1 INFO-VAX	Sun, 07 May 2006	Volume 2006 : Issue 253       Contents:8 Re: DCL versus Unix CLIs, was: Re: File output like Unix8 Re: DCL versus Unix CLIs, was: Re: File output like Unix8 Re: DCL versus Unix CLIs, was: Re: File output like Unix# Re: DHCP client - Help configuring. # Re: DHCP client - Help configuring. # Re: DHCP client - Help configuring.  FIBRE-SCSI V8.0 patch P Re: H P To Launch Multi Million Dollar Ad Campaign For The PC  [WAS   Re: OT: InP Re: H P To Launch Multi Million Dollar Ad Campaign For The PC  [WAS   Re: OT: InL Re: This post will self-destruct in 10secs (Was Re: X windows vulnerability) Re: X windows vulnerability  Re: X windows vulnerability  Re: X windows vulnerability   F ----------------------------------------------------------------------   Date: 7 May 2006 08:38:04 -0700 $ From: "AEF" <spamsink2001@yahoo.com>A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix B Message-ID: <1147016284.107652.80100@g10g2000cwb.googlegroups.com>   JF Mezei wrote:  > AEF wrote:I > > If a file ends in .DIR;1 then it means (implied or otherwise) that it D > > will be looked for by an appropriate directory lookup operation. >   > The wording should instead be: > G > If a file ends in .DIR it does not imply that it is a directory file. A > If you are trying to access a directory, then VMS will seek the ' > directory name with file type ".DIR".    That's what I said.    > I > However, if that file does not have the "directory" attribute, then the   > directory operation will fail.   That's what I said.    >  > create /dir [.cake]  > create [.cake]chocolate.txt  > miumm !!! 
 > <ctrl-z> >   > set file cake.dir /nodirectory > B > dir [.cake]  results in an error because cake.dir is not a valid > directory file.  > I > The "classic" MACoS file system was truly hiearchical and there were no I > real directory "files" that you could play with.  They were a node in a I > database. This makes for an extremely fast file system when looking for J > files. But it also means that corruption in that oen database file meansH > your disk is lost and you need big tools to recover as much as you canF > from the disk. In VMS, it is easier to recover from errors since theE > file system is spread amongst many files. And having separate files I > probably also makes directory ownership/ACLs etc easier to manage since I > used don't need to update some root directory database when they create H > their own files in their own directory. (but they still need to updateG > indexf.sys, so perhaps there is not much of an advantage in the end).  >  >  > E > VMS doesn't have that level of sophistication and getting a list of C > directories involves a DIR *.DIR (and that will erroneously catch I > non-directory files that happen to have .DIR file type.  You can easily I > test this with FILEVIEW in decwidnows that displays any file in .DIR as  > a directory. > I > However, at the DCL level, you can type, edit, print etc any file, even J > if it ends in .DIR (as long as it doesn't have the directory attribute).   ------------------------------   Date: 7 May 2006 09:07:34 -0700 $ From: "AEF" <spamsink2001@yahoo.com>A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix C Message-ID: <1147018054.539381.217330@g10g2000cwb.googlegroups.com>    david20@alpha2.mdx.ac.uk wrote: l > In article <1146934213.961533.202380@i40g2000cwc.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes: > > " > >davi...@alpha2.mdx.ac.uk wrote:o > >> In article <1146872269.637151.135640@g10g2000cwb.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes:  > >> >% > >> >david20@alpha2.mdx.ac.uk wrote: r > >> >> In article <1146616707.415114.288360@v46g2000cwv.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes:( > >> >> >david20@alpha2.mdx.ac.uk wrote: > >> >> >> In article <445656D0.A6669895@NeOaSrPtAhMlNiOnWk.net>, "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes:+ > >> >> >> >david20@alpha2.mdx.ac.uk wrote: 
