      -- John Travell" Independent VMS crashdump analyst. john- at - jomatech - dot - comc +44-(0)23-92552229 http://www.jomatech.com/       --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.520 / Virus Database: 318 - Release Date: 18/09/2003l   ------------------------------    Date: 22 Sep 2003 14:13:49 -0700, From: vinit.adya@mizuhocbus.com (Vinit Adya)4 Subject: Re: Oracle 8.1.7.4 on VMS 7.2-2 Performance= Message-ID: <eb8f4d7b.0309221313.52b1ba03@posting.google.com>s   thans Hein,r  E But how does TCP/IP or Disk IO have an impact on a simple query like - SQL> select * from dual; -  D I specifically am testing this beacuse i did see serious performance@ differences when the same apllication was running VMS Oracle vs. Windoze Oracle.u  D I am absolutely frustated. PC based Oracle beats my 2 cpu GS140 with@ 4GB memory, by more than 10 times. 3 seconds on PC vs 40 Secs on GS140.  p Hein van den Heuvel <hein_netscape@eps.zko.dec.com> wrote in message news:<3F6F3487.CFB32E4A@eps.zko.dec.com>... > Vinit Adya wrote:p > F > > We have noticed that a query like "select * from dual;" takes much/ > > longer on OpenVMS than on Windows based PC.  >  n > :i >   A > > Any ideas what might be wrong (with VMS or this test itself?)m > > > The TCP stack would be my guess. You can perhpas verify that= > with statspack CPU counts or show system / moni mode to getf2 > an impression where the system is psending time. > J > I could try to be funny and say that VMS systems, perhpas unlike Windows
 > systems,H > do not expect to be asked to do frivolous works, and are not optimized > for that.i > G > The VMS provided TCP stack may be little heavy (spinlocks and stuff).tH > A test as described would  exacerbate that by making the net-to-oracleA > proportion be for example 80:20 versus perhaps 10:90 for a morer > typical business transaction.a > 0 > Alternative stacks may or might not do better. > G > For SMP systems the next VMS version (opal?) with the next UCX (5.4?)eE > will be significantly improved from a SCALING point of view, due to 	 > reduced> > IOLOCK8 usage. >  > fwiw,f > Hein.    ------------------------------  % Date: Tue, 23 Sep 2003 00:49:45 -0400u$ From: Hein <hein_cov@eps.zk.dec.com>4 Subject: Re: Oracle 8.1.7.4 on VMS 7.2-2 Performance. Message-ID: <3F6FD0E9.113C3901@eps.zk.dec.com>   Vinit Adya wrote:   
 > thans Hein,a >nF > But how does TCP/IP or Disk IO have an impact on a simple query like > SQL> select * from dual;  I > Well, Like John Travell pointed out, you did not give us much to go on.r  I I simply assumed you used a third computer as driver to either VMS on NT,s< coming in though the network stack for each select. Correct?  N If also assumed you leave the connection open, once connected through SQL*net.  T If you just do a connect, select from dual, and close, then you are involking a fullL process creation per query and VMS will be slower. Bu such test woudl not beR very realistic... in general. Most apps connect once and do a lot of work, or theyQ have a transaction manager for (near) stateless quick queries which keep the linkt open.o  S I'm afraid you'd have explain a whole lot more about your test setup to give bettertR advise, and if you do, it may sound too much like real work for use wierdos to try to help.   ------------------------------  % Date: Mon, 22 Sep 2003 23:43:13 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>+ Subject: OT: Verisign's contract with ICANNi) Message-ID: <3F6FC131.684783A4@istop.com>r  N Here is the link to the agreement betwene Verisign and ICANN for the provision of the .com domain registry.  ; http://www.icann.org/tlds/agreements/verisign/com-index.htma  G Just skimming though it, I found plenty of areas where ICANN, it it hadwM gonads, could have pounced on Verisign the day they put in the redirection ofi& non existant sites on a verisign site.  I Notably: the total need of separation (even physical offices) between theaM subsidiary of verisign that has the responsability of central domain registryoO and all other verisign businesses, especially its domain registration business.   N Notably, the domain registry must be neutral and in no way must Verisign'otherH businesses benefit from Verisign's position as a neutral operator of the central registry.e  C Another responsability is to provide for inactive domains. They aretM essentially reserved names that do not have any DNS entries. The change voidssH this mandate. Many companies have registered some of their trademarks inL "inactive" mode (to prevent their trademark from being registered by someoneS else). Well, Verisign now makes use of those domains by pointing to their own site.r  G There were many other areas which at first glance can have been used to N declare verisign "guilty". However, i suspect Verisign has its own laywers whoG went though the dots on the is and saw a way to justify their move. But>I hopefully, they knew it was illegal but assumed ICANN would remain a goodm! little poodle and not lift a paw.   2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/MonthR x--   UNLIMITED DOWNLOAD   ------------------------------  % Date: Tue, 23 Sep 2003 00:17:57 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>/ Subject: Re: OT: Verisign's contract with ICANN>) Message-ID: <3F6FC952.8D6500B0@istop.com>   I Verisign has responded to ICANN's "polite" request to remove the wildcardt! until proper study has been made.   ? http://www.icann.org/correspondence/lewis-to-twomey-21sep03.htm   > That link also contains a link to ICANN's request to Verisign.  L Essentially, Verisign states that everyone is happy and benefiting from this, change and will not remove this wildcarding.  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Monthr x--   UNLIMITED DOWNLOAD   ------------------------------  % Date: Tue, 23 Sep 2003 00:32:59 -0400o* From: JF Mezei <jfmezei.spamnot@istop.com>/ Subject: Re: OT: Verisign's contract with ICANNn( Message-ID: <3F6FCCD7.F953BCB@istop.com>  0 ICANN has released a second, stronger statement.  ? http://www.icann.org/correspondence/secsac-to-board-22sep03.htmT  H This is from the Security and Advisory Comittee of ICANN. And the letterK includes many of the negative impacts of the Verisign change. They also askeL the IETF to update the DNS standards to restrict use of wildcard at root and top level domain levels.  J But it looks like this may take some time. There is a public meeting to be  held October 7th on this matter.  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Month  x--   UNLIMITED DOWNLOAD   ------------------------------  % Date: Mon, 22 Sep 2003 22:03:42 -0700c% From: Dean Woodward <deanw@rdrop.com>e/ Subject: Re: OT: Verisign's contract with ICANN4( Message-ID: <3F6FD42E.2090508@rdrop.com>   JF Mezei wrote:l  2 > ICANN has released a second, stronger statement. > A > http://www.icann.org/correspondence/secsac-to-board-22sep03.htm  > C > This is from the Security and Advisory Comittee of ICANN. And thetF > letter includes many of the negative impacts of the Verisign change.G > They also ask the IETF to update the DNS standards to restrict use ofl/ > wildcard at root and top level domain levels.e > F > But it looks like this may take some time. There is a public meeting( > to be held October 7th on this matter.  ( Meanwhile, ICANN is asking for feedback:  B ! To gather information on security and stability implications, we< ! invite inputs from all interested parties. Send inputs to: !o ! secsac-comments@icann.orgg   ------------------------------  % Date: Mon, 22 Sep 2003 16:02:50 -0400h* From: JF Mezei <jfmezei.spamnot@istop.com> Subject: Re: P.C. Market share) Message-ID: <3F6F5568.ACDD7B60@istop.com>a   "John N." wrote:7 > Does anyone wonder why HP is losing PC market share? r  I In one way, this is to be expected. Nothwidthstanding any quality/servicetN issues, HP could not expect to hold on to the combined market share of both HPF and Compaq. Any merger of 2 competing companies results in loss of theN combined market share. There is the period of uncertaintly about which productI lines will survive, any negative impact of one company's influence on the K other, as well as the loss of customers who had chosen B because they hatedhR A's products/service and are now find themselves dealing with A. (and vice versa).  K A company rises to the top not by eating others, but by beating others. You-N need better products, better prices, better service than others to rise to theH top. If you have flaws that prevent you from rising, no matter how manyt> companies you buy, those flaws will bring you back down again.  N My only question is whether HP's current performance is what Carly and CompanyK had trully planned behind the scenes. (as opposed to the rosy figures Carly 9 and Curly were parading around to try to sell the merger.m  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Month6 x--   UNLIMITED DOWNLOAD   ------------------------------  # Date: Mon, 22 Sep 2003 23:13:33 GMT-9 From: "Joe Headrick" <william.j.headrick@lmco.nospam.com>r Subject: PCI Device AccessC Message-ID: <xmLbb.64890$NM1.8457@newsread2.news.atl.earthlink.net>6   Hi,oK    Need some help (duh), and I can't find a copy of the "new" device driver  bookF (found a couple of used 1996 incarnations, but figure I should get the second editionJ assuming it has more PCI and map_io/read_io stuff for PCI). Anyway, I need to be I able to turn off the horiz and vert sync on a couple of types of graphicsu
 cards, andF figured writing the IO registers would be the best bet (no source, and
 apparently no E DPMI in 6.2-1H3). Well, I have become very proficient in crashing the- computer in-I all sorts of ways :-) I have found the ADP and the device block, and evenu figured outcG how to work device_lock (I think). But when I try to map/unmap - crash.1 ProbablyI something simple, but there is apparently enough difference in PCI vs ISA  that all theK examples I have found so far have led me astray (all ISA of course). And ofR courseL actual IO crashes too (although using the PCI config calls work fine to read
