1 INFO-VAX	Sat, 23 Dec 2006	Volume 2006 : Issue 704       Contents:" Re: Audio Cast #2 is now available Grow your christmas tree !. Re: Hobbyist CD-ROM (data) for VMS V8.3 Alpha?, Re: Hobbyist PAK (whining, minor) complaints, Re: Hobbyist PAK (whining, minor) complaints0 Re: Multiple time zones from a single VMS server0 Re: Multiple time zones from a single VMS server0 Re: Multiple time zones from a single VMS server0 Re: Multiple time zones from a single VMS server- Re: Need help with DHCP Client & Name servers - Re: Need help with DHCP Client & Name servers - Re: Need help with DHCP Client & Name servers - Re: Need help with DHCP Client & Name servers - Re: Need help with DHCP Client & Name servers , Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?- Re: VMS Alpha V8.3 v. USB flash storage gizmo   F ----------------------------------------------------------------------    Date: 22 Dec 2006 14:20:08 -0800( From: "Ian Miller" <gxys@googlemail.com>+ Subject: Re: Audio Cast #2 is now available B Message-ID: <1166826008.055358.42110@i12g2000cwa.googlegroups.com>   Show notes are at 9 http://www.openvms.org/stories.php?story=06/12/22/6415112    Feedback form is at  http://www.openvms.org/feedback    ------------------------------  % Date: Fri, 22 Dec 2006 22:08:44 -0500 . From: JF Mezei <jfmezei.spamnot@vaxination.ca># Subject: Grow your christmas tree ! 8 Message-ID: <e8fcd$458c9de3$cef8887a$23249@TEKSAVVY.COM>  L In this festive season, we all know that the focus switches from the length H of a certain organ to the size of one's chritsmas tree  which is put on ( display for everyone to see and measure.  H Neighbours try to figure out who'se got the biggest christmas tree, but  nobody will admit to it.    J During the holiday season, our Men's Health Pharmacy has converted itself I into a Tree Growth Pharmacy and is remarketing our world famous pills to  > help men grow their christmas tree during this festive season.  G You're one of the few lucky  to have been selected, out of millions of  L candidates, to receive this special introductory offer for our world famous H tree growing pills which are garanteed to add 3" to your christmas tree.  I Everyone knowns that the ornaments that hang from bigger christmas trees  D are more appreciated and make the neighbours jealous.  And a bigger / christmas tree supports bigger christmas balls.   & For a very limited time, if you go to L http://www.itch-pee.com/go/alphaserver and order any alpha server machines, K you will receive a bottle of our world famous tree growing pills that will  , make you the envy of all of your neighbours.  ; But hurry up ! this offer is valid for a limited time only.    ------------------------------  + Date: Fri, 22 Dec 2006 19:52:01 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda)7 Subject: Re: Hobbyist CD-ROM (data) for VMS V8.3 Alpha? 2 Message-ID: <06122219520183_2020028F@antinode.org>  H    My thanks to those folks who made various offers.  I sucked down whatH I needed from the semi-secret HP FTP server.  Converting a BACKUP /IMAGEC save set to a CD image using LD is less fun than getting a CD image   directly, but not a big problem.  0    If I need anything else, I'll be sure to ask.  H ------------------------------------------------------------------------  3    Steven M. Schweda               sms@antinode-org 4    382 South Warwick Street        (+1) 651-699-9818    Saint Paul  MN  55105-2547    ------------------------------  + Date: Fri, 22 Dec 2006 17:05:29 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda)5 Subject: Re: Hobbyist PAK (whining, minor) complaints 2 Message-ID: <06122217052966_2020028F@antinode.org>   From: davidc@montagar.com   @ > > 1.  E-mail response is slower than might be expected.  [...]  F > Not sure what I can do about that.  I'm going to retire "the birds",I > since they are being replaced by DS10L's.  That might help, but some of H > the issue could be due to spam loading...  I had 27,000 spams rejected" > by spamhaus.org yesterday alone.  G    Since the major annoyance source stopped, my SMTP connection rate is H down to about 1000/12h, so your load seems to be higher than mine.  (Buy more hardware, I always say.)   @ > >    3.  I miss the /HARDWARE_ID=serial_number qualifier [...]  F > Is that missing?  It should be in the Base O/S PAK's but the LayeredD > Products can be loaded on any system/cluster for all participating  > nodes and have no hardware id.  " $ LICENSE REGISTER OPENVMS-ALPHA - /ISSUER=OPENVMS_HOBBYIST -( /AUTHORIZATION=DECUS-USA-328018-991781 - /PRODUCER=DEC - 