 > >> >> >> >> e > >> >> >> >> In article <4452d96f@usenet01.boi.hp.com>, "FredK" <fred.nospam@nospam.dec.com> writes:  > >> >> >> >> > A > >> >> >> >> >"Rob Brown" <mylastname@gmcl.com> wrote in message N > >> >> >> >> >news:Pine.LNX.4.61.0604281211340.19111@localhost.localdomain... > >> >> >> >> >  > >  > >[......] (big snip) > >  > >> >What's your point? > >> > > >>H > >> The point was in refuting the claim in the top section of this post > >> > >> "Q > >> >> >> >> >> > Why include Unix?  In Unix, no filename (or extension) has any T > >> >> >> >> >> > implicit meaning.  Applications may impart meaning, but the OS ju > >> >> >> >> >> > don't care. > >> >> >> >> >>= > >> >> >> >> >> I think that the same is true in VMS or RSX.  > >> >> >> >> >> > >> >> >> >> >  > >> >> >> >> >    .DIR  > >> >> >> >> >  > >> > > >> " > >>? > >> which suggested that .DIR  on VMS has an implicit meaning.  > >  > > - > >OK, I think we can agree on the following:  > > H > >If a file ends in .DIR;1 then it means (implied or otherwise) that itC > >will be looked for by an appropriate directory lookup operation.  > >Example: The file > >   > >    DISK$DATA1:[AAA]BBB.DIR;1 > > A > >will be the target of an operation attempting to inerptret the  > >directory-spec  > >  > >    DISK$DATA1:[AAA.BBB]  > > H > >IOW, a file ending in .DIR;1 will be expected (or assumed!) by RMS toF > >be a directory. If said file is not a bona fide directory, an error > >message will be issued: > >  > >$ DIR [AAA.BBB]E > >%DIRECT-E-OPENIN, error opening DISK$DATA1:[AAA.BBB]*.*;* as input 2 > >-RMS-E-FND, ACP file or directory lookup failed3 > >-SYSTEM-W-BADIRECTORY, bad directory file format  > >$ > > H > >VMS (or RMS) obviously considers BBB.DIR to be a corrupted directory. > > J > >Furthermore, *if* no one has done something silly like COPY NL: BBB.DIRF > >(or other silly or purposefully destructive operations and that theI > >file has not been corrupted somwhow), then .DIR;1 implies that BBB.DIR ! > >is a bona fide directory file.  > > H > >Obviously, the "*if*" part makes all the difference to you. In normal. > >VMS life, the *if* part is rarely relevant. > > D > >Now, while .DIR;1 may not strictly imply a bona fide, uncorruptedH > >directory file, it is very useful to know that in most circumstances,> > >such a file will be found to be a bona fide directory file. > >  > 
 > ON ODS-5 >  > $ create/dir [.test] > $ create/dir [test.b]  > $ create/dir [test.b.c]  > $ ren test.dir test.axp;2/lo > $ dir/full test.axp;2  > " > Directory SYS$SYSDEVICE:[000000] > 5 > TEST.AXP;2                    File ID:  (9724,40,0)  > .  > .  > .  >  > $  > $ set proc/parse=ext > $ dir [9724,40,0]  > % > Directory SYS$SYSDEVICE:[9724,40,0]  > 	 > B.DIR;1  >  > Total of 1 file. > $ dir [9724,40,0.b]  > ' > Directory SYS$SYSDEVICE:[9724,40,0.B]  > 	 > C.DIR;1  >  > Total of 1 file. > I > Shows that a directory search can start from a directory which does not E > have a .dir;1 extension - in this case one with a .axp;2 extension.   E OK, in ODS-5 (which I have no experience with) you can do some screwy ? thing to access files in a purposely mis-named directory that's F floating randomly thru a file system. Do your VMS systems have lots ofE these directories with the wrong file type? Are your users constantly F renaming directories? Do you want OpenVMS engineering to find a way to