 the config registers).iL   If anyone has a IO example that they can share... It would be appreciated.      Joe Headrick    ------------------------------  + Date: Mon, 22 Sep 2003 20:22:43 -0500 (CDT)/ From: sms@antinode.org1 Subject: Peculiar (header-free) junk e-mail.  Eh?n) Message-ID: <03092220224383@antinode.org>E  E    Amid the current junk flood, I got a couple of unusually brief (no/2 header) messages.  Here's how VMS MAIL shows them:  P     #40         22-SEP-2003 14:55:32.28                                     MAIL From:   SMTP%"angileri@clas.it"  To:     sms2 CC:  Subj::   MAIL>)  P     #41         22-SEP-2003 14:58:39.82                                     MAIL From:   SMTP%"angileri@clas.it". To:     smsb CC:i Subj:a   MAIL>s     ALP $ tcpip show version  ?   Compaq TCP/IP Services for OpenVMS Alpha Version V5.3 - ECO 2N4   on a AlphaStation 200 4/233 running OpenVMS V7.3-1    G    I'd be interested in an explanation of the lack of header data.  DuetG to the current junk flood, relevant log files have probably been purged> by now,e  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode,org"    Saint Paul  MN  55105-2547    ------------------------------  % Date: Mon, 22 Sep 2003 22:11:49 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>5 Subject: Re: Peculiar (header-free) junk e-mail.  Eh?n) Message-ID: <3F6FABCA.E35FAE08@istop.com>o   sms@antinode.org wrote:u > G >    Amid the current junk flood, I got a couple of unusually brief (no 4 > header) messages.  Here's how VMS MAIL shows them: > R >     #40         22-SEP-2003 14:55:32.28                                     MAIL! > From:   SMTP%"angileri@clas.it"s
 > To:     smse > CC:e > Subj:d   telnet/port=25 your.node.com <welcome banner> HELO clas.it MAIL FROM: angileri@clas.itr RCPT TO: sms@pronode.com DATA .    QUIT  L Will do the trick. (eg: an empty DATA phase). It is interesting that the VMSG software won't add at least the Reply-path: and Received: header to the % contents, which is it supposed to do.   2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Monthe x--   UNLIMITED DOWNLOAD   ------------------------------  + Date: Mon, 22 Sep 2003 22:01:22 -0500 (CDT)I From: sms@antinode.org5 Subject: Re: Peculiar (header-free) junk e-mail.  Eh?-) Message-ID: <03092222012230@antinode.org>0  * From: JF Mezei <jfmezei.spamnot@istop.com>  	 > > [...]oT > >     #40         22-SEP-2003 14:55:32.28                                     MAIL# > > From:   SMTP%"angileri@clas.it"l > > To:     smsf > > CC:t	 > > Subj:  >  > telnet/port=25 your.node.com > <welcome banner> > HELO clas.it > MAIL FROM: angileri@clas.it  > RCPT TO: sms@pronode.com > DATA > .s >  > QUIT > N > Will do the trick. (eg: an empty DATA phase). It is interesting that the VMSI > software won't add at least the Reply-path: and Received: header to thep' > contents, which is it supposed to do.c  ,    Ah.  Sounds like a defect to me.  Thanks.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode,org-    Saint Paul  MN  55105-25472   ------------------------------  # Date: Mon, 22 Sep 2003 21:02:10 GMTRA From: "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk>.0 Subject: Possible e-mail scam allegedly from HP.= Message-ID: <mrJbb.2480$hM7.23261426@news-text.cableinet.net>   
 Hello All,  H I think that this could be an e-mail scam. I know nothing of any invoiceH 00179, nor have I ever dealt knowingly with anyone at Bangalore, despiteI dealing frequently with HP. The 'English' at the bottom is a bit odd too.e  L My apologies to one Siriam Gautam if I'm being alarmist, but I fear I'm not.   -- s   Hope this helps, Colin.o) colin DOT butcher AT xdelta DOT co DOT ukmL Systems Archaeologist - Investigation & troubleshooting of older systems and	 networks.c         -----Original Message-----@ From: GAUTAM,SRIRAM (HP-India,ex3) [mailto:sriram.gautam@hp.com]( Sent: Monday, September 22, 2003 3:10 PM To:  Subject: Creationa    # Return-path: <sriram.gautam@hp.com>h1 Received: from punt-3.mail.demon.net by mailstorel  for  id 1A1ROF-0004g7-B7;   Mon, 22 Sep 2003 19:53:49 +0000= Received: from [217.174.192.68] (helo=webmail2.amenworld.com)e7  by punt-3.mail.demon.net with smtp id 1A1ROF-0004g7-B7-&  for ; Mon, 22 Sep 2003 14:10:11 +0000F Received: (qmail 10523 invoked by uid 501); 22 Sep 2003 14:10:07 -0000
 Delivered-To::H Received: (qmail 10518 invoked from network); 22 Sep 2003 14:10:06 -0000; Received: from unknown (HELO sngrel7.hp.com) (192.6.86.111) A   by webmail2.amenworld.com with SMTP; 22 Sep 2003 14:10:06 -0000tE Received: from XSGBRG2.sgp.hp.com (xsgbrg2.sgp.hp.com [15.85.49.112])_3  by sngrel7.hp.com (Postfix) with ESMTP id 1501220Cs.  for <>; Mon, 22 Sep 2003 22:10:07 +0800 (SGP)H Received: by xsgbrg2.sgp.hp.com with Internet Mail Service (5.5.2656.59)/  id <TJYQYDPP>; Mon, 22 Sep 2003 22:10:06 +0800aG Message-ID: <EA68A1CBC8B50346BCB125ACA5A913B38BCAF1@xin05.india.hp.com>i= From: "GAUTAM,SRIRAM \(HP-India,ex3\)" <sriram.gautam@hp.com> 	 To: "" <>l Subject: Creation$% Date: Mon, 22 Sep 2003 22:10:04 +0800  MIME-Version: 1.0R- X-Mailer: Internet Mail Service (5.5.2656.59)a$ Content-Type: multipart/alternative;1  boundary="----_=_NextPart_001_01C38113.385F0960"O     Hi, ' This is with reference Invoice - 00179.   K Please can you give the following details as they are very important for us  to make payments   1. Full Name  
 2. Address   3. Bank detailse   4. Vat Number.   Thanks and regards Gautam       PO Management UK & Ireland  $ Hewlett Packard Financial Operations  $ Global e-Business Centre - Bangalore   Telnet -  +91-80-504 6875$   Fax     - +91-80-520 4088c  ! IDD    - +91-80-520 4050 Ext-6875h      I If you have purchase order or invoice related queries, kindly contact ouriL Customer Response Centre either by telephone at 0800 89 0011 followed by 800C 325 5372( toll free number) or by email at uk-ireland-vp.crc@hp.comV  K This message contains confidential information and is intended only for the D individual named. If you are not the named addressee, you are herebyG notified that disclosing, copying, distributing or taking any action in F reliance on the contents of this e-mail is strictly prohibited. PleaseK notify the sender immediately by e-mail if you have received this e-mail by I mistake, and delete this material from your any computer. Hewlett-Packard.D accepts no liability for errors or omissions in the contents of this8 message, which arise as a result of e-mail transmission.   ------------------------------  % Date: Mon, 22 Sep 2003 18:05:30 -0400f* From: JF Mezei <jfmezei.spamnot@istop.com>4 Subject: Re: Possible e-mail scam allegedly from HP.) Message-ID: <3F6F7220.F3EB0A1A@istop.com>,   Colin Butcher wrote:+ > I think that this could be an e-mail scam-N > My apologies to one Siriam Gautam if I'm being alarmist, but I fear I'm not.  H I'd send a copy of this message to abose@hp.com as well as a copy to theJ person who is the apparent originator. This is clearly a scam that did notM originate from hp. But they probably used a legitimate email from HP (or thatd% person) as a template for their scam.i  L *IF* the message ID is valid, perjaps HP's logs might be able to see that itL was a legitimate message sent from mr Gautam to some person, and that personM might be the guilty party who would have then used the message from Mr GautamaG to compuse the fraudulent message and make it appear to be from Gautam.k  3 > Received: from punt-3.mail.demon.net by mailstore   ? > Received: from [217.174.192.68] (helo=webmail2.amenworld.com)a9 >  by punt-3.mail.demon.net with smtp id 1A1ROF-0004g7-B7_( >  for ; Mon, 22 Sep 2003 14:10:11 +0000  L amenworld is a paris based outfit. This seems to be like that last trustable> Received: in the header. The rest below is most likely forged.    H > Received: (qmail 10523 invoked by uid 501); 22 Sep 2003 14:10:07 -0000 > Delivered-To:m  E The above "Received" is suspicious since it lacks a "from". Also, the)J Delivered-To: is out of place since there are other "Received" below. (and since it is blank)  J > Received: (qmail 10518 invoked from network); 22 Sep 2003 14:10:06 -0000    again, another suspect Received:    = > Received: from unknown (HELO sngrel7.hp.com) (192.6.86.111) C >   by webmail2.amenworld.com with SMTP; 22 Sep 2003 14:10:06 -0000e  I 192.6.86.111 does belong to HP.  This is an attempt to make amerworld.comnM appear to be a relay, to try to shift responsability from amenworld to hp.com-  G > Received: from XSGBRG2.sgp.hp.com (xsgbrg2.sgp.hp.com [15.85.49.112])y5 >  by sngrel7.hp.com (Postfix) with ESMTP id 1501220Cs0 >  for <>; Mon, 22 Sep 2003 22:10:07 +0800 (SGP)  M 15.85.49.112 does belong to HP, but again, the "for <>" is highly suspicious.s    J > Received: by xsgbrg2.sgp.hp.com with Internet Mail Service (5.5.2656.59)1 >  id <TJYQYDPP>; Mon, 22 Sep 2003 22:10:06 +0800r  D 5.5.2656.59 is not a valid IP address to begin with. (2656 is a deadW giveaway). Secondly, 5.5.*.* is not an assigned IP address/network and is not routable.a  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Month. x--   UNLIMITED DOWNLOAD   ------------------------------  % Date: Mon, 22 Sep 2003 23:01:23 +0100s) From: Antonio Carlini <arcarlini@iee.org>u4 Subject: Re: Possible e-mail scam allegedly from HP.> Message-ID: <_iKbb.2221$%G1.598@newsfep4-winn.server.ntli.net>   Colin Butcher wrote:J > I think that this could be an e-mail scam. I know nothing of any invoiceJ > 00179, nor have I ever dealt knowingly with anyone at Bangalore, despiteK > dealing frequently with HP. The 'English' at the bottom is a bit odd too.f  , Wow - the Nigerians have been outsourced :-)  4 Can't you call the freefone number and see what they have to say for themselves?r   AntonioE   --   ---------------e- Antonio Carlini             arcarlini@iee.orgM   ------------------------------  + Date: Mon, 22 Sep 2003 16:05:21 +0000 (UTC)u= From: jlw@psulias.psu.edu (j.lance wilkinson, (814) 865-1818)*M Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ? . Message-ID: <bkn6k1$1du6$1@f04n12.cac.psu.edu>  K In article <3F6EE95F.B25018D1@Omond.net>, Roy Omond <Roy@Omond.net> writes:n= >The current flood of spam seems to be totally out of control-  >(doomed, we're all doomed ...). >09 >I happened to venture over to http://www.process.com and ; >their PreciseMail AntiSpam Gateway sounds *very* promisingD9 >(coming from Hunter Goatley, it's something I would havee >a lot of faith in ...). >e# >Anyone had any experience so far ?i >s