 /UNITS=0 -  /TERMINATION_DATE=20-DEC-2007  -
 /ACTIVITY=A -  /OPTIONS=(NO_SHARE) -  /CHECKSUM=2-xxxx-xxxx-xxxx-xxxx     F > As David D. pointed out, a wget command could do this, or lynx.  YouH > can use "GET" or "POST" method, but you could script out the post dataF > into a series of files, and do "lynx --post-data" in a command file.  F    I'd probably select wget, but the problem is more remembering to doF it than it is visiting the Web page.  I'm pretty sure that someone hasG posted an expiring-soon-license-checking procedure, and I assume that I + could create one, if only I were less lazy.   1 > So far, I haven't gotten any IA64 complaints...       Well, if _I_ had one, ...  @    Thanks for the response (and the hobbyist effort in general).  H ------------------------------------------------------------------------  3    Steven M. Schweda               sms@antinode-org 4    382 South Warwick Street        (+1) 651-699-9818    Saint Paul  MN  55105-2547    ------------------------------  % Date: Fri, 22 Dec 2006 17:38:39 -0800 % From: DeanW <dean.woodward@gmail.com> 5 Subject: Re: Hobbyist PAK (whining, minor) complaints F Message-ID: <3f119ada0612221738j3a8029k6bd7654ffd3ec81@mail.gmail.com>  8 On 12/22/06, Steven M. Schweda <sms@antinode.org> wrote:B > > >    3.  I miss the /HARDWARE_ID=serial_number qualifier [...] > H > > Is that missing?  It should be in the Base O/S PAK's but the LayeredF > > Products can be loaded on any system/cluster for all participating" > > nodes and have no hardware id.  I I just recently got base PAKs for Alpha & IA64- no hardware ID on either.   H > > As David D. pointed out, a wget command could do this, or lynx.  YouJ > > can use "GET" or "POST" method, but you could script out the post dataH > > into a series of files, and do "lynx --post-data" in a command file. > H >    I'd probably select wget, but the problem is more remembering to doH > it than it is visiting the Web page.  I'm pretty sure that someone hasI > posted an expiring-soon-license-checking procedure, and I assume that I - > could create one, if only I were less lazy.   E Hm- something to do it automagically would be cool- but an extra week D or two before expiration would be most appreciated. Of course, givenF that, some of us would then find our dates marching forward... for the= same reason.- forgetting to renew them before they expire. ;)   B >    Thanks for the response (and the hobbyist effort in general).   Wot 'e said!   ------------------------------    Date: 22 Dec 2006 13:47:24 -0800' From: "mooneydl" <mooneydl@hotmail.com> 9 Subject: Re: Multiple time zones from a single VMS server A Message-ID: <1166824044.897545.43430@73g2000cwn.googlegroups.com>    Bob Koehler wrote:m > In article <1166545625.506816.94300@73g2000cwn.googlegroups.com>, "mooneydl" <mooneydl@hotmail.com> writes: J > > We have a VMS server located in the U.S. Eastern Time zone (GMT-05:00)J > > hosting an application used in the U.S., Switzerland (GMT+01:00),  andJ > > Japan (GMT+09:00). Some dates and times need to be shown in the user'sJ > > local time zone. What libraries can be used to take a date and time asF > > expressed in one time zone and express it in a different time zoneF > > while accounting for the different starting and ending periods for3 > > daylight saving time in the various time zones?  > G >    The is built into the C run time library.  See the C RTL utilities  >    manual for starters.   G I would like to thank everyone for their suggestions! Using the C RTL I C now have a solution that meets our needs. Sometimes we just want to D show the current date and time to the user in their local time zone.> Sometimes we want to convert a stored date and time (stored asG US/EASTERN) to the user's local time zone and then show them the result  of the conversion.  