 prevent this?    > O > Also the fact that a non-directory file with a .DIR;1 extension will cause an K > error message is no more relevent to this discussion than the fact that a P > non-executable with a .exe extension will cause an error if you attempt to run > it.  >  > $ edit test.exe  > aaaaa  > [End of file] 9 > 1 line written to file SYS$SYSDEVICE:[000000]test.exe;1  >  > $ run test.exe2 > %DCL-W-ACTIMAGE, error activating image test.exe4 > -CLI-E-IMGNAME, image file DSA0:[000000]test.exe;1: > -IMGACT-F-NOTNATIVE, image is not an OpenVMS Alpha image >   F It shows that .DIR and .EXE are special file types in VMS. No, they doG not in 100% of cases imply the files their full file-specs point to are A directories and executables, respectively, but it is still a very  important fact for VMS users.    > " > I think I have demonstrated that > > > 1) Files with an extension of .DIR;1 need not be directories   I never disputed this.  O > 2) Renaming a valid directory to an extension other than .DIR;1 does not stop M > it being a directory as such. In particular in ODS-5 it is possible to move @ > into that directory and to access files within that directory.P > (the fact that the command to move into the directory is slightly different ie" > utilising the DID is irrelevent)  E I am not familiar with ODS-5, so fine, there is a trick to read files D and file-ids from a dangling, mis-named directory. Everything else I said holds.    >  > This demonstrates that > 0 > .DIR;1 does not imply  The file is a directory >  > and  > B > The file is a directory does not imply it has a .DIR;1 extension   I never said otherwise.    >  >  > (  > " > The real implication seems to be > * > Directory bit set => file is a directory >  > and  > * > File is a directory => directory bit set > H > with the .DIR;1 requirement just added in to speed up directory searchP > operations (so every file doesn't need to be opened to check for the directory > bit) >  > )  >  > P > I fully agree that the normal state of affairs is for a directory to have both> > the directory bit set and for it to have a .DIR;1 extension.   That was my point.  H > However the thing which tells the OS that a file is a directory is the, > directory-bit not just the file extension.K > Remember this discussion is about filenames or extensions having implicit M > meanings for the OS. For a meaning to be implicit it must apply 100% of the = > time otherwise you are just talking about default meanings.   G Your just saying that just because it can be broken, it doesn't matter.   B Like I said, the "*if* part" makes all the difference to you. I amF concerned with normal situations in which case .DIR;1 and .EXE do haveE special meanings. In a file system in which you can give any file any B legal name you are not going to meet your strict criterion, but inG normal usage it holds, and therefore is a very important aspect of VMS.   D I think we'll just have to disagree to agree. (And I do mean that inD that order.) Well, we disagree on the importance of broken directoryC files. I've never encountered in 21 years as a VMS user a mis-named A directory file that I myself didn't create just to see what would G happen. Like I said last time, the "*if* part" makes all the difference  to you.   D Also, when driving a car: Stepping on the brake pedal doesn't alwaysB stop the car (it could be broken some how -- maybe someone cut theD brake lines!). Nevertheless, this is a very important feature on theG car! So I guess you'd say that stepping on the brake doesn't imply that F the car will slow down and stop. But you can't say that's not what the@ brake pedal is for! Similarly, .DIR;1 is for directory files and1 therefore has special meaning to the file system.   @ Anyway, if you want to be so strict, let's reexamine what you're	 refuting:   H > >> The point was in refuting the claim in the top section of this post > >> > >> "Q > >> >> >> >> >> > Why include Unix?  In Unix, no filename (or extension) has any T > >> >> >> >> >> > implicit meaning.  Applications may impart meaning, but the OS ju > >> >> >> >> >> > don't care. > >> >> >> >> >>= > >> >> >> >> >> I think that the same is true in VMS or RSX.  > >> >> >> >> >> > >> >> >> >> >  > >> >> >> >> >    .DIR  > >> >> >> >> >  > >> > > >> " > >>? > >> which suggested that .DIR  on VMS has an implicit meaning.   B It says "any implicit meaning". Well, "any" includes "under normalE operating circumstances" in which case you yourself admit .DIR;1 does G mean directory. Also, normal (ODS-2, at least) directory-specs will not G see directory files unless they end in .DIR;1. It's not 100%, but it is 1 ceratinly *some*, which is all the claim claimed.      > David Webb > Security team leader > CCSS > Middlesex University >  >  > >> David Webb  > >[...] > >AEF > >  AEF    ------------------------------   Date: 7 May 2006 10:22:31 -0700 $ From: "AEF" <spamsink2001@yahoo.com>A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix C Message-ID: <1147022551.096797.203100@j73g2000cwa.googlegroups.com>    Wilm Boerhout wrote:  > AEF wrote on 7-5-2006 18:07... > [snip] > >> $ run test.exe 5 > >> %DCL-W-ACTIMAGE, error activating image test.exe 7 > >> -CLI-E-IMGNAME, image file DSA0:[000000]test.exe;1 = > >> -IMGACT-F-NOTNATIVE, image is not an OpenVMS Alpha image  > >> > > J > > It shows that .DIR and .EXE are special file types in VMS. No, they doK > > not in 100% of cases imply the files their full file-specs point to are E > > directories and executables, respectively, but it is still a very ! > > important fact for VMS users.  > F > No, it does not show that .EXE is a special file type. It just showsJ > that the RUN DCL command will check any file type (including .EXE) for aE > valid image header in order to activate the image. In this case, an   < And that's part of what makes .EXE special. Try this: renameF SYSBOOT.EXE to have some other file type and reboot. Good luck. (Note:E I haven't tried this. Maybe it actually works anyway, but I doubt it. 4 If I have time later, I'll try it on a test system!)   > image header is not present. > H > As has been demonstrated before somewhere in this thread, $RUN FOO.BARH > will succeed provided an image header and a transfer address are found > by the RUN command.   F And I already acknowledged that fact and included it in my commentary.  B Nothing you have said contradicted anything I said except that youE don't think that .EXE being the default type for executables makes it A special. Don't you think it is important to know that .EXE is the @ default file type for executables? That makes it special! NB: IfC something it important to know, that makes it special! Try $ RENAME : SYS$SYSTEM:*.EXE .BAR and see what happens to your system.  G Let's try it in other words: If you have a file whose type is .EXE, you D don't have to specify the file type when you tell VMS to execute theF executable. THAT MAKES IT SPECIAL, OKAY? The fact that you can call anG executable FOO.BAR and still run it doesn't change that. OKAY? The fact E that you can rename a non-executable file to .EXE doesn't change that 
 either. OKAY?    > /Wilm    AEF    ------------------------------   Date: 6 May 2006 23:18:24 -0700  From: shofu_au@yahoo.com.au , Subject: Re: DHCP client - Help configuring.C Message-ID: <1146982704.200490.320680@e56g2000cwe.googlegroups.com>    Hi All,   + Yep I have read the above mentioned manual.   : >From the DHCP startup log TCPIP$DHCP_CLIENT_RUN.LOG I get  
 $ Set NoOn/ $ VERIFY = F$VERIFY(F$TRNLNM("SYLOGIN_VERIFY")) $ debug level                      = 0) class ID                         = TCPVMS % start_delay                      = 10 $ retries                          = 2$ lease desired                    = 0. timeouts                         = 4,8,16,32,0( use_saved_config                 = false' accept_bootp                     = true . rv                               = 1,3,6,12,151 Configuration of WE0 going to sleep for 9 seconds  Configuration of WE0 awake now DHCP started on WE0  DHCP configuring WE0#          subnet mask     :  0.0.0.0 #          IP address      :  0.0.0.0 ,          broadcast address:  255.255.255.255B initReboot for interface WE0 proceeding. No other interfaces doing DHCP.  