 >Roy Omond >Blue Bubble Ltd.v >c    > 	Don't recall exactly what we're limited to say as Beta users,< 	but I certainly don't think they'd be upset with a positive 	reaction being voiced.n  D 	My site ran it since 4th July 2003.  Liked it.  Bought it.  Waiting/ 	anxiously for the purchased license to arrive.e  C 	Did this without evaluating the BAYESIAN Filters or the QUARANTINE2? 	aspect.  Just the inserted headers assigning a spam likelyhoodl@ 	score (SPAMICITY).  My users had mixed reviews, generally good.? 	Will likely add BAYESIAN Filters and/or Quarantine after laterc4 	releases make them more practical for my user base.  > 	Using it did point out some performance issues on our system,7 	which were resolvable once we realized what they were.i  P +----"Never Underestimate the bandwidth of a station wagon full of mag tapes"--+B | J.Lance Wilkinson ("Lance")		InterNet:  Lance.Wilkinson@psu.edu < | Systems Design Specialist - Lead	AT&T:      (814) 865-18189 | Digital Library Technologies		FAX:       (814) 863-3560i2 | 3 Paterno Library				"I'd rather be dancing..." B | Penn State University		    A host is a host from coast to coast,K | University Park, PA 16802	    And no one will talk to a host that's closefC | <postmaster@psulias.psu.edu>	    Unless the host that isn't closet: | EMail Professional since 1978	    Is busy, hung or dead.P +---------"He's dead, Jim. I'll get his tricorder. You take his wallet."-------+9                 [apologies to DeForest Kelley, 1920-1999]t3 <A Href="http://perdita.lcs.psu.edu">home page</a> oJ <a Href="http://perdita.lcs.psu.edu/junkdec.htm">junk mail declaration</a> --	/"\ 	\ /	ASCII RIBBON CAMPAIGN 	 X	AGAINST HTML MAILp 	/ \   ------------------------------   Date: 22 Sep 2003 13:15:17 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)mM Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ? 0 Message-ID: <bkmsl5$dj1$1@n.ruf.uni-freiburg.de>  K In article <3F6EE95F.B25018D1@Omond.net>, Roy Omond <Roy@Omond.net> writes:e= >The current flood of spam seems to be totally out of controlh  >(doomed, we're all doomed ...). >"9 >I happened to venture over to http://www.process.com andy; >their PreciseMail AntiSpam Gateway sounds *very* promisingc9 >(coming from Hunter Goatley, it's something I would havem >a lot of faith in ...). >e# >Anyone had any experience so far ?l  I As a beta-tester I do recommend it. You will need PMDF as a prerequisite,e though. O Concerning the recent SWEN virus flood: PMDF will be sufficient to deal with its! or you might want Sophos for VMS.c   Regardsh    Christoph GartmannC   --  E  Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452   ImmunbiologieA  Postfach 1169                 Internet: gartmann@immunbio.mpg.ded  D-79011  Freiburg, Germanyp9                http://www.immunbio.mpg.de/home/menue.htmlp   ------------------------------  % Date: Mon, 22 Sep 2003 21:50:23 -0400:* From: "Bill Todd" <billtodd@metrocast.net>M Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?n2 Message-ID: <ZuednaI5fOd9O_KiXTWJiw@metrocast.net>  , "Roy Omond" <Roy@Omond.net> wrote in message# news:3F6EE95F.B25018D1@Omond.net....> > The current flood of spam seems to be totally out of control! > (doomed, we're all doomed ...).0 >2: > I happened to venture over to http://www.process.com and< > their PreciseMail AntiSpam Gateway sounds *very* promising: > (coming from Hunter Goatley, it's something I would have > a lot of faith in ...).b >g$ > Anyone had any experience so far ?  K It's worth noting that ISP quarantine measures don't necessarily solve thiss particular problem  L My ISP has faithfully caught all (I think) of the infected emails and almostH all the uninfected garbage generated as a by-product.  However, this hasH merely resulted in my having about 10,000 (as of now) quarantined emailsI that I have to deal with (I'd just leave them in quarantine to be deleted6K automatically after a couple of weeks, but unfortunately the virus versions L keep the non-virus quarantined emails from being displayed until I bring theI infected count down under 500, and since I have the filter set to maximum I sensitivity *once* in a while an uninfected message gets quarantined that  shouldn't be).  H And the particular quarantining mechanism my ISP uses doesn't allow bulkJ discard of infected emails (though it does allow it for uninfected ones) -J so I have to discard them one page (10) at a time (both these issues couldL of course be dealt with by enhancing the software interface, but as of right  now that's what I'm stuck with).  G While on the general subject, I've noted (from recipient lists) that myoJ deluge seems to have started with the harvesting of comp.sys.intel - whichD may be why most people in c.o.v. haven't been hit as hard as I have.J Unfortunately, today comp.arch seems to have been harvested as well, which7 probably explains why my volume just took another jump.r  I I'd vote for that $.01 per email fee that was proposed:  I think it wouldeI stop most spammers in their tracks, at least to the degree of making themmL select for people who at least *might* be interested in their product ratherI than just spam the world.  And it would give infected sites a far greater + incentive to clean themselves up *quickly*.A   - bill   ------------------------------   Date: 23 Sep 2003 02:19:59 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)M Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?d9 Message-ID: <bkoakf$42ov5$1@ID-135708.news.uni-berlin.de>d  2 In article <ZuednaI5fOd9O_KiXTWJiw@metrocast.net>,- 	"Bill Todd" <billtodd@metrocast.net> writes:i >  > K > I'd vote for that $.01 per email fee that was proposed:  I think it wouldaK > stop most spammers in their tracks, at least to the degree of making themtN > select for people who at least *might* be interested in their product rather > than just spam the world.   F And how is it proposed to do this?  How do you know how many emails myE machine sends out?  Especially when I hide the originating machine innH the first place.  Hint:  There is no central place on the INTERNET whereI the FBI or NSA is monitoring all the traffic.  The telephone is regulatedaH by Federal Law and they still haven't stopped telemarketers from callingF my phone, not with a state no-call system or a Federal no-call system.K I'll say it once more... You can't solve social problems with technologicals
 solutions.  K >                            And it would give infected sites a far greatere- > incentive to clean themselves up *quickly*.n  G Most infected sites are not even aware that they are infected, where is=D the incentive?  And once again, Who collects the fee and how do they! determine who to collect it from?t   bill   -- oJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Mon, 22 Sep 2003 23:02:38 -0400n* From: JF Mezei <jfmezei.spamnot@istop.com>M Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?r) Message-ID: <3F6FB7B0.10B557CB@istop.com>k   Bill Gunshannon wrote:J > the first place.  Hint:  There is no central place on the INTERNET where/ > the FBI or NSA is monitoring all the traffic.   D That is what *they* want you to think, says the tin-foil-hat wearing conspiracist. :-)   L Haven't certain countries with an orwelian government declared stuff such asK "total information awareness" where they want to monitor everything that isa. going on, with the excuse to find terrorists ?  K I suspect that NSA/CIA/whatever have developped close working relationshipsmM with Cisco to intercept/log all traffic. Of course, they still need to have apM peek inside AOL and other large ISPs since intra-ISP emails don't flow out onV mail inter-ISP trunks.  K However, it is probably fairly possible to intercept all traffic flowing in-< and out of a country since those trunks are not so numerous.  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Montha x--   UNLIMITED DOWNLOAD   ------------------------------    Date: 22 Sep 2003 21:41:57 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)sM Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?n3 Message-ID: <m0uIVEotocho@eisner.encompasserve.org>   d In article <bkoakf$42ov5$1@ID-135708.news.uni-berlin.de>, bill@cs.uofs.edu (Bill Gunshannon) writes:4 > In article <ZuednaI5fOd9O_KiXTWJiw@metrocast.net>,/ > 	"Bill Todd" <billtodd@metrocast.net> writes:O >>   >> .L >> I'd vote for that $.01 per email fee that was proposed:  I think it wouldL >> stop most spammers in their tracks, at least to the degree of making themO >> select for people who at least *might* be interested in their product rathera >> than just spam the world.   > H > And how is it proposed to do this?  How do you know how many emails my > machine sends out?  = Receiving ISPs charge the backbone for each message received.h? Backbones charge each other (and sending ISPs) for each messagea	 received.s   ------------------------------  % Date: Mon, 22 Sep 2003 23:55:53 -0400r* From: "Bill Todd" <billtodd@metrocast.net>M Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ? 2 Message-ID: <A96cneEWSfHWWfKiU-KYgg@metrocast.net>  5 "Bill Gunshannon" <bill@cs.uofs.edu> wrote in messages3 news:bkoakf$42ov5$1@ID-135708.news.uni-berlin.de...i4 > In article <ZuednaI5fOd9O_KiXTWJiw@metrocast.net>,. > "Bill Todd" <billtodd@metrocast.net> writes: > >  > >eG > > I'd vote for that $.01 per email fee that was proposed:  I think ite wouldtH > > stop most spammers in their tracks, at least to the degree of making themI > > select for people who at least *might* be interested in their productt rather > > than just spam the world.n >S$ > And how is it proposed to do this?  K The sender's ISP should be able to - it should be aware of which connection.J the out-going email is arriving on, though whether any mechanism currentlyI exists for tracking this and charging on a per-destination basis exists I 