C I'm finding that switching between time zones results in a bit of a F performance penalty. I've set the DECC$TZ_CACHE_SIZE to various valuesG (2,6,10,100) but I don't see any improvement in performance. If you set @ the time zone and leave it there then performance is great. I'veD include a sample program below that demonstrates the performance. IfC the time zone is set and left alone then the program completes in a C second. If the time zone is switched for each iteration of the loop F then the program completes in about 15 seconds. Is there anything else, that can be done to improve the performance?   Thanks,  Dan   6 Running on an es45, vms 7.3-2, Compaq C V6.5-001-48BCD   (LNM$PROCESS_TABLE)      "DECC$TZ_CACHE_SIZE" = "6"   CC TESTTZ.C  LINK TESTTZ 
 RUN TESTTZ   #include <stdio.h> #include <string.h>  #include <time.h> @ int setenv (const char *name, const char *value, int overwrite);( void *calloc(size_t nmemb, size_t size); void free(void *ptr); ! int main (int argc, char *argv[])  { 	 time_t t;  typedef struct tm tm_t; 
 tm_t *tml;
 tm_t *tmr; int i;   setenv ("TZ", "US/EASTERN", 1);  tzset(); t = time((time_t)0);  tml = calloc( sizeof(tm_t), 1 ); tmr = localtime_r(&t,tml);, printf("%d-%02d-%02d-%02d:%02d:%02d.%03d\n",7         tml->tm_year+1900, tml->tm_mon+1, tml->tm_mday, /         tml->tm_hour, tml->tm_min, tml->tm_sec, 
         (0)); 
 free(tml);   for (i = 0; i < 1001; i++) { !   setenv ("TZ", "US/EASTERN", 1); 
   tzset();   t = time((time_t)0);"   tml = calloc( sizeof(tm_t), 1 );   tmr = localtime_r(&t,tml); /*.   printf("%d-%02d-%02d-%02d:%02d:%02d.%03d\n",9           tml->tm_year+1900, tml->tm_mon+1, tml->tm_mday, 1           tml->tm_hour, tml->tm_min, tml->tm_sec,            (0));  */  C   setenv ("TZ", "JAPAN", 1);              // Switching time zones -  slower performanceG //  setenv ("TZ", "US/EASTERN", 1); // Not switching time zones - great  performance 
   tzset();   tmr = localtime_r(&t,tml); /*.   printf("%d-%02d-%02d-%02d:%02d:%02d.%03d\n",9           tml->tm_year+1900, tml->tm_mon+1, tml->tm_mday, 1           tml->tm_hour, tml->tm_min, tml->tm_sec,            (0));  */     free(tml); }    setenv ("TZ", "JAPAN", 1); tzset(); t = time((time_t)0);  tml = calloc( sizeof(tm_t), 1 ); tmr = localtime_r(&t,tml);, printf("%d-%02d-%02d-%02d:%02d:%02d.%03d\n",7         tml->tm_year+1900, tml->tm_mon+1, tml->tm_mday, /         tml->tm_hour, tml->tm_min, tml->tm_sec, 
         (0)); 
 free(tml);   }    ------------------------------    Date: 22 Dec 2006 14:02:30 -0800- From: "Doug Phillips" <dphill46@netscape.net> 9 Subject: Re: Multiple time zones from a single VMS server B Message-ID: <1166824950.625989.261390@42g2000cwt.googlegroups.com>   David J Dachtera wrote:  > Doug Phillips wrote:
 > > [snip]G > > So, until each station or user gets a GPS chip, there is no perfect H > > solution. And, I can think of a few folks here who wouldn't go along > > with a GPS solution. > E > Seeking "perfect solutions" results in "the paralysis of analysis".  >   C A "perfect" solution is not necessarily the "only" solution. In the G above, I meant "a simple solution that completely solves the problem of G identifying users locations." AFA analysis paralysis: Like Deep Thought D said when his Ultimate Answer was mocked (paraphrased), "I think theD problem . . . is that [we've] never actually known what the question is."  E Sometimes seeking even a "best" solution isn't cost efficient because @ we don't have time to analyze every option and the problem isn'tD critical or isn't stable, and we know we can revisit it again, so we> just "settle" on something that works; like, sitting down in a@ restaurant and being handed a menu and wine list the size of theD OpenVMS V7 Debugger Manual, and not really being hungry for just oneF thing; well, that's where analysis paralysis can ruin a nice evening:)  L >  Perhaps a key reason why we do not have OpenVMS-IA32 today, nor are there > plans for  OpenVMS-x64.   8 I think you and I have similar feelings on this subject.  