HECnet

hecnet@lists.dfupdate.se

5 participants
15580 discussions

OpenVMS Volume Shadowing of your system disk

by bqt＠softjar.se

Hi. Mark Wickens wrote: Guys, Thought someone might be interested in my 'the minimum you need to know' guide about setting up a RAID-1 system disk using volume shadowing under OpenVMS: http://lakesdev.blogspot.com/2009/09/volume-shadowing-system-disk-on-vax.ht… Nice writeup. Maybe I shouldn't tell how easy it is under RSX... :-) Johnny

15 years, 11 months

OpenVMS Volume Shadowing of your system disk

by mark＠wickensonline.co.uk

Guys, Thought someone might be interested in my 'the minimum you need to know' guide about setting up a RAID-1 system disk using volume shadowing under OpenVMS: http://lakesdev.blogspot.com/2009/09/volume-shadowing-system-disk-on-vax.ht… Cheers, Mark. --

15 years, 11 months

Security hole in CSWS

by bqt＠softjar.se

gerry77 at mail.com wrote: On Wed, 23 Sep 2009 10:25:36 +0200, you wrote: Wow. I apologize. I thought I had that under control... :-) I wonder if VMS have changed along the way, or if it has always been like that...? It's my feeling that it has been always like that, anyway I can certify that VAX/VMS V4.7 works as I expected: [...Same tests as done by Bob] FLORRY::RPT$ type test.dat;-0 one FLORRY::RPT$ show system VAX/VMS V4.7 on node FLORRY 23-SEP-2009 10:21:58.14 Uptime 15 08:30:54 Pid Process Name State Pri I/O CPU Page flts Ph.Mem 00000100 NULL COM 0 0 15 08:29:27.66 0 0 00000101 SWAPPER HIB 16 0 0 00:00:00.06 0 0 [...] Bye, :) G. P.S.: This V4.7 was obtained by upgrading a V4.6, so I'm assuming that even on that release the behaviour would have been the same. I would expect that if anything, the change would have happened between V3 and V4 of VMS. V3 had the same 9+3 filename length limits as RSX... (Started using VMS at V3.7 (I think it was), a long long time ago...) Johnny

15 years, 11 months

Security hole in CSWS

by gerry77＠mail.com

On Wed, 23 Sep 2009 10:25:36 +0200, you wrote: Wow. I apologize. I thought I had that under control... :-) I wonder if VMS have changed along the way, or if it has always been like that...? It's my feeling that it has been always like that, anyway I can certify that VAX/VMS V4.7 works as I expected: [...Same tests as done by Bob] FLORRY::RPT$ type test.dat;-0 one FLORRY::RPT$ show system VAX/VMS V4.7 on node FLORRY 23-SEP-2009 10:21:58.14 Uptime 15 08:30:54 Pid Process Name State Pri I/O CPU Page flts Ph.Mem 00000100 NULL COM 0 0 15 08:29:27.66 0 0 00000101 SWAPPER HIB 16 0 0 00:00:00.06 0 0 [...] Bye, :) G. P.S.: This V4.7 was obtained by upgrading a V4.6, so I'm assuming that even on that release the behaviour would have been the same.

15 years, 11 months

Security hole in CSWS

by bqt＠softjar.se

Bob Armstrong wrote: Actually, in VMS (as well as in RSX), ;-1 means the oldest version available. I think you only have ;-1 and ;0 which are special. Can't speak for RSX's problems, but for VMS none of this is true. Wow. I apologize. I thought I had that under control... :-) I wonder if VMS have changed along the way, or if it has always been like that...? In RSX it is as I described, however... .dir login.cmd;* Directory DU4:[BQT] 23-SEP-09 10:23 LOGIN.CMD;48 3. 26-FEB-09 10:27 LOGIN.CMD;47 3. 16-FEB-09 11:37 LOGIN.CMD;46 3. 13-DEC-08 06:22 Total of 9./15. blocks in 3. files .dir login.cmd;0 Directory DU4:[BQT] 23-SEP-09 10:23 LOGIN.CMD;48 3. 26-FEB-09 10:27 Total of 3./5. blocks in 1. file .dir login.cmd;-0 Directory DU4:[BQT] 23-SEP-09 10:23 LOGIN.CMD;48 3. 26-FEB-09 10:27 Total of 3./5. blocks in 1. file .dir login.cmd;-1 Directory DU4:[BQT] 23-SEP-09 10:23 LOGIN.CMD;46 3. 13-DEC-08 06:22 Total of 3./5. blocks in 1. file . Johnny -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt at softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol

15 years, 11 months

Security hole in CSWS

by bob＠jfcl.com

Actually, in VMS (as well as in RSX), ;-1 means the oldest version available. I think you only have ;-1 and ;0 which are special. Can't speak for RSX's problems, but for VMS none of this is true. ----- $ create foo.txt one ^Z $ create foo.txt two ^Z $ create foo.txt three ^Z $ dir foo.txt Directory USERS:[BOB] FOO.TXT;3 1/80 22-SEP-2009 18:52:28.94 [BOB] (RWED,RWED,RWED,RE) FOO.TXT;2 1/80 22-SEP-2009 18:52:26.34 [BOB] (RWED,RWED,RWED,RE) FOO.TXT;1 1/80 22-SEP-2009 18:52:21.09 [BOB] (RWED,RWED,RWED,RE) Total of 3 files, 3/240 blocks. $ type foo.txt;1 one $ type foo.txt;0 three $ type foo.txt three $ type foo.txt; three $ type foo.txt;-1 two $ type foo.txt;-2 one $ type foo.txt;-3 %TYPE-W-SEARCHFAIL, error searching for USERS:[BOB]FOO.TXT;-3 -RMS-E-FNF, file not found $ type foo.txt;-0 one ------- Give it a try for yourself (there are lots of VMS systems on HECnet!). Bob

15 years, 11 months

Security hole in CSWS

by bqt＠softjar.se

gerry77 at mail.com wrote: On Tue, 22 Sep 2009 16:07:03 -0700, you wrote: That's legal. The longer story is that version numbers start with 1, and ";0" is shorthand for the most recent version. "FOO.TXT;" is just shorthand for "FOO.TXT;0". It's also legal to say "FOO.TXT;-1" meaning the previous version (i.e. most recent -1), as is ";-2", ";-3", etc. It's legal ";-0" too, meaning the oldest version available :) Actually, in VMS (as well as in RSX), ;-1 means the oldest version available. Around here is where VMS differs from TOPS-20. ;0 also have different meanings on TOPS-20 compared to VMS. On VMS, ;0 means the most recent version if you open a file for reading. But if you try to create a file, ;0 means one higher than what exist right now. In TOPS-20, unless my memory fails me, ;0 will overwrite the most recent version that do exist. The ;-0, ;-1, ;-2, ;-3 and so on to refer to older versions is how I think TOPS-20 did it. I don't remember that there is a way of specifying something like that in VMS. I think you only have ;-1 and ;0 which are special. But I might be remembering wrong from playing too much with RSX. :-) Johnny -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt at softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol

15 years, 11 months

Security hole in CSWS

by drb＠msu.edu

It's legal ";-0" too, meaning the oldest version available :) One is starting to hope that PHP gets _some_ of the filespec stuff right. :) But, while I'm willing to believe (at least until I have time to study it a bit more) that it's PHP which looks to see if it owns the requested extension, I'm reluctant to believe that Apache hands over full pathname processing. That should reduce the size of the problem space a little. I wonder if case matters... The rule might better read: RewriteRule (.*\.[Pp][Hh][Pp])([;.][0-9-]*)(.*) $1$3 De

15 years, 11 months

Security hole in CSWS

by gerry77＠mail.com

On Tue, 22 Sep 2009 16:07:03 -0700, you wrote: That's legal. The longer story is that version numbers start with 1, and ";0" is shorthand for the most recent version. "FOO.TXT;" is just shorthand for "FOO.TXT;0". It's also legal to say "FOO.TXT;-1" meaning the previous version (i.e. most recent -1), as is ";-2", ";-3", etc. It's legal ";-0" too, meaning the oldest version available :) G.

15 years, 11 months

Security hole in CSWS

by bob＠jfcl.com

As in "FOO.TXT;" ? That certainly *used* to be a valid filename, back in V4/V5 days anyway, That's legal. The longer story is that version numbers start with 1, and ";0" is shorthand for the most recent version. "FOO.TXT;" is just shorthand for "FOO.TXT;0". It's also legal to say "FOO.TXT;-1" meaning the previous version (i.e. most recent -1), as is ";-2", ";-3", etc. And, FWIW, it's also legal to use angle brackets in VMS path specifications (e.g. "DISK:<DIR.DIR.DIR>FOO.TXT"). Both the use of angle brackets and using a second "." in place of the semicolon (e.g. "FOO.TXT.1") were put into VMS long ago as an aid to ex-TOPS20 users. Bob

15 years, 11 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

HECnet