 have no idea.n   ...n  E > >                            And it would give infected sites a fars greaterI/ > > incentive to clean themselves up *quickly*.  >tI > Most infected sites are not even aware that they are infected, where ist > the incentive?  C The incentive is that they're sending the spam or virus, so they'reaI responsible for paying for doing so - whether they know it's going out oruA not.  That creates an incentive not only to put a stop to it ASAPtL (presumably with some helpful feedback from the ISP just in case they're notK aware that their send rate just ballooned) but to protect better against itu in the first place.e  G It's a similar philosophy to that which makes people at least partiallymJ responsible for damages caused by dangerous items when they're negligentlyK left lying around.  To make the parallel even closer, perhaps charges couldyL be waived until a first notice of the abuse had been given (which would giveI the ISP some incentive to get it out quickly, now that I think of it...). J And/or the ISP could just shut off out-going email service until getting aK response (this would cover cases where the customer wasn't in a position to7J respond immediately); when the customer actually *wanted* the volume, s/he  could notify the ISP in advance.  K There are, of course, two only slightly related areas being discussed here: L spam, where the source of the spam is clear (at least to the spammer's ISP),J and virii (and their noxious relatives), where some third-party is trickedJ into being the source.  The $.01 per-email charge would certainly help theF former (and could not even be circumvented by, say, coopting a zillionF individual to send it), and the infrastructure required to track thoseH charges would at least provide some support for dealing with the latter.   - bill   ------------------------------  % Date: Tue, 23 Sep 2003 00:38:19 -0400C. From: Mike Bartman <omni@foolie.omniphile.com>M Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?.8 Message-ID: <fgjvmvosdko5k2lvohq699o94cvtil6bmk@4ax.com>  F On 23 Sep 2003 02:19:59 GMT, bill@cs.uofs.edu (Bill Gunshannon) wrote:  L >I'll say it once more... You can't solve social problems with technological >solutions.   F Sure you can!  There is no personal problem so severe that it can't be9 solved by a suitable application of high explosives!  :^)m   -- Mikee@ ----------------------------------------------------------------=   To reply via e-mail, remove the 'foolie.' from the address.m%   I'm getting sick of all the SPAM...r@ ----------------------------------------------------------------   ------------------------------  % Date: Tue, 23 Sep 2003 00:43:28 -0400). From: Mike Bartman <omni@foolie.omniphile.com>M Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?.8 Message-ID: <bijvmv0psgpf71dm2pieeigu9uln7t0n35@4ax.com>  F On 22 Sep 2003 21:41:57 -0500, Kilgallen@SpamCop.net (Larry Kilgallen) wrote:  e >In article <bkoakf$42ov5$1@ID-135708.news.uni-berlin.de>, bill@cs.uofs.edu (Bill Gunshannon) writes:i5 >> In article <ZuednaI5fOd9O_KiXTWJiw@metrocast.net>,T0 >> 	"Bill Todd" <billtodd@metrocast.net> writes: >>> M >>> I'd vote for that $.01 per email fee that was proposed:  I think it wouldtM >>> stop most spammers in their tracks, at least to the degree of making themrP >>> select for people who at least *might* be interested in their product rather >>> than just spam the world.  >> oI >> And how is it proposed to do this?  How do you know how many emails my  >> machine sends out?l >,> >Receiving ISPs charge the backbone for each message received.@ >Backbones charge each other (and sending ISPs) for each message
 >received.  > How do any of these know that a given packet is part of a mail< message?  The port number used?  What about those who run on@ non-standard ports, or over other protocols for part of the trip (web-mail for instance)?  B What if the sender isn't an ISP?  I run my own mailserver...my ISPB isn't involved except in moving packets at the TCP/IP level.  Am I supposed to bill myself?  B What will be your reaction when you find that your home system got? infected by a virus that sent a commercial e-mail to 27 million3F addresses while you were at the beach over the weekend?  Will you just, pay the $270,000.00 bill your ISP sends you?  ? There may be a solution to the spam problem, but that isn't it..   -- Miket  @ ----------------------------------------------------------------=   To reply via e-mail, remove the 'foolie.' from the address.e%   I'm getting sick of all the SPAM... @ ----------------------------------------------------------------   ------------------------------  % Date: Tue, 23 Sep 2003 01:29:39 -0400i* From: JF Mezei <jfmezei.spamnot@istop.com>M Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?l) Message-ID: <3F6FDA1B.359A6B91@istop.com>i   Mike Bartman wrote:oA > There may be a solution to the spam problem, but that isn't it.a  G Educate ALL users to delete spam message before even reading them. As aiM result, the business of sending spam will not be viable since they woN't sell < products or generate clicks to some web site or some advert.  M Another issue is that of viruses. It would be best to simply educate everyonerN to simply avoid anything microsoft. But if that won't happen, there isn't muchB to be done to prevent viri since microsoft is such an easy target.  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Months x--   UNLIMITED DOWNLOAD   ------------------------------    Date: 22 Sep 2003 13:09:52 -0700) From: jamliz01@peoplepc.com (James Faulk)tO Subject: Re: Search all Files for any Lines containing specific set of Strings.r= Message-ID: <217fe68d.0309221209.15cd3e4d@posting.google.com>r  F Thanks to each of you that replied.  I will review each suggestion and> determine which is most appropriate for my search.  Additional comments will be welcomed.   ------------------------------  % Date: Mon, 22 Sep 2003 20:18:04 -050061 From: "David J. Dachtera" <djesys.nospam@fsi.net>d+ Subject: Re: source for 431 pin CPU socket?l' Message-ID: <3F6F9F4C.B52034CD@fsi.net>    Bob Koehler wrote: > ^ > In article <87pthwxf05.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes: > >rF > > If you do get one Bob, how are you going to remove the old one and8 > > re-solder the new one in? What sort of system is it? > G >    I've removed the top layer of the ZIF socket and have found that I 7 >    can remove individual pins.  It will take a while.  > H >    I don't recall the model of this one of the top of my head (it's at >    home and I'm not).u > B >    OBTW, it's not AMD 431, it's AMP and it labeled "431 socket".  E Do you have a desoldering tool? Spring-loaded sucker - heat the jointDG until the solder flows then suck it away. Let it cool and break teh pin G loose with a small screw driver or something else. Do that on each pin,sB being careful to cool the board so it doesn't heat the surrounding! components. Kinda tedious, but...a   -- t David J. Dachtera  dba DJE Systemsh http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Mon, 22 Sep 2003 20:29:23 -0500e1 From: "David J. Dachtera" <djesys.nospam@fsi.net>o, Subject: Re: Strange problem with DLT drives' Message-ID: <3F6FA1F3.E5C5B2C8@fsi.net>t   Malcolm Dunnett wrote: > J >   I've noticed an odd problem with some of our DLT tape drives recently. > N >   I run a VMS backup job on the drive. I do a compare pass. Everything looksN > fine - BACKUP reports no errors - no "clean tape drive" lights on the drive.N > But when I reload the tape into the same drive it bounces around for a whileJ > trying to load the tape - then finally loads it but with the "clean tapeP > drive" light on. When I attempt to mount the tape mount VMS tells me the labelO > is blank. Any attempt to read the tape produces many minutes of tape spinning  > followed by a parity error.e > M >   I can then take this tape to a different drive and it will mount and reade > just fine. > ) >   Anyone else seen something like this?i   Yes.    What DLT drives are you running?  E I saw this on Quantum SuperDLT-320s. Turned out to be the SCSI cablese7 attaching the drives to the NSRs (believe it, or else).    -- e David J. Dachterat dba DJE SystemsS http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/e   ------------------------------  % Date: Mon, 22 Sep 2003 20:48:16 -0500:1 From: "David J. Dachtera" <djesys.nospam@fsi.net>B, Subject: Re: Strange problem with DLT drives' Message-ID: <3F6FA660.3B8B3086@fsi.net>r   Malcolm Dunnett wrote: > ) > In article <3F6FA1F3.E5C5B2C8@fsi.net>,/8 >    "David J. Dachtera" <djesys.nospam@fsi.net> writes: >  > > Malcolm Dunnett wrote: > >>M > >>   I've noticed an odd problem with some of our DLT tape drives recently.e > >>Q > >>   I run a VMS backup job on the drive. I do a compare pass. Everything looks Q > >> fine - BACKUP reports no errors - no "clean tape drive" lights on the drive.uQ > >> But when I reload the tape into the same drive it bounces around for a while M > >> trying to load the tape - then finally loads it but with the "clean tapebS > >> drive" light on. When I attempt to mount the tape mount VMS tells me the labeltR > >> is blank. Any attempt to read the tape produces many minutes of tape spinning  > >> followed by a parity error. > >>P > >>   I can then take this tape to a different drive and it will mount and read > >> just fine.r > >>, > >>   Anyone else seen something like this? > >h > > Yes. > > $ > > What DLT drives are you running? > >  > D >    I'm seeing it on a DLT 2000XT and a DLT 4000. Both are directly- > connected to (different) Alphaserver DS20s.