home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.os2.misc
- Path: sparky!uunet!newsflash.concordia.ca!mizar.cc.umanitoba.ca!news!buhr
- From: buhr@umanitoba.ca (Kevin Andrew Buhr)
- Subject: Re: OS/2 2.1 Beta: Bugs and Comments
- In-Reply-To: francis@ese.ogi.edu's message of 2 Jan 93 17:02:37 GMT
- Message-ID: <BUHR.93Jan2183051@ccu.umanitoba.ca>
- Sender: news@ccu.umanitoba.ca
- Nntp-Posting-Host: ccu.umanitoba.ca
- Organization: University of Manitoba, Canada
- References: <47886@ogicse.ogi.edu>
- Date: Sun, 3 Jan 1993 00:30:51 GMT
- Lines: 111
-
- In article <47886@ogicse.ogi.edu> francis@ese.ogi.edu (Francis Moraes) writes:
-
- | What is _it_? A complete system crash, which happens pretty much any time,
- | doing pretty much anything. Here are the details:
- |
- | Location: ##0160:fff60967 - oood:9967.60000, 9084
- |
- | 038600d1
- | Internal revision 6.479, 92/12/09
-
- This is very likely a problem with flaky memory. It probably showed
- up when you started using the beta simply because before that you were
- lucky with which particular byte of which particular program was or
- wasn't affected. Since I began running Netware, TCP/IP, and seamless
- Windows programs at once, for example, I had three crashes like yours
- in a very short span of time, even though I'd never had one before
- (running any of those components separately).
-
- My solution was to flip the banks of SIMMS--my motherboard had eight
- slots, four cards per bank, so I just reversed them all back to
- front--and I haven't had it crash again. This is either because the
- SIMMS were okay and there was just a bad connection which was fixed
- when I swapped them or because the bad memory is no longer having an
- effect in the new card order.
-
- Try doing the same sort of thing. Also, try cleaning the contacts of
- your memory boards with something like rubbing alcohol. Be careful to
- ground yourself before you handle memory, since even a small static
- charge can wreck it.
-
- If you still have trouble, see if your computer came with any memory
- test software. (The test will probably have to be quite rigorous to
- duplicate the kind of stress OS/2 places on memory.) If you know a
- friendly computer technician, see if he or she will give your memory a
- vigorous workout. If he or she finds a problem, your memory may still
- be under warranty (many come with a 1 year warranty against these
- kinds of defects).
-
- | DELL 325D, 12 MB ram, 80 MB Hard Disk with 14 MB free
- ^^^^^^^^^
-
- Do you have two banks of different speeds or types of memory? IBM has
- strongly suggested that OS/2 users not mix memory of different types
- since this seems to cause a lot of failures when the memory is worked
- hard. You may want to consider upgrading the 4 MB bank to an 8 MB
- bank of the same type as your other bank (assuming your current
- configuration is a bank of 8 and a bank of 4) if you have the cash.
- Not only will this probably solve your problem, but OS/2 runs a lot
- nicer in 16 MB and none of the memory banks will be worked quite as
- hard. Be sure to match the memory speed. Preferably, get exactly the
- same model.
-
- | This crashing business continues, even without the network software loaded.
-
- If you have a friend with compatible SIMMS, try borrowing them to use
- in your machine. This will settle the business of whether this is or
- isn't an OS/2 problem.
-
- As a last resort, try removing your 8 MB bank and starting a couple of
- DOS applications and a WIN-OS/2 session. It'll be very slow, but
- it'll probably work. Then try removing the 4 MB bank and doing the
- same thing with the 8 MB bank. If you crash under one configuration
- but not the other then you've certainly narrowed it down. If you
- crash under neither configuration, it's a problem with mismatched
- memory speeds. Finally, if you crash under both, you are either the
- extremely unlucky owner of two bad banks of memory, or you've found a
- very strange bug in the beta.
-
- | At boot, I get the following message:
- |
- | SYS1201: The device driver "C:/OS2/MDOS/VSVGA.SYS" specified in
- | the IFS command on line 101 of the CONFIG.SYS file was not
- | installed. Line 101 is ignored.
-
- For one thing, I would change all the forward slashes in your file
- names to backward slashes.... or is this what you meant? ;)
-
- Why not try posting your entire CONFIG.SYS file? This might help
- narrow down the problem.
-
- | PMX cannot be killed (or _is_ not anyway) after a SHUTDOWN command. So
- | I have to kill it manually. After I kill it, the SHUTDOWN command
- | continues (!) and shuts the computer down the rest of the way. This is
- | very strange, since it _appears_ that the SHUTDOWN command had
- | stopped.
-
- Hmmmm. I had this problem with the original GA (without even the
- service pack) at another site. It isn't all that strange, though.
- The way the OS/2 shutdown procedure is designed, any program can
- prevent the shutdown from completing. Evidently, PMX is doing
- something rude with the shutdown request. While this may be
- irritating, there's nothing wrong with doing a Ctrl-Alt-Del after a
- reasonable length of time (say 15 seconds without disk activity).
-
- | NOTE: I decided to re-install my network software when I originally had
- | problems. This then required that I re-install the CSDs for the network
- | software. Unfortunately, the PMX CSD will _not_ install; it crashes the
- | system at the same place every time. Still, I believe that these things
- | (Bugs three and four) are with OS/2 2.1 since the problems I had before
- | (using PMX with OS/2 2.0) were quite different.
-
- Well... It isn't necessarly an OS/2 2.1 bug. A buggy application may
- work under version 2.0 + SP and not under future versions if the
- programmer did something silly that just *happened* to work with that
- particular version of the operating system. The PMX CSD might very
- well fix your problems, especially since one of them is obviously
- focus-related.
-
- Good luck.
-
- Kevin <buhr@ccu.UManitoba.CA>
-