DHCP configuring WE0#          subnet mask     :  0.0.0.0 #          IP address      :  0.0.0.0 ,          broadcast address:  255.255.255.255  * packet arrived on Saturday May 07 00:26:01   DHCP configuring WE0)          subnet mask     :  255.255.255.0 (          IP address      :  192.168.1.12*          broadcast address:  192.168.1.255'          default gateway :  192.168.1.1 > The DHCP server did not pass a host name for this host to use..   Using the requested hostname "ds10" instead.;   DNS lookups for the requested host name may not function. 9 %TCPIP-E-NAMEERROR, error processing name service request / -TCPIP-E-BINDDISABLE, BIND resolver is disabled 9 %TCPIP-E-NAMEERROR, error processing name service request 6 -TCPIP-E-INSBINDDATA, insufficient data to enable BIND    Do I need to configure BIND?????  D >From $ TCPIP SHOW ROUTE  (with static route to 192.168.1.1 defined)B %TCPIP-E-ROUTEERROR, error accessing routes database (TCPIP$ROUTE)  < Yes I can ping 192.168.1.1.  Locally I can telnet to my ds10
 192.168.1.12.    Any ideas on what I need to do?    Mark   ------------------------------  % Date: Sun, 07 May 2006 03:35:20 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> , Subject: Re: DHCP client - Help configuring., Message-ID: <445DA335.2DE64694@teksavvy.com>   shofu_au@yahoo.com.au wrote: > DHCP configuring WE0+ >          subnet mask     :  255.255.255.0 * >          IP address      :  192.168.1.12, >          broadcast address:  192.168.1.255) >          default gateway :  192.168.1.1 @ > The DHCP server did not pass a host name for this host to use.0 >   Using the requested hostname "ds10" instead.= >   DNS lookups for the requested host name may not function. ; > %TCPIP-E-NAMEERROR, error processing name service request 1 > -TCPIP-E-BINDDISABLE, BIND resolver is disabled ; > %TCPIP-E-NAMEERROR, error processing name service request 8 > -TCPIP-E-INSBINDDATA, insufficient data to enable BIND  G You don't need to consigure a bind server on VMS., but you need to tell  it about a bind resolver.   H From the above, it appears that VMS DHCP client isn't getting DNS serverG IPs. (either because the router isn't sending them, or because the DHCP 2 client doesn't know what to do with those fields).  L Do you have a disk:[TCPIP$DCHP]CLIENT.PCY file ? It shoudl contain at least: $ TYPE CLIENT.PCY  class_id TCPVMS # lease_desired 86400 # 24 hour lease  request routers  request host_name  request dns_servers  request dns_domain_name   # (this is straight from the manual).    $set def disk:[TCPIP$DHCP]I $lib/extract=client_pcy/output=client.pcy sys$library:tcpip$templates.tlb   F this will give you the above client.pcy. There are more parameters you can add in it.    / After the DHCP negotiation, you would want to :    TCPIP SHOW INTERFACE WEA0/FULL TCPIP SHOW ROUTE TCPIP SHOW NAME      You shoudl get something like: $ tcpip show name    BIND Resolver Parameters    Local domain: chocolate.com    System      State:     Started, Enabled      Transport: UDP   Domain:    chocolate.com   Retry:     4   Timeout:   4   Servers:    bike   Path:       No values defined     Process     State:     Enabled     Transport:	   Domain:    Retry:
   Timeout:
   Servers:   Path:     H The "Domain" is the domain that automatically gets added when you make a request for just a node name. H (eg: asking for "pastry" results in "pastry.chocolate.com" being sent to the DNS server.   E The "servers" portion is where the DNS servers get listed. Seems that G TCPIP servres really wants a host name in there, but DHCP might be able C to put in tthe IP address. To have a "host name" in there, you must  TCPIP SET HOST node ip.address  H This defines it ins the local hosts databnase so that the translation is= available to the stack before the interface is fully defined.     G Note, it isn't your fault you are having problems. Many people have had  problems with the DHCP client.K (in the end, you may find it MUCH easier to defined fixed IP for your lan).    ------------------------------   Date: 7 May 2006 01:41:47 -0700  From: shofu_au@yahoo.com.au , Subject: Re: DHCP client - Help configuring.C Message-ID: <1146991307.701719.110170@j33g2000cwa.googlegroups.com>    Hi JF,  3 >Note, it isn't your fault you are having problems.  Yes it just worked on 7.3.    $ TCPIP SHOW INTERFACE WE0 /full  Interface: WE0 ?    IP_Addr: 192.168.1.12      NETWRK: 255.255.255.0     BRDCST: 
 192.168.1.255 E                        Ethernet_Addr: 08-00-2B-86-31-53    MTU:  1500 *      Flags: UP BRDCST NOTRL RUN MCAST SMPX5                                   RECEIVE        SEND 5    Packets                            115         131 5      Errors                             0           0 )    Collisions:                          0        $ TCPIP SHOW name    BIND Resolver Parameters    Local domain: * Mismatch *     System      State:     Started, Disabled     Transport: UDP	   Domain:    Retry:     4   Timeout:   4&   Servers:    220.233.0.4, 220.233.0.3   Path:       No values defined     Process     State:     Disabled      Transport:	   Domain:    Retry:
   Timeout:
   Servers:   Path:     G Looks like there is an issue with the domain.  Mismatch - why???  I see G there are some other thread on this.  But behind my owner ADSL firewall   / NAT router what is the domain?   TCPIP SHOW comm    Communication Parameters  / Local host:                             Domain:   @                                  Maximum     Current        Peak( Proxies                               20     Remote Terminal    Large buffers:           0   UCBs:                    0   Virtual term:     disabled    ( $ TCPIP SHOW CONFIGURATION COMMUNICATION   Communication Configuration   9 Local host:                             Domain:   dchpool    Cluster timer:               0  9 Interfaces:                  0          Type:     Default  Device_sockets:              0 Routes:                      0 Services:                    0 Proxies:                     0  4                           Free     Maximum   Minimum4 Large buffers                0           0         04 Small buffers                0           0         0* IRPs                         0           0 Non TCPIP buffers            0   Remote Terminal    Large buffers:             0   UCBs:                      0   Virtual term:     disabled   $ type CLIENT.PCY;1  #  # File name:      CLIENT.PCY0 # Product:        HP TCP/IP Services for OpenVMS # Version:        V5.5-11  # B # ? Copyright 1976, 2004 Hewlett-Packard Development Company, L.P. #    class_id TCPVMS  request routers  request subnet_mask  request host_name  request dns_servers  request dns_domain_name    ------------------------------  * Date: Sun, 7 May 2006 07:04:11 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: FIBRE-SCSI V8.0 patch$ Message-ID: <e3k65b$4vm$1@online.de>  I What's up with the FIBRE-SCSI V8.0 patch?  It appears with a . before it  H in the FTP listing, which means it has been withdrawn, but I don't know I why.  (I would have expected to get an email, since I got one (actually,  A two---strange since otherwise I only get one) when the patch was   initially announced.   ------------------------------  % Date: Sun, 07 May 2006 09:59:38 +0200 ( From: Michael Kraemer <M.Kraemer@gsi.de>Y Subject: Re: H P To Launch Multi Million Dollar Ad Campaign For The PC  [WAS   Re: OT: In / Message-ID: <e3k9cr$921$02$1@news.t-online.com>    Bill Todd schrieb: > I > No:  throwing money at a product line which has resisted many years of  E > intense efforts to make it succeed while starving a line which has  J > managed to persist in profitability despite over a decade of neglect is K > utterly incompetent business logic.  "All show and no go" is not just an   > automotive concept, it seems.   > apparently HP thinks that after another decade the PC businessF will turn profitable. Under his assumption their decision makes sense.I HP have a couple of product lines they decided to keep (be it reasonable   or not),A and it's the least successful line that needs the most attention.    ------------------------------  % Date: Sun, 07 May 2006 04:40:05 -0400 ( From: Bill Todd <billtodd@metrocast.net>Y Subject: Re: H P To Launch Multi Million Dollar Ad Campaign For The PC  [WAS   Re: OT: In = Message-ID: <ltCdnf3QHPuiL8DZRVn-tg@metrocastcablevision.com>    Michael Kraemer wrote: > Bill Todd schrieb: >>J >> No:  throwing money at a product line which has resisted many years of F >> intense efforts to make it succeed while starving a line which has H >> managed to persist in profitability despite over a decade of neglect G >> is utterly incompetent business logic.  "All show and no go" is not  ( >> just an automotive concept, it seems. > @ > apparently HP thinks that after another decade the PC businessH > will turn profitable. Under his assumption their decision makes sense.  F If you're so willing to accept incompetent premises, perhaps you feel  you can justify anything.   B But that is utterly incompetent logic:  if the premise is absurd, ' nothing based on it can be any less so.    - bill   ------------------------------  % Date: Sun, 07 May 2006 09:56:43 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> U Subject: Re: This post will self-destruct in 10secs (Was Re: X windows vulnerability) 0 Message-ID: <eKednSQX1NQGYcDZRVn-jA@comcast.com>   Richard Maher wrote:   > Hi,  > 5 > (Listen very carefully! I shall say siss only once)  > N > It has been my painful experience that, more often than not, VMS engineeringL > simply refuses to discuss "a particular vulnerability". Heads-in-the-sand,H > platitudes about "this not being the right forum" or "details would be& > fodder for script kiddies" etc, etc. > H > But then I forgot that we're in denial - Where are my affirmations and
 > mantras? >  > a) VMS doesn't have bugs > b) VMS has never had a bug > c) VMS will never have a bug3 > d) Decimation of testing staff means no more bugs < > e) Porting of crap UNIX and C practises means no more bugsA > f) No one developing on VMS anymore is good news - no more bugs L > g) "Don't ask - Don't tell" - What you do with other bugs in your own home > is your business > h) Loose Lips Sink ShipsJ > i) Massaging pig ignorant customer's egos with NDAs really placates themL > j) If anyone questions the ethos spread FUD, attack credibility, disparageN > k) Be suspicious of anyone who purports to be developing software on VMS for > VMS  >  > Regards Richard Maher  >  <snip>  C Would you prefer that the details of each vulnerability and how to  ) exploit it were published on comp.os.vms?   : IF you can't find out, the crackers probably can't either.  D If you could find out, do you have the resources to fix the problem?  I Not being part of the "in group" may sting but limiting the distribution  @   of the details seems to have worked quite well over the years.   ------------------------------   Date: 7 May 2006 06:54:25 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen) $ Subject: Re: X windows vulnerability3 Message-ID: <SX2PfVsM4jkL@eisner.encompasserve.org>    In article <1146976841.571442.122550@i39g2000cwa.googlegroups.com>, "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> writes:  >  > JF Mezei wrote: $ >> "johnhreinhardt@yahoo.com" wrote:I >> > Note that it states that versions X11R6.8.2 and prior are apparently K >> > not vulnerable.  