+  @ Hhmmm... Dunno. Next guess time is see if you can clean the headG manually. Remember to use a non-aqueous solvent. Water will destroy the-- adhesives that hold the head layers together.3   -- - David J. Dachterac dba DJE Systemsd http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------    Date: 22 Sep 2003 13:56:54 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)? Subject: Re: strangeness with temporary text files used by MAIL < Message-ID: <b096a4ee.0309221256.956bef7@posting.google.com>  m John Laird <nospam@laird-towers.org.uk> wrote in message news:<tahtmvsl6q29e6p2lb21k48udk5uaa5580@4ax.com>... K > On Mon, 22 Sep 2003 05:41:54 +0000 (UTC), helbig@astro.multiCLOTHESvax.de,3 > (Phillip Helbig---remove CLOTHES to reply) wrote:b >   > >$ sea sys$login:*.edt summary* > >%SEARCH-I-NOMATCHES, no strings matched > M > "set nosu" is sufficient, so your search is too limiting.  Also, edtini.edtcH > is invoked in the default directory, if found, not sys$login (unless aI > logical points there).  Use of set nosummary would seem the most likelye& > explanation for the lack of message.  F There is also no guarantee that a user will use the .EDT file type forE his EDT init file, or for that matter, there is no guarantee that therC startup file will be in SYS$LOGIN. Check if he has EDTSYS or EDTINI  defined as logical names.r     Disclaimer: JMHO Alan E. Feldman    ------------------------------  % Date: Mon, 22 Sep 2003 22:05:56 +0100j- From: John Laird <nospam@laird-towers.org.uk> ? Subject: Re: strangeness with temporary text files used by MAIL 8 Message-ID: <luoumv8ce3q2m0svslmc6ndfq0f362k10e@4ax.com>  G On 22 Sep 2003 13:56:54 -0700, spamsink2001@yahoo.com (Alan E. Feldman)  wrote:  n >John Laird <nospam@laird-towers.org.uk> wrote in message news:<tahtmvsl6q29e6p2lb21k48udk5uaa5580@4ax.com>...L >> On Mon, 22 Sep 2003 05:41:54 +0000 (UTC), helbig@astro.multiCLOTHESvax.de4 >> (Phillip Helbig---remove CLOTHES to reply) wrote: >> C! >> >$ sea sys$login:*.edt summaryi+ >> >%SEARCH-I-NOMATCHES, no strings matchedr >> VN >> "set nosu" is sufficient, so your search is too limiting.  Also, edtini.edtI >> is invoked in the default directory, if found, not sys$login (unless a.J >> logical points there).  Use of set nosummary would seem the most likely' >> explanation for the lack of message.- >-G >There is also no guarantee that a user will use the .EDT file type for F >his EDT init file, or for that matter, there is no guarantee that theD >startup file will be in SYS$LOGIN. Check if he has EDTSYS or EDTINI >defined as logical names.  L Phillip's post (which I heavily snipped) did show no *edt* logicals defined.L Assuming he was logged on as the user in question, that really only left the "local" edtini.edt possibility.    --   John Mail john rather than nospam...o   ------------------------------   Date: 23 Sep 03 00:26:00 +0200) From: p_sture@elias.decus.ch (Paul Sture)P? Subject: Re: strangeness with temporary text files used by MAILm) Message-ID: <7+S9SRjRY7m8@elias.decus.ch>e  h In article <luoumv8ce3q2m0svslmc6ndfq0f362k10e@4ax.com>, John Laird <nospam@laird-towers.org.uk> writes:I > On 22 Sep 2003 13:56:54 -0700, spamsink2001@yahoo.com (Alan E. Feldman)  > wrote: > o >>John Laird <nospam@laird-towers.org.uk> wrote in message news:<tahtmvsl6q29e6p2lb21k48udk5uaa5580@4ax.com>... M >>> On Mon, 22 Sep 2003 05:41:54 +0000 (UTC), helbig@astro.multiCLOTHESvax.deP5 >>> (Phillip Helbig---remove CLOTHES to reply) wrote:9 >>> " >>> >$ sea sys$login:*.edt summary, >>> >%SEARCH-I-NOMATCHES, no strings matched >>> O >>> "set nosu" is sufficient, so your search is too limiting.  Also, edtini.edtIJ >>> is invoked in the default directory, if found, not sys$login (unless aK >>> logical points there).  Use of set nosummary would seem the most likelyp( >>> explanation for the lack of message. >>H >>There is also no guarantee that a user will use the .EDT file type forG >>his EDT init file, or for that matter, there is no guarantee that the E >>startup file will be in SYS$LOGIN. Check if he has EDTSYS or EDTINIt >>defined as logical names.i > N > Phillip's post (which I heavily snipped) did show no *edt* logicals defined.N > Assuming he was logged on as the user in question, that really only left the! > "local" edtini.edt possibility.d >    Not quite true:t   From HELP EDIT/EDT/COMMAND  *        $ EDIT/COMMAND=XEDTINI.EDT MEMO.DAT  A        If you do not include the /COMMAND=command file qualifier,-B        EDT looks for the EDTSYS logical name assignment. If EDTSYSC        is not defined, EDT processes the systemwide startup command D        file SYS$LIBRARY:EDTSYS.EDT. If this file does not exist, EDTE        looks for the EDTINI logical name assignment. If EDTINI is notaG        defined, EDT looks for the file named EDTINI.EDT in your defaultgH        directory. If none of these files exists, EDT begins your editing$        session in the default state.  C        To prevent EDT from processing either the systemwide startupgE        command file or the EDTINI.EDT file in your default directory,E/        use the /NOCOMMAND qualifier as follows:o           $ EDIT/NOCOMMAND MEMO.DAT  E I have a feeling that many moons ago (V2), SYS$LIBRARY:EDTSYS.EDT waso. elsewhere (SYS$MANAGER ?), but could be wrong.   ------------------------------  % Date: Mon, 22 Sep 2003 20:01:58 +0200w( From: "H Vlems" <hvlems.nieuw@zonnet.nl>7 Subject: Re: TCPIP:DNS Dual interfaces and domain namesm9 Message-ID: <bkndhq$3m0u2$1@ID-143435.news.uni-berlin.de>u  / "John Duh" <jduh@nobody.com> schreef in berichtp$ news:3F6EAB4A.A179D185@nobody.com...L > OK, I have node cake.chocolate.com at primary address 10.0.0.10 . But thatJ > node also supports a SLIP interface with IP address 10.1.0.11 on the vax side.i >eF > When I do a traceroute from my PDA (who is 10.1.0.20 attached to the	 10.1.0.10eI > SLIP port), it is unable to backtranslate 10.1.0.11, the first "hop" ine the: > route to a remote host.v >d > Question:) >2G > Is there any gotchas to adding the reverse transation of 10.1.0.11 to9I > "cake.chocolate.com" in the in0-addr_arpa.db file ? (there would be two0 IPs G > pointing to same domain name). Is this expected practice, or shoudl Ib leavel; > the vax-side of the SLIP interface undefined in the dns ?S >2H I'd think that the problem with traceroute is an indication of a routing  problem rather than a DNS error. >P > Second issue:0K > from the PDA, if I try to get the IP address of cake.chocolate.com, I get. its.L > primary IP address of 10.0.0.10 instead of the slip side IP address. Would? > this cause any problems ? If so, are there ways around this ?   J IP addresses are related to an interface, not to the host. For a host withF just one interface there's no difference but on a system with multipleI interfaces it starts to matter. DECnet issues addresses to hosts, it doesyI not matter how many NIC's or other DECnet capable interfaces it has, theya< all respond to one address and hence one name is sufficient.L IP assignes an address to each interface, the address plus the mask provides' the routing information. The DNS server G translates the appropriate name to an IP address. In your case the SLIP K address must be in a different network to be routable. That means that yourvI mask must be at least  8 bits, 255.0.0.0 in order to make the source nodeRL know that 10.0.0.0 and 10.1.0.0 are two different networks and that a router  is needed to deliver the packet. >r > Third issue:K > On my old mac, I have a default route pointing to the router for anythingoL > outside of 10.0.0.*, so when the mac tries to access the PDA at 10.1.0.20, theiD > call goes to the router. In the router, I have defined a route via$ > cake.chocolate.com so things work.  8 So the Mac has a route for 0.0.0.0 mask 0.0.0.0 , right? >yL > However, what sort of routing configuration (gated, routed ?) would I need onK > cake.chocolate.com so that it would broadcast a route to the PDA that thei macsK > would understand resulting in the mac routing a call to the PDA duirectlys via K > the VAX, bypassing the router ?  Right now, I only have static routes, sor I . > guess they don't get broadcasted on the lan.  G Routed (or RIP version 1) should handle this quite well. OTOH you use aiL class A network, probably with a 24 bit mask so your LAN may be a lot larger! thanis obvious from your post :-)u   ------------------------------  % Date: Mon, 22 Sep 2003 16:35:12 -0400e* From: JF Mezei <jfmezei.spamnot@istop.com>7 Subject: Re: TCPIP:DNS Dual interfaces and domain namesy) Message-ID: <3F6F5CFC.22985D3D@istop.com>t   H Vlems wrote:J > I'd think that the problem with traceroute is an indication of a routing" > problem rather than a DNS error.  L If I can ping the PDA from any machine on my lan, and if my PDA can ping anyI machine on my lan, doesn't this mean that the routing is properly setup ?   N On VMS, I can manually set routes so that BIKE knows that to send to PSION, itN must route through VELO. But on the old MAC, I don't have the ability to setup static routes.  G So the way I did it is tell the mac it is in a 10.0.*.* subnet and send K anything else to the router. (default gateway field).  And on the router, I0J defined a static route to 10.1.*.* that goes via VELO. It works, but is an extra step.u  L Just wondering what is needed to get VELO to broadcast the route to 10.1.*.*K so that I would not have to hardcode that route as static routes everywhereeK else, and so that hopefully, the mac would see the route broadcast and know  how to send to 10.1.*.*t    I > Routed (or RIP version 1) should handle this quite well. OTOH you use a N > class A network, probably with a 24 bit mask so your LAN may be a lot larger# > thanis obvious from your post :-)e  M No, it only has 5 nodes and a router. But Since I am allowed to use 10.