M > Some solutions work in the situations for which they were designed, but are B > totally incompatible with other situations. That's their nature. >   D A "perfect" solution for one problem isn't a "universal" solution toD all problems. Except maybe, annihilation of the universe might solveG everything, except, maybe everything goes someplace else when it ceases F to exist here, so, it seems we don't yet have enough data to formulateG a universal solution to all problems. Hmm. Could our Big Bang have been D the manifestation of a universal solution to all problems in anotherF universe? Maybe The Supreme Being will show up and explain everything.  P > Classic example: bound volume sets had their use, but now seem archaic to many( > while some folks still depend on them. >   G A bound volume set could be the perfect solution to a specific problem, A at a specific time and working within specific parameters. If the E problem changes, or certain constraints are removed (i.e. the cost of B storage, affordable RAID,...) then a new perfect solution might be found.  G Thanks for the thought-stream. If you think about it enough, everything 
 is weird:)   ------------------------------  % Date: Fri, 22 Dec 2006 17:18:09 -0500 . From: JF Mezei <jfmezei.spamnot@vaxination.ca>9 Subject: Re: Multiple time zones from a single VMS server 8 Message-ID: <24895$458c59c8$cef8887a$13206@TEKSAVVY.COM>   Larry Kilgallen wrote:J > I see no reason to audit something that is merely a user display assist.= > File metadata should be in a constant timezone (I use GMT).    <devil's advocate>  L If you must enter a contest before midnight LOCAL TIME,  and you're allowed M to change your own time zone, you can then cheat. Auditing is not irrelevant.    </devil's advocate>    ------------------------------  # Date: Sat, 23 Dec 2006 06:33:30 GMT   From: John Santos <john@egh.com>9 Subject: Re: Multiple time zones from a single VMS server % Message-ID: <_44jh.98$6_.10@trnddc07>    JF Mezei wrote:  > Larry Kilgallen wrote: > K >> I see no reason to audit something that is merely a user display assist. > >> File metadata should be in a constant timezone (I use GMT). >  >  > <devil's advocate> > F > If you must enter a contest before midnight LOCAL TIME,  and you're H > allowed to change your own time zone, you can then cheat. Auditing is  > not irrelevant.  >  > </devil's advocate>   H All the contests I've seen (computer-based ones, not ones where you mailG in the entries, and they go by the postmark) the deadlines are based on E the timezone of the contest sponser, not the timezone of the entrant.   < (All entries must be received by midnight Dec 24, 2006 EST.)       --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------    Date: 22 Dec 2006 16:51:16 -0800 From: Ken.and.Ann@gmail.com 6 Subject: Re: Need help with DHCP Client & Name serversB Message-ID: <1166835076.896846.262410@48g2000cwx.googlegroups.com>   John Santos wrote: > JF Mezei wrote:    [snip]  < > The Linksys router defaults to 192.168.1.1 (IIRC, or maybe   That's right, 192.168.1.1 ...   D > 192.168.0.1).  If you hard-code some other address into your AlphaF > in that subnet, the Linksys should route for it, and the DHCP serverJ > in the Linksys should notice that something on the LAN is using whateverF > address you chose for your Alpha, and not try to serve it to another	 > client.   E OK, that's interesting.  I didn't try that because I didn't know what  the Linksys C might do if the port was set up for DHCP.  OTOH, if there's no DHCP  request  the Linksys shouldn't care???    [...] J > The Linksys uses 255.255.255.0 (/24).  I think it's 192.168.1.1, but I'm > not certain.   Yes, this is what I found.   [...] C > Before zapping all the config files, and re-running TCPIP$CONFIG, E > did you already run it the first time?  Lots of TCPIP configuration D > problems seem to be from people installing it and then diving into? > the various configuration stuff without first running through B > TCPIP$CONFIG, which seems to set up some things that there is no > other way of initializing.  ? TCPIP$CONFIG is/was run the first time I did TCPIP$STARTUP, and E configured DHCP on the WE0 interface.  But it doesn't configure BIND.   ` >                                                          In particular, I think it creates theG > config files JF mentions above.  It also creates various VMS accounts G > used by some (but not all) of the TCPIP clients and servers.  I don't F > think the DNS client (resolver) needs a special VMS account, but the > DNS server definitely does.   