home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!usc!elroy.jpl.nasa.gov!news.claremont.edu!nntp-server.caltech.edu!SOL1.GPS.CALTECH.EDU!CARL
- From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick)
- Newsgroups: comp.os.vms
- Subject: Re: HELP!!! Security problem for gurus. [Directories]
- Date: 4 Jan 1993 09:31:00 GMT
- Organization: HST Wide Field/Planetary Camera
- Lines: 70
- Distribution: world
- Message-ID: <1i904kINNdbj@gap.caltech.edu>
- References: <1i6dd2INNrct@gap.caltech.edu>,<14628002@zl2tnm.gen.nz>
- Reply-To: carl@SOL1.GPS.CALTECH.EDU
- NNTP-Posting-Host: sol1.gps.caltech.edu
-
- In article <14628002@zl2tnm.gen.nz>, don@zl2tnm.gen.nz (Don Stokes) writes:
- >RMS won't crash a system at all (unless BUGCHECKFATAL is set). The XQP might.
- >RMS lives in exec mode; an unhandled condition will at worst take out the
- >process.
-
- That's technically true, but since RMS calls the XQP, the distinction isn't as
- clear as it might otherwise seem. For example, the first time I was really
- scorched in an INFO-VAX flame, a guy by the name of Jerry Leichter on the
- machine (I think; I'm probably pretty close to the right name)
- venus.ycc.yale.edu took strong exception to one of my posts. In the post to
- which he was responding, I pointed out that having a directory with zero blocks
- allocated to it could crash you system. This would happen any time you tried
- to access a file in the directory (i.e., any time you used a directory lookup
- involving that directory as a directory, not simply as a file in another
- directory; I hope this makes sense; I've gotten a bit verbose because there's
- another thread going on about the distinction between dealing with a directory
- as a file, and dealing with it as a directory). Jerry's gripe wasn't that I
- pointed this out; he objected to my including code that used an unsupported RMS
- routine to truncate the directory). Somehow it doesn't seem fair: I got
- flamed for including all the necessary code to reproduce my problem :-).
-
- >As for case (4), a corrupt system disk is _not_ a good reason to crash
- >the system. In many cases keeping the thing up could be the difference
- >between a clean recovery and having to restore from backup (possibly
- >causing loss of data from between the last backup and the crash, at
- >best losing valuable production time);
-
- But if a trusted image has been corrupted, it can potentially do nasty things
- to files on ALL your disks. Especially if the image in question is the one
- that handles all disk I/O on the system. I'd much rather have to do an image
- restore of my system disk (I can always get it restored to within a week ago)
- than risk having corrupted data in the files my users are *REALLY* interested
- in.
-
- >ANALYZE/DISK/REPAIR _does_ work on a system disk.
-
- Usually. But what if the contents of a file are corrupted? What if RMS
- somehow decides to write to SYSUAF.DAT instead of reading from it (yeah, I
- know, if it does that, at worst SYSUAF.DAT becomes unusable; but there are
- other files that might not be so robust). If the stuff that handles all disk
- I/O knows that it's sick, it should quit. RIGHT NOW (well, YESTERDAY would be
- better, but that doesn't seem possible).
-
- >Of course if it's really scrozzled you might need to
- >crash it and restore anyway. I can't think of any reason for the system
- >to crash because it discovers a damaged file structure (as opposed to
- >crashing because it ate something poisonous as a result), nor can I recall
- >anyone losing a system because of file structure damage.
-
- Well, I don't have the listings at hand, but, let's suppose: RMS gets damaged
- in such a way that when JOE_USER logs in, when RMS looks up the rightslist
- identifiers he holds (part of loginout), it misreads RIGHTSLIST.DAT in such a
- way as to assign him an identifier we'd created to allow someone to modify,
- say, SYSTARTUP_V5.COM? Could that perhaps be considered a problem? I'm not
- claiming that it's likely, just that it's possible. What if we had a file in
- SYS$COMMON:[*...] that was normally supposed to be used, and one in
- SYS$SPECIFIC:[*...] that we used for development purposes, and somehow the
- system disk got hosed in such a way that we could only find the one in
- SYS$SPECIFIC?
-
- In my humble (STOP THE SNICKERING!) opinion, if the system disk is found to
- have been hosed, the system should shut down NOW!
- --------------------------------------------------------------------------------
- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL
-
- Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My
- understanding of astronomy is purely at the amateur level (or below). So
- unless what I'm saying is directly related to VAX/VMS, don't hold me or my
- organization responsible for it. If it IS related to VAX/VMS, you can try to
- hold me responsible for it, but my organization had nothing to do with it.
-