The X Windows used in VMS's Motif V1.5 is based on an > >> > earlier version (X11R6.6) according to the Release Notes. >>H >> If VMS engineering were still fully staffed, then they woudl have hadF >> the resources to post a notice announcing to their customers that aH >> particular vulnerability does not (or does) affect the VMS version of >> the software. >>J >> This is not the first time that public vulnerabilities of software suchI >> as X, motif, Bind etc have had no sign of life from VMS engineering to 1 >> confirm that the VMS version was NOT affected.  >>I >> It would be nice PR for HP to make public the fact that VMS version of K >> such software was not vulnerable whenever such warnings are made public.  > C > True, but the fact that VMS is secure because it's 3 dot releases A > behind the current verions might not be the best PR either.  :)   G I thought the vulnerability is to give root access, and that DECwindows  does not run as System on VMS.   ------------------------------   Date: 7 May 2006 08:38:26 -0700  From: davidc@montagar.com $ Subject: Re: X windows vulnerabilityB Message-ID: <1147016306.216221.43540@e56g2000cwe.googlegroups.com>  F That nature of this bug is a check for geteuid() that was mis-coded asD "geteuid == 0" rather than "geteuid() == 0", which is testing if the@ address of the geteuid function is 0 rather than the return fromC geteuid is 0.  Since OpenVMS does not have a "root" UID, but rather E UIC's (which tend to always be non-zero), even an OpenVMS system with D this bug would likely be unaffected.  Since OpenVMS uses a privilege@ system, and not effective UID's, to gain privileges, the code is essentially a no-op.  F So, even an OpenVMS system with this bug in the code is still probably would not be vulnerable.   ------------------------------   Date: 7 May 2006 12:24:50 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen) $ Subject: Re: X windows vulnerability3 Message-ID: <9gusGKbxSsyc@eisner.encompasserve.org>   ^ In article <1147016306.216221.43540@e56g2000cwe.googlegroups.com>, davidc@montagar.com writes:  H > That nature of this bug is a check for geteuid() that was mis-coded asF > "geteuid == 0" rather than "geteuid() == 0", which is testing if theB > address of the geteuid function is 0 rather than the return from > geteuid is 0.   A Another bug due to choice of a weakly typed programming language.    ------------------------------   End of INFO-VAX 2006.253 ************************                                                                                                                                                                                                                                                                                                                                rZٶu-QdS
 W
D+25^gOѬM|ͺWn(+{bB(%M'{>>^o&8-
,J!WWN䉝j;S )/(QMK_25l&ߝ
R@@]e 1~HO}ȓH(݅/M9?Z~XmGn|e|{߰KNff.v81еtNOtHvue	;8A;(X:񄒕U]鎜xGcA15Iߞk	/g`G
+BGtt6H4$]SֹKp
=8C*VEGRiS܇\o	&{oUZ-UJxNq)EecDfxvc|گ?(T]wYEޯZ"H'.Zm,_i565El	'/|z~nFNW<X¼}\P%XQ.%I-	Uղ5f&`(aw:mZNX 4L<fN{dUa7SpA)/RM|)Bm:&ۨg̢Dj󇏃+)3CL?^1WF`?u1P:sups]8y)Cok*j
뚙ǎ$5w=!F=O@LЉrRոkY19ۮtרYR֤npzyl|ϥUo)upN9)$'5s2RN[VSa
"%
WiEUY8 q98 !%R\Hn$	?0HBRcWoΈJcKNLY|X@Z)QͥGhlw}sԑ:pk#rpxړaM@ค!J|ϕ
MAE4XpD^o*}oD%q  Z;
zKburS3؃9O92;PgfLSx6ÀUe2n:CZ=/:rҖrJ|U5UvVˉ	'j
 $ 'ӓG]Ph5\[3[YTrxьuކ2IHR΋ܗB5:ۚf=IWWꁶde>!҆oZ4Lq$	҅X@$XёV$@Z$?Ymu
#x"nlP*3rT`pB2ṷj~ٺ6O
H
V1Ƥ>F(qH|@˧h>rв!ʯ"胈$~Qe,$HSց4rI
=~͂G<:*x0io?̏Y߉^xʾH\R=53TieLvɪ^lqǩHeJ}]njn8'5@G%+%hO:|b\ivsmPwjKWMZH骇MoK^T}C#qan[ˆH]hk(V$BɅs]nZ7q/g:1qj1+}nH5+MLMBk[n[/	^GaaadhIz䅈L:E࿜lA-{-)`'ߕ80BbPN:w\svFj7pه=v.U	;7Aڔ-̔
-ÞBȻI\0S$#	eR*<;krߜ+W}8U5{lMdLGQ
}:H'<em,Al0ȷxaonIWT#z[5,i;#F?0lU+,fs!)ic-H
#Oө0[!IQ}QnjbZMXôTUTȖ&GCJPv`
08<Nqf~Mޘ).%uѨ i(jlat`3  FFF!E,DrA̲Bgd`M1TܠsHCCQ*_*\%;{ZyuOj$%#!z

z4CmK" !Vy1.M&7GQ*MH#XDmtW\6$O`9ݻ-'DЀ~-{=K&dѷ8.ﱳ`jACc%v+-/ޏuJѪB5ER`zozʭ 
*qMʕ@QU~dao<oZtQekn,ou9g=\
z35QƅM-8:@էBҹ腗HS"c, !n]b̙ɹd+ Z,t8ŴٔN$O:Sۂ489IaUjcm0ɏdXG_]Hy>Q>Mɭa+)ڌ&