F I've done a little debugging and found that even though the client.pcyB file does a "request DNS_domain" (I think), no domain is returned,) just the IP addresses of the DNS servers.      [...] J > > You then define the bind resolver. You can then enter the two IPs fromI > > your ISP as bind servers. It may prompt you for host names (which get B > > defined in the hosts database to point to those IP addresses). > >  > C > Don't remember if it asks for host names.  It might, but they are B > definitely not needed for name resolution.  In fact, it *has* toB > use the IP addresses, because if it only knew the host names, it6 > couldn't resolve them to locate the name servers :-)  E I think as JF mentioned elsewhere, I need to configure a local domain 8 to make BIND happy, at least, that's what I'll try next.   [...] A > My Linksys doesn't talk telnet, just HTTP as far as I can tell.   @ Precisely, which is why I'm in this "fix".  I can telnet/port=80 192.168.1.1,= but I can't do the required authentication with no browser...   E Thanks to John, JF, and Steven M. Schweda (who replied privately) for D all your suggestions.  I'll do some more experimentation and let you all C know the results.  If worse comes to worst. my duaghter will arrive  with herA iBook on Sunday and I'll be able to browse to the router from her  system.    Thanks again, Ken  --  
 Ken Fairfield    ------------------------------  % Date: Fri, 22 Dec 2006 20:15:18 -0700 $ From: Jeff Campbell <n8wxs@arrl.net>6 Subject: Re: Need help with DHCP Client & Name servers3 Message-ID: <1166843802_16977@sp6iad.superfeed.net>    Ken.and.Ann@gmail.com wrote: > John Santos wrote: >> JF Mezei wrote: >  > [snip] > = >> The Linksys router defaults to 192.168.1.1 (IIRC, or maybe  >  > That's right, 192.168.1.1 ...  > E >> 192.168.0.1).  If you hard-code some other address into your Alpha G >> in that subnet, the Linksys should route for it, and the DHCP server K >> in the Linksys should notice that something on the LAN is using whatever G >> address you chose for your Alpha, and not try to serve it to another 
 >> client. > G > OK, that's interesting.  I didn't try that because I didn't know what 
 > the Linksys E > might do if the port was set up for DHCP.  OTOH, if there's no DHCP 	 > request  > the Linksys shouldn't care???    Correct.  D According to the user guide on the Linksys web site, the DHCP serverH uses 192.168.1.100 to 192.168.1.149 as the default address pool for DHCPC clients (assuming the unit is still on the default IP base address, F 192.168.1.1.) The router will route the whole subnet minus the gateway and broadcast addresses.   >  > [...] K >> The Linksys uses 255.255.255.0 (/24).  I think it's 192.168.1.1, but I'm  >> not certain.  >  > Yes, this is what I found. >  > [...] D >> Before zapping all the config files, and re-running TCPIP$CONFIG,F >> did you already run it the first time?  Lots of TCPIP configurationE >> problems seem to be from people installing it and then diving into @ >> the various configuration stuff without first running throughC >> TCPIP$CONFIG, which seems to set up some things that there is no  >> other way of initializing.  > A > TCPIP$CONFIG is/was run the first time I did TCPIP$STARTUP, and G > configured DHCP on the WE0 interface.  But it doesn't configure BIND.  \ E Well, the core configuration option will configure the BIND resolver. E The BIND server doesn't need to be configured unless you want/need to  do so.  B If you didn't do all 5 of the core configuration items, you shouldD do so when you reuse TCPIP$CONFIG to change the interface from using DHCP to using a static IP.   > a >>                                                          In particular, I think it creates the H >> config files JF mentions above.  It also creates various VMS accountsH >> used by some (but not all) of the TCPIP clients and servers.  I don'tG >> think the DNS client (resolver) needs a special VMS account, but the  >> DNS server definitely does. > H > I've done a little debugging and found that even though the client.pcyD > file does a "request DNS_domain" (I think), no domain is returned,+ > just the IP addresses of the DNS servers.   