*.*.*,oN it was much easier to take big chunks to define my subnets instead of splitingG  10.0.0.* into two subnets and dealing with ugly bit masks and lat byteyB numbering limits (eg: from 0-31 is one subnet, 32 to 255 another).  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Month  x--   UNLIMITED DOWNLOAD   ------------------------------    Date: 22 Sep 2003 11:49:52 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)(T Subject: Re: There is no Joy in Sunville -- mighty Billy has struck out [on his own]= Message-ID: <cf15391e.0309221049.5a44257d@posting.google.com>S   Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bkmlif$gse$1@new-usenet.uk.sun.com>...,? > What you probably also didn't notice is that the whole markets> > dipped by ~5% on Sept 9-10. 3% below the market would hardly@ > prove your point in fact it just illustrates that you are just > grasping at straws.   E No, the "whole market" didn't dip ~5% on Sept. 9-10.  One can readily A see that the S&P 500 and Dow Jones Industrial average didn't drop.  much, judging from Yahoo's data:K http://finance.yahoo.com/q/bc?t=3m&s=SUNW&l=on&z=m&q=l&c=&c=%5ESPX&c=%5EDJI7  D And even a drop of 3% of your market cap. in a single day would sillC seem to indicate the market felt the news was significant, and bad.t  B > You still havn't explained someone as dilligent as you failed to, > notice that Bill is going for over a week.  C I have a job as a engineer at HP.  I don't spend full-time watchingaC Sun, and I don't read (and, even more significantly, I don't spread-E FUD in) comp.unix.solaris or any of the other Sun-related newsgroups;.F I typically only notice a news item about Sun when someone else within HP points it out.a   ------------------------------  # Date: Tue, 23 Sep 2003 00:30:10 GMTd2 From: rdeininger@mindspring.com (Robert Deininger)3 Subject: Re: Trying to figure out license transfers L Message-ID: <rdeininger-2209032038590001@user-uinj5tu.dialup.mindspring.com>  = In article <ea44f5a1.0309220602.6aaae26a@posting.google.com>,e& tadamsmar@yahoo.com (Tom Adams) wrote:  > >rdeininger@mindspring.com (Robert Deininger) wrote in messageH news:<rdeininger-1909032104470001@user-105n86e.dialup.mindspring.com>...@ >> In article <ea44f5a1.0309191008.3d911fcb@posting.google.com>,) >> tadamsmar@yahoo.com (Tom Adams) wrote:r >> b3 >> >Can you still get DS10/466 systems new from HP?i >> >C >> >So far, the only quote that I have is for a refurbished system.e >> >7 >> >I thought I saw a new system listed on the HP site.e >> rM >> This question has come up several times lately.  I just (re)checked the HP 5 >> Alphaserver web page, and the DS10 is still shown.k >> o) >> http://h18002.www1.hp.com/alphaserver/  >> wK >> Are you able to talk to an above-room-temperature person at HP and get a & >> direct answer to the DS10 question? >fF >A while back, I got an unsolicted email from a guy at HP that said weE >should consider him our HP point of contact.  He was the first guy I C >send a request for quotation to.  He never acknowledged my email. g< >Meanwhile, I have been in touch with some resellers.  Had a! >conversation with an Island rep.S  I I think it would be fair to post his name.  Something might have happenedA8 beyond his control, or he might have just blown you off.  M Island is very good to work with, but they can't sell what they can't obtain.o  6 >I am still trying to figure out the licensing issues. > C >The Island guy says I should be able to just load licenses from myoF >dead box, but get the idea from the HP site that I am suppose to send >in a $300 dollar- >application, at least.   G It really depends on the license terms.  Do you still have the original  paperwork from the old system?  E Different types of licences have different terms.  Some are tied to atF single machines, but can be transfered with the machine to a new ownerE (for a fee).  Some can be redeployed at the owner's discretion.  Some-H licenses can't be transfered at all, but have substantial trade-in value toward new licences.   >j@ >I am trying to go from a Alphaserver 400/233 to a DS10/466.  WeD >already have one DS10/466.  I am trying to figure out if I can justC >buy a base VMS license for the new box and then load paks from the ! >dead box and configure my way to F >the system that I want.  We already get the binary distributions from >HP/Compaq.u  I If the AS 400/233 licenses are (legally) transferrable to another system,o6 they might not have the right number of license units.  I Do you have your heart set on a 466 MHz system?  Those _are_ EOL, but 600eJ MHz systems are still (supposed to be) available.  I expect the (used) 466H would be significantly cheaper, so that may settle the question for you.    E >The ADL license pak look almost identical. Same number of units. But  >theF >product number is *3E vs *BE and the Product name has CONC instead of >ADL.9 >1C >Also, what does the "8" mean in OpenVMS Conc 8 LIC".  Is that an 8 C >user license?  When I test the system, I can have many more than 8 E >users.  The interactive limit can be set very high.  It is currently- >at 64.- >-G >The VOLSHAD license pak has the same model number and unit count.  Butd >theC >product name is a bit different.  Same situation with the UCX pak.   E Check the Software Product Descriptions (SPDs) for VMS and Shadowing, J etc.  The license options are described there, including old license types that are no longer sold.    > >Beyond that, it is confusing because the DS10 seems to have aC >different set of paks from the 400.  I cannot tell what I need.  IsC >wonder if all these PAKs come if you buy the base license.  I needgA >Motif and DECnet phase IV.  Do these come with the base license?o  I The Base license doesn't include Motif, networks, or users.  But most newnA VMS systems are bundled with Base license and an add-on package. aD Depending on the year of purchase and the type of system, the bundleF varies in name and contents.  Motif is typically bundled with anythingE sold as a workstation.  NET-APP-SUPPORT-nnn bundles typically include J networks, Motif, and some form of clustering.  (Bigger nnn bundles includeG more stuff.  Cheaper, smaller systems have a smaller nnn.)  More recenttG server systems include EIP (Enterprise Integration Package), which is a J bunch of individual PAKs for products -- similar to the older N-A-S-nnn in concept, but different name.      M >> Did you fill out the "buy online" form on the Alpha web page?  Did you get ! >> any response?  Was it helpful?r >aG >Have not tried that.  I may contact my "point of contact" by phone andmD >see if he will send a quote.  I would like to have 3 quotes for the >purchasing process.  D I filled in the form late Friday, and have heard nothing so far.  OfI course, my email box has been filled up for most of that time.  We'll see.C if HP will expend effort to sell an upgrade to my DEC 3000 systems.h  F If you really want a used system, it may take a while for them to find? it.  That's still no excuse for not responding to your request.r  s- >> Are you trying to buy a DS10 with OpenVMS?t >a >Yes.   I Buying a refurb system without licenses, and then adding on licenses fromuH HP, might cost more than buying a new system bundled with licenses.  But1 how will you know if they won't give you a quote?a   ------------------------------  + Date: Mon, 22 Sep 2003 20:14:07 +0000 (UTC)e: From: "Murray Styman" <murray.nospam.styman@btintrnet.com> Subject: Virtul Tape Driver 2 Message-ID: <bknl6e$hhd$1@hercules.btinternet.com>  : Does anyone have or know of a virtual tape driver for vms?  $ Preferably freeware and with source.   ------------------------------  % Date: Mon, 22 Sep 2003 15:57:52 -0500e( From: Wayne Sewell <wayne@tachysoft.com> Subject: Re: Virtul Tape Driveru0 Message-ID: <00A264AA.BE2B36BF.10@tachysoft.com>   >t; >Does anyone have or know of a virtual tape driver for vms?  > % >Preferably freeware and with source.f >t >   O Depends on what you need it for.  Do you need the driver itself for purposes of L changing it, or would an existing virtual tape drive work for you?  SoftwareM Partners has three different virtual tape drives layered on the same driver. pM Thruway provides a remote tape drive across a unclustered network.  TSHAD/CMPiO provides a tape shadowing device where records written to the virtual drive areoO replicated on two or more physical tape drives, allowing you to create multiplecK copies of a backup saveset with one backup command.  TS11 is a virtual tape G drive that I use for testing of tapesys backups.  It currently is not aaL product, but of course it could be turned into one with sufficient interest.  0 Not freeware and source not included, of course.   Wayne O ===============================================================================sN Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html   sO ===============================================================================lN Butler:"Gentlemen!"  Curly(as he and other Stooges look around):"Who came in?"   ------------------------------  # Date: Mon, 22 Sep 2003 21:39:57 GMTo" From:   VAXman-  @SendSpamHere.ORG Subject: Re: Virtul Tape Driverl0 Message-ID: <00A264B9.050C740C@SendSpamHere.ORG>  o In article <bknl6e$hhd$1@hercules.btinternet.com>, "Murray Styman" <murray.nospam.styman@btintrnet.com> writes:a; >Does anyone have or know of a virtual tape driver for vms?3 >B% >Preferably freeware and with source.n >   @ So that you can perform virtual backups and virtual restores? ;)  " What are you trying to accomplish?   --L VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)COM            g5   "Well my son, life is like a beanstalk, isn't it?" a   ------------------------------  + Date: Tue, 23 Sep 2003 00:22:14 +0000 (UTC) : From: "Murray Styman" <murray.nospam.styman@btintrnet.com> Subject: Re: Virtul Tape Driver . Message-ID: <bko3nm$jv$1@titan.btinternet.com>  L Our developers are constantly using the tape units as part of the applicaionG they are developing and keep dismounting the tapes which means that ourtI operators have to reload.  Part of there job, but a pain in the arse whenpI you have to do it every 5 minutes.  They also seem to get there processescJ into varying states which lock up the tape unit and often the only way out