E The Linksys may or may not have the capability to supply the info VMS C is asking for. I know my DSL router, a Zoom ADSL X6, can supply the B domain name, router (gateway) addresses, name server addresses andD host name. But VMS, being standards compliant 8-) , doesn't like the0 way the Zoom returns the data and so ignores it.   >  >  > [...] J >>> You then define the bind resolver. You can then enter the two IPs fromI >>> your ISP as bind servers. It may prompt you for host names (which get B >>> defined in the hosts database to point to those IP addresses). >>> D >> Don't remember if it asks for host names.  It might, but they areC >> definitely not needed for name resolution.  In fact, it *has* to C >> use the IP addresses, because if it only knew the host names, it 7 >> couldn't resolve them to locate the name servers :-)  > G > I think as JF mentioned elsewhere, I need to configure a local domain : > to make BIND happy, at least, that's what I'll try next.  D The BIND server wants a domain name, but the resolver won't care. ItC will use the name - address mappings in the local hosts table. When = you reconfigure the BIND resolver using the TCPIP$CONFIG core F configuration option, use the IP addresses of the two ISP name serversC you are using in answer to the name server prompt. The routine will F ask if you want to assign a name to each name server address. Here youE assign a name such as NS1, ns2.myisp.com, etc. These will be added to  the local hosts table.   >  > [...] B >> My Linksys doesn't talk telnet, just HTTP as far as I can tell. > B > Precisely, which is why I'm in this "fix".  I can telnet/port=80 > 192.168.1.1,? > but I can't do the required authentication with no browser...  > G > Thanks to John, JF, and Steven M. Schweda (who replied privately) for F > all your suggestions.  I'll do some more experimentation and let you > all E > know the results.  If worse comes to worst. my duaghter will arrive 
 > with herC > iBook on Sunday and I'll be able to browse to the router from her 	 > system.  >  > Thanks again, Ken    Jeff  Q ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------  % Date: Fri, 22 Dec 2006 22:26:39 -0500 . From: JF Mezei <jfmezei.spamnot@vaxination.ca>6 Subject: Re: Need help with DHCP Client & Name servers8 Message-ID: <50121$458ca217$cef8887a$24241@TEKSAVVY.COM>   Jeff Campbell wrote:C > The BIND server wants a domain name, but the resolver won't care.   J Actually it cares.  Not 100% sure it requires one, but it wants one. This K is because it has the built-in ability to tag a domain name to a host name     when resolving dns queries.   J So, when I ask to TELNET ROUTER1 , it is the bind *resolver* that tags my @ domain name and then makes the DNS request to get an IP address.   ------------------------------  % Date: Fri, 22 Dec 2006 21:40:28 -0700 $ From: Jeff Campbell <n8wxs@arrl.net>6 Subject: Re: Need help with DHCP Client & Name servers3 Message-ID: <1166848911_17027@sp6iad.superfeed.net>    JF Mezei wrote:  > Jeff Campbell wrote:D >> The BIND server wants a domain name, but the resolver won't care. > G > Actually it cares.  Not 100% sure it requires one, but it wants one.  H > This is because it has the built-in ability to tag a domain name to a ( > host name  when resolving dns queries. > I > So, when I ask to TELNET ROUTER1 , it is the bind *resolver* that tags  E > my domain name and then makes the DNS request to get an IP address.   D It doesn't require one. Localhost is not a member of any domain, forG example. When you TELNET ROUTER1, the resolver looks in the local hosts A file TCPIP$HOSTS.DAT (perhaps cached?) first. If the name doesn't C resolve, then it will ask the name servers it knows for the address G translation. Why would the resolver append the local domain name to the B requested name if the requested name is not *in* the local domain?) Shirley, it would know it's own zone? 8-)    Jeff  Q ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------  # Date: Sat, 23 Dec 2006 06:14:59 GMT   From: John Santos <john@egh.com>6 Subject: Re: Need help with DHCP Client & Name servers% Message-ID: <DP3jh.96$6_.88@trnddc07>    Ken.and.Ann@gmail.com wrote: > John Santos wrote: >  >>JF Mezei wrote:  >  >  > [snip] >  > < >>The Linksys router defaults to 192.168.1.1 (IIRC, or maybe >  >  > That's right, 192.168.1.1 ...  >  > D >>192.168.0.1).  If you hard-code some other address into your AlphaF >>in that subnet, the Linksys should route for it, and the DHCP serverJ >>in the Linksys should notice that something on the LAN is using whateverF >>address you chose for your Alpha, and not try to serve it to another	 >>client.  >  > G > OK, that's interesting.  I didn't try that because I didn't know what 
 > the Linksys E > might do if the port was set up for DHCP.  OTOH, if there's no DHCP 	 > request  > the Linksys shouldn't care???  >   B Yes.  I think it is because the Linksys doesn't have any permanent< memory, so if you have it configured to DHCP serve addresses= 192.168.1.2 through 192.168.1.100 and it sees something using B 192.168.1.50, it just assumes it must have issued that address and> then been rebooted, so it marks the address as "leased" in its? list.  At least, this theory is consistent with what I've seen.      > [...]  > J >>The Linksys uses 255.255.255.0 (/24).  I think it's 192.168.1.1, but I'm >>not certain. >  >  > Yes, this is what I found. >  > [...]  > C >>Before zapping all the config files, and re-running TCPIP$CONFIG, E >>did you already run it the first time?  Lots of TCPIP configuration D >>problems seem to be from people installing it and then diving into? >>the various configuration stuff without first running through B >>TCPIP$CONFIG, which seems to set up some things that there is no >>other way of initializing. >  > A > TCPIP$CONFIG is/was run the first time I did TCPIP$STARTUP, and G > configured DHCP on the WE0 interface.  But it doesn't configure BIND.  >   C Option 3 (server components), Item 1 (BIND), at least on my version C of TCP/IP Services 5.6 on I64, VMS 8.3.  But it's been this way for @ a long time.  This option doesn't create and populate the domain> database files or the TCPIP$BIND.CONF file, you need to createB those by hand, from the templates, but it does a bunch of magic toD allow the service to be enabled, and may, the first time it is done,B create the TCPIP$BIND VMS account and the sys$sysroot:[tcpip$bind]
 directory.   > ` >>                                                         In particular, I think it creates theG >>config files JF mentions above.  It also creates various VMS accounts G >>used by some (but not all) of the TCPIP clients and servers.  I don't F >>think the DNS client (resolver) needs a special VMS account, but the >>DNS server definitely does.  >  > H > I've done a little debugging and found that even though the client.pcyD > file does a "request DNS_domain" (I think), no domain is returned,+ > just the IP addresses of the DNS servers.  >  >   F Sounds like a DHCP server problem.  Maybe the Linksys doesn't know its9 own domain name yet, so it can't serve it to its clients.      > [...]  > I >>>You then define the bind resolver. You can then enter the two IPs from H >>>your ISP as bind servers. It may prompt you for host names (which getA >>>defined in the hosts database to point to those IP addresses).  >>>  >>C >>Don't remember if it asks for host names.  It might, but they are B >>definitely not needed for name resolution.  In fact, it *has* toB >>use the IP addresses, because if it only knew the host names, it6 >>couldn't resolve them to locate the name servers :-) >  > G > I think as JF mentioned elsewhere, I need to configure a local domain : > to make BIND happy, at least, that's what I'll try next. >  > [...]  > A >>My Linksys doesn't talk telnet, just HTTP as far as I can tell.  >  > B > Precisely, which is why I'm in this "fix".  I can telnet/port=80 > 192.168.1.1,? > but I can't do the required authentication with no browser...  > G > Thanks to John, JF, and Steven M. Schweda (who replied privately) for F > all your suggestions.  I'll do some more experimentation and let you > all E > know the results.  If worse comes to worst. my duaghter will arrive 
 > with herC > iBook on Sunday and I'll be able to browse to the router from her 	 > system.  >  > Thanks again, Ken   