 is a reboot!.f  @ I work in a very security contious environment and giving access5 un-necessaryily to the tape units is not a good idea.-  J I thought that a virtual tape unit would be a good solution.  No where forK the test data to go, reduced access to the real tape units, faster responseP/ and if they lock it up jut create a new device.i    5 "Wayne Sewell" <wayne@tachysoft.com> wrote in messager* news:00A264AA.BE2B36BF.10@tachysoft.com... > >n= > >Does anyone have or know of a virtual tape driver for vms?  > >a' > >Preferably freeware and with source.n > >  > >i >TE > Depends on what you need it for.  Do you need the driver itself fort purposes ofiD > changing it, or would an existing virtual tape drive work for you? SoftwareF > Partners has three different virtual tape drives layered on the same driver.tD > Thruway provides a remote tape drive across a unclustered network.	 TSHAD/CMPnG > provides a tape shadowing device where records written to the virtuala	 drive are H > replicated on two or more physical tape drives, allowing you to create multipleH > copies of a backup saveset with one backup command.  TS11 is a virtual tapeI > drive that I use for testing of tapesys backups.  It currently is not aoD > product, but of course it could be turned into one with sufficient	 interest.  >n2 > Not freeware and source not included, of course. >K > Waynec >.L ============================================================================ ===@: > Wayne Sewell, Tachyon Software Consulting  (281)812-0738 wayne@tachysoft.comn: > http://www.tachysoft.com/www/tachyon.html and wayne.html >eL ============================================================================ === K > Butler:"Gentlemen!"  Curly(as he and other Stooges look around):"Who camem in?"   ------------------------------  + Date: Tue, 23 Sep 2003 00:22:35 +0000 (UTC) : From: "Murray Styman" <murray.nospam.styman@btintrnet.com> Subject: Re: Virtul Tape Drivero2 Message-ID: <bko3ob$bra$1@hercules.btinternet.com>  L Our developers are constantly using the tape units as part of the applicaionG they are developing and keep dismounting the tapes which means that our I operators have to reload.  Part of there job, but a pain in the arse wheniI you have to do it every 5 minutes.  They also seem to get there processesuJ into varying states which lock up the tape unit and often the only way out
 is a reboot!.   @ I work in a very security contious environment and giving access5 un-necessaryily to the tape units is not a good idea.   J I thought that a virtual tape unit would be a good solution.  No where forK the test data to go, reduced access to the real tape units, faster response / and if they lock it up jut create a new device.n  , <VAXman- @SendSpamHere.ORG> wrote in message* news:00A264B9.050C740C@SendSpamHere.ORG...D > In article <bknl6e$hhd$1@hercules.btinternet.com>, "Murray Styman", <murray.nospam.styman@btintrnet.com> writes:= > >Does anyone have or know of a virtual tape driver for vms?l > > ' > >Preferably freeware and with source.  > >  >iB > So that you can perform virtual backups and virtual restores? ;) >i$ > What are you trying to accomplish? >I > --2 > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM >t6 >   "Well my son, life is like a beanstalk, isn't it?"   ------------------------------  % Date: Mon, 22 Sep 2003 20:25:32 -0500l( From: Wayne Sewell <wayne@tachysoft.com> Subject: Re: Virtul Tape Driver,/ Message-ID: <00A264D0.22DC3792.2@tachysoft.com>    >sM >Our developers are constantly using the tape units as part of the applicaionr >they are developing d  0 Ah, got it.  That's exactly what I use TS11 for.  4 >and keep dismounting the tapes which means that ourJ >operators have to reload.  Part of there job, but a pain in the arse when% >you have to do it every 5 minutes.     D I would have this problem when testing tapesys with real drives, butK fortunately the sysbak parameter file allows you to specify /nounload.  Not-M normally something you would do with production tapesys, but really handy forBM testing, especially since I'm a geezer and my real tape drives are upstairs.  < :-)  So I can run a dozen tests and the tape remains loaded.  M I have the pseudo drives set up to unload after use, but I don't have to loadBO those anyway.  I have a virtual jukebox that the virtual tape drives reside in,mF so I can use MRU and JB to load the virtual drives automatically.  :-)    & >They also seem to get there processesK >into varying states which lock up the tape unit and often the only way out- >is a reboot!. >-A >I work in a very security contious environment and giving access 6 >un-necessaryily to the tape units is not a good idea. > K >I thought that a virtual tape unit would be a good solution.  No where for L >the test data to go, reduced access to the real tape units, faster response0 >and if they lock it up jut create a new device. >R  F Yes, I sometimes wind up with a dozen dead pseudodrives before finally rebooting the test machine.e  M In addition to the speed, I have added some other features to ts11 to make itaL useful for testing applications that write to tape drives.  For instance, toL more closely simulate a tape drive, I can deliberately induce parity errors,N fatal controller errors, or any other tape drive error at any time to test theO software's error handling.  The virtual tapes can be set to different lengths. lL You can make them short to test end-of-volume processing, or endless so thatN any amount of data will go into them.  I was using this recently because I hadK a problem during the backup recording pass.  I wanted a big backup so there L would be lots of files to record, but with ts11 I didn't really have to wait" for the actual backup to complete.  M TS11 even simulates the LCD panel for the tape drive.  Tapesys writes to thissM when loading tapes, and with ts11 I can test that, even though my actual tapea drives don't have displays.    WayneaO =============================================================================== N Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html   kO ===============================================================================cN Butler:"Gentlemen!"  Curly(as he and other Stooges look around):"Who came in?"   ------------------------------    Date: 22 Sep 2003 20:11:15 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)0 Subject: Re: Virtul Tape Driver 3 Message-ID: <IFf55FvrKH1$@eisner.encompasserve.org>   k In article <bko3nm$jv$1@titan.btinternet.com>, "Murray Styman" <murray.nospam.styman@btintrnet.com> writes:iN > Our developers are constantly using the tape units as part of the applicaionI > they are developing and keep dismounting the tapes which means that our-K > operators have to reload.  Part of there job, but a pain in the arse whenF$ > you have to do it every 5 minutes.  B Have they tried specifying MNT$M_NOUNLOAD within the application ? It is described at  E 	http://h71000.www7.hp.com/doc/731FINAL/4527/4527pro_081.html#jun_389s  G They could make inclusion of that bit depend on presence of some systemlF logical name used only for development, on the UIC under which the jobH is running, or some other indication of what constitutes the development environment.   ------------------------------  # Date: Mon, 22 Sep 2003 19:01:23 GMTu2 From: "Ken Farmer" <KFarmer@NOSPAM.SpyderByte.com>' Subject: VMS mentioned in Linux article : Message-ID: <7GHbb.6164$vq1.4687@twister.southeast.rr.com>   LinuxInsider.com1 http://www.linuxinsider.com/perl/story/31634.html0  J ... "You can correlate systems knowledge with age," explains Avery Lyford,L CEO of Linuxcare, which develops management software for Linux environments.K "It's a gross generalization, but if you talk to someone in their 20s, theylK know Linux; in their 30s, they know Microsoft; in their 40s, Unix; in their H 50s, big systems like VMS [Virtual Memory System]." So in theory, LyfordL says, you could gauge your Linux migration costs by figuring out the average' age of your system administrators.  ...s  # I thought this was a funny analogy.i   -- Kenneth Farmer  <><o SpyderByte.com    EnterpriseUnix.org  |  Tru64.org OpenVMS.org  |  dcl.OpenVMS.orgm5 EnterpriseLinux.org  |  LinuxHPC.org  |  BigBlueLinux:   ------------------------------  % Date: Mon, 22 Sep 2003 17:26:17 -0400t* From: JF Mezei <jfmezei.spamnot@istop.com>+ Subject: Re: VMS mentioned in Linux articleo' Message-ID: <3F6F68F1.14FA62@istop.com>    Ken Farmer wrote: 3 > http://www.linuxinsider.com/perl/story/31634.html  > L > ... "You can correlate systems knowledge with age," explains Avery Lyford,N > CEO of Linuxcare, which develops management software for Linux environments.M > "It's a gross generalization, but if you talk to someone in their 20s, theyeM > know Linux; in their 30s, they know Microsoft; in their 40s, Unix; in their 5 > 50s, big systems like VMS [Virtual Memory System]."v  N Since VMS is younger than Unix by about 10 years, you'd think that the 50 yearM old would be Unix types, and it woudl be the 40 year olds that are VMS types.d  N On the other hand, you could draw the picture that as you become more and moreM experienced, you evolved from Linux to VMS and by the time to get to VMS, youa( gather experience from Windows and Unix.  