 Good luck!   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  # Date: Fri, 22 Dec 2006 19:41:33 GMT ( From: Alan Greig <greigaln@netscape.net>5 Subject: Re: Open sourced VMS as a business concept ? > Message-ID: <NxWih.135706$bz5.75022@fe3.news.blueyonder.co.uk>   Michael Austin wrote:   G > On another similar topic, configuring a Sun active-active Oracle RAC  J > cluster takes 15-20 times longer on a two node cluster than configuring * > a 4- to 10-node VMS Cluster running RDB.  1 Which is a big plus if you're paid by the hour...    --  
 Alan Greig   ------------------------------  % Date: Fri, 22 Dec 2006 16:59:33 -0600 2 From: pechter@pechter.dyndns.org (William Pechter)5 Subject: Re: Open sourced VMS as a business concept ? : Message-ID: <4L-dnbLNS_vI_hHYnZ2dnUVZ_qjinZ2d@comcast.com>  3 In article <EWXV2WFvmzqu@eisner.encompasserve.org>, : Bob Koehler <koehler@eisner.aspm.encompasserve.org> wrote:D >In article <zpCih.20012$HV6.6909@newsfe1-gui.ntli.net>, ChrisQuayle ><nospam@devnul.co.uk> writes: > 6 >> Come to think of it, Dec were orignally a hardware ! >> company, so what went wrong ?.  > E >   DEC separated the OS license from the hardware so that you didn't @ >   have to buy both.  So many people bought VAXen without a VMSH >   license that DEC looked into it and found they were all running BSD,( >   which lead them to market ULTRIX-32. > C >   And back in PDP-11 days, you had your choice of several OS with G >   different capabilities.  Its easier to sell you the one you want if G >   you list them as different parts, and it makes sense to charge more  >   for the more complex OS. > C >   Even VAXen supported three DEC OS and a couple of non-DEC UNIX.   % They pretty much had two I thought...                VAX/VMS  			Ultrix-32  G VAXELN was pretty much for target embedded apps... Don't think 11/7x0's  ran it.      > F >   Sun didn't have such capabilities to offer, all they could do was D >   UNIX.  No choices, so it might as well ship with the hardware.  D >   Since they started out selling a BSD, there was no market users 5 >   to buy hardware from Sun and the OS from Berkley.  >   = Well, they did change that with the AT&T deal and the move to  Solaris2.x.   G Also, NextStep was available for a while... and Solaris was (along with . WinNT) developed for PowerPC for a short time.   So the OS could be decoupled.      Bill   --   --  <   "When I think back on all the crap I learned in Vax schoolG   It's a wonder I fixed anything at all."   (to the tune of Kodachrome)    pechter-at-ureach.com    ------------------------------   Date: 23 Dec 2006 01:23:36 GMT From: healyzh@aracnet.com 5 Subject: Re: Open sourced VMS as a business concept ? , Message-ID: <emi0eo017v0@enews1.newsguy.com>  3 William Pechter <pechter@pechter.dyndns.org> wrote: I > Also, NextStep was available for a while... and Solaris was (along with 0 > WinNT) developed for PowerPC for a short time.   > So the OS could be decoupled.   I Wasn't that limited to the Sun Sparc 5, and not even all models of it?  I K believe the number of S-Bus widgets supported was extremely small as well.  6 BTW, NeXTStep also supported at least one model of HP.   	Zane    ------------------------------  + Date: Fri, 22 Dec 2006 17:19:14 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda)6 Subject: Re: VMS Alpha V8.3 v. USB flash storage gizmo2 Message-ID: <06122217191421_2020028F@antinode.org>  ' From: David Mathog <mathog@caltech.edu>   F > Did you verify that the USB card works by plugging it into a PC and H > testing it there?  I've used several similar NEC cards and while they F > typically "just work" sometimes you'll hit a bad one.  In this case,( > on VMS, that's one too many variables.  
 > [...]  When J > (while the MOUNT was hung) I pulled the gizmo so I could try it on a MacB > (where it seemed to work just fine), the system crashed, with an8 > INVEXCEPTN ("exception while above ASTDEL") complaint.  1    I thought that the Mac test would be adequate.   F    Interestingly (or not), this is the same gizmo I used while workingG on mtools for VMS, so I did have more (though not unlimited) success in G the past with the same hardware (but older/different software).  In any F case, the informal (not-really-) support process has begun, so there's, some hope for the future.  Watch this space.      SMS.    ------------------------------   End of INFO-VAX 2006.704 ************************