L What I find most interesting though is how the marketing portrays Linux as a; different system than Unix. Isn't Linux just another Unix ?,  2 x--   100 Proof News - http://www.100ProofNews.com< x--   3,500+ Binary NewsGroups, and over 90,000 other groups/ x--   Access to over 800 Gigs/Day - $8.95/Montht x--   UNLIMITED DOWNLOAD   ------------------------------   Date: 23 Sep 03 00:04:06 +0200) From: p_sture@elias.decus.ch (Paul Sture)j+ Subject: Re: VMS mentioned in Linux articlei) Message-ID: <QcCvMvUg919Y@elias.decus.ch>   o In article <7GHbb.6164$vq1.4687@twister.southeast.rr.com>, "Ken Farmer" <KFarmer@NOSPAM.SpyderByte.com> writes:r > LinuxInsider.com3 > http://www.linuxinsider.com/perl/story/31634.html" > L > ... "You can correlate systems knowledge with age," explains Avery Lyford,N > CEO of Linuxcare, which develops management software for Linux environments.M > "It's a gross generalization, but if you talk to someone in their 20s, they M > know Linux; in their 30s, they know Microsoft; in their 40s, Unix; in theireJ > 50s, big systems like VMS [Virtual Memory System]." So in theory, LyfordN > says, you could gauge your Linux migration costs by figuring out the average) > age of your system administrators.  ...e > % > I thought this was a funny analogy.  >    Oi! I ain't 50 yet.. :-)    ------------------------------   Date: 22 Sep 2003 22:59:18 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)+ Subject: Re: VMS mentioned in Linux articler9 Message-ID: <bknus5$3t1ju$3@ID-135708.news.uni-berlin.de>R  ' In article <3F6F68F1.14FA62@istop.com>,-- 	JF Mezei <jfmezei.spamnot@istop.com> writes:n > Ken Farmer wrote:a4 >> http://www.linuxinsider.com/perl/story/31634.html >> rM >> ... "You can correlate systems knowledge with age," explains Avery Lyford, O >> CEO of Linuxcare, which develops management software for Linux environments. N >> "It's a gross generalization, but if you talk to someone in their 20s, theyN >> know Linux; in their 30s, they know Microsoft; in their 40s, Unix; in their6 >> 50s, big systems like VMS [Virtual Memory System]." > P > Since VMS is younger than Unix by about 10 years, you'd think that the 50 yearO > old would be Unix types, and it woudl be the 40 year olds that are VMS types.e  D Actually, I think he was really refering to Big Iron VM Systems like% 4300 family and 3080 family from IBM.i   > P > On the other hand, you could draw the picture that as you become more and moreO > experienced, you evolved from Linux to VMS and by the time to get to VMS, yous* > gather experience from Windows and Unix. > N > What I find most interesting though is how the marketing portrays Linux as a= > different system than Unix. Isn't Linux just another Unix ?)  I If you consider reinventing the wheel (badly) with no recourse to severalaH decades of serious computer science research as being the same, then youF are probably right.  I doubt anyone who has seriously worked with real Unix would agree with you.   bill   -- wJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   e   ------------------------------  # Date: Mon, 22 Sep 2003 23:33:58 GMT72 From: "Ken Farmer" <KFarmer@NOSPAM.SpyderByte.com>+ Subject: Re: VMS mentioned in Linux article26 Message-ID: <GFLbb.104$eB.37@twister.southeast.rr.com>  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in message ! news:3F6F68F1.14FA62@istop.com...t > Ken Farmer wrote: 5 > > http://www.linuxinsider.com/perl/story/31634.htmlS > >sL > What I find most interesting though is how the marketing portrays Linux as aA= > different system than Unix. Isn't Linux just another Unix ?n  J As far as basic functionality...looks like, smells like, taste like.  Must be Unix.   -- Kenneth Farmer  <><o SpyderByte.com    EnterpriseUnix.org  |  Tru64.org OpenVMS.org  |  dcl.OpenVMS.orgT$ EnterpriseLinux.org  |  LinuxHPC.org   ------------------------------   Date: 23 Sep 2003 01:04:49 GMT, From: bill@gw5.cs.uofs.edu (Bill Gunshannon)+ Subject: Re: VMS mentioned in Linux articlem9 Message-ID: <bko67h$4018j$1@ID-135708.news.uni-berlin.de>-  6 In article <GFLbb.104$eB.37@twister.southeast.rr.com>,5 	"Ken Farmer" <KFarmer@NOSPAM.SpyderByte.com> writes:e9 > "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messageo# > news:3F6F68F1.14FA62@istop.com...o >> Ken Farmer wrote:6 >> > http://www.linuxinsider.com/perl/story/31634.html >> >M >> What I find most interesting though is how the marketing portrays Linux asq > a2> >> different system than Unix. Isn't Linux just another Unix ? > L > As far as basic functionality...looks like, smells like, taste like.  Must
 > be Unix.  G A pile of dog poop looks like expensive french pate'.  You want that on C toast or a cracker?  An similarity between Linux and Unix is purelys accidental..   bill   -- eJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Mon, 22 Sep 2003 20:33:07 -0500h1 From: "David J. Dachtera" <djesys.nospam@fsi.net> + Subject: Re: VMS mentioned in Linux articleI' Message-ID: <3F6FA2D3.9A089E0F@fsi.net>    JF Mezei wrote:l > ! > Isn't Linux just another Unix ?e  G Yes and no. Linus did his best to emulate the UN*X kernel. Linux is not  UN*X and UN*X is not Linux.p   -- i David J. Dachtera  dba DJE Systemst http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------  # Date: Mon, 22 Sep 2003 17:30:15 GMTl' From: Don Sykes <anonymous@pacbell.net>m Subject: Re: VMS Upgrade Needed<+ Message-ID: <3F6F322C.D7C13D33@pacbell.net>J   JF Mezei wrote:a > J > > >AND the only reason to use James instead of HP/Compaq's TCPIP/SMTP isG > > >because I CAN'T MAKE THAT FACILITY SAY "NO SUCH USER" AND DROP THErL > > >CONNECTION, during the connect phase, when there really IS no such user
 > > >anymore!S >   > We had this discussion before. > L > The way I understand how the Digital TCPIP SMTP server works, the receiverO > merely receives the message with minimal processing and queues it to the SMTPtM > queues for later delivery. It is the SMTOP symbiont in that queue that does, > all the work.  > N > However, when you look at the control files that are generated, the receiverN > does parse the RFC822 header since the subject of the message as well as theE > "nice" To:  are stored in the control file as well as the received:r > information. > K > And the receiver is the one that does the RBL processing as well as relaytT > prevention since those are done during the SMTP negotiation before the DATA phase. > M > However, the receiver is given a list of "local" hosts to which relaying isoO > allowed.  Those hosts may or may not be part fo a cluster, hence the receiveraM > doesn't have the ability to verify the validity of a recipient because suche> > recipient doesn't necessarily reside on a node with the same4 > vmsmail_profile.data as the receiving smtp server.  H That should not be a problem. I could still have the option of providingH a file of valid users for system B even if they are not also valid usersE on system A (running the SMTP service). My complaint has been I'm note given that option.   > L > And there are security implications in allowing the receiver to validate a > username given in RCPT TO:O > since it would allow for some rogue process to start issuing tons of RCPT TO:T& > to gather a list of valid usernames.  G They can already do that, it just takes longer. But that should be easyeH to check after some number of attempts, drop any further connections, orH just use the delay tactic. And in any event there are an infinite numberE of user names. I just don't think this would be a realistic problem. S   -- A   Have VMS, Will Travelh Wire paladin, San Franciscod   (paladinATalphaseDOTcom)   ------------------------------  % Date: Mon, 22 Sep 2003 13:17:57 -0700 % From: Dean Woodward <deanw@rdrop.com>T Subject: Re: VMS Upgrade Neededi( Message-ID: <3F6F58F5.1020807@rdrop.com>   Don Sykes wrote:1 > They can already do that, it just takes longer.e  F Not necessarily. Some major ISPs are non-conformant and don't send "noI such user" messages- they just drop the incoming mail in the bit-bucket. II (When they get enough extra bits, they stir 'em around a bit, then stamp  F them onto a CD and send it out snail-mail, perpetuating the cycle. ;-)  E > And in any event there are an infinite number of user names. I just 0 > don't think this would be a realistic problem.  I I believe there is software that will take email addresses, separate the 9E usernames from the domains, then apply each username to each domain,  E against the odd chance there's a hit. If there isn't, no skin of the   spammer's back, eh?    ------------------------------  # Date: Mon, 22 Sep 2003 21:04:43 GMTn' From: Don Sykes <anonymous@pacbell.net>iA Subject: Re: We can stop these SPAMS by blocking all the SPAMBOTso+ Message-ID: <3F6F6470.125DDC92@pacbell.net>h   Dominique GEORGES wrote: >  > Hi,e > = > I'm also victim of this new virus/spam... as MANY otherone.e > A > I found, on a french PHP site, an explanation on how it works :rF > Robots like Google, ... scans all the sites, and store ALL the textsB > like x.y@a.b.  store it in a db, and finally sell the db to spam > users. > B > There is noway to stop it, but if we attack them by giving theseE > robots tons of invalid addresses, they wont be credible at all, andeD > their customers won't be interrested at all by buying these lists. >   ( I don't think this is a very good idea. F 1. How can you verify your giving these robots invalid addresses. EvenG if generated randomly some poor bastard could be overwhelmed with crap.aE 2. Emails, even invalid ones, can take up bandwidth which we all use.aH 3. Judging by the amount of stuff already sent out and largely unread, I< don't think they care if they include lots of bad addresses.  G > Here is the pointer of PHP script that detect the robot scanning your G > site.  If the robot is known as a kind one (a la Google), it does NOToE > provide him bad info, but if the robot is unknown, it automaticallye% > generates dummay email addresses...g > G > Have a look here, it's in french, but I think it's a good way to stope > these SPAMs. >  > http://anti-spam.nachix.com/ > 
 > Regards, > Dominique    -- o   Have VMS, Will Traveln Wire paladin, San Franciscod   (paladinATalphaseDOTcom)   ------------------------------   End of INFO-VAX 2003.527 ************************