Rom, in my PC running 98SE what seems to be the only thing that blocks the PC is when the client tries to make an update (automatically or manually). Apparently, it closes the SETI client and starts getting very slow instantly.

Everything else worked fine. I had 3 WU already processed, after the server started responding the WU were successfully uploaded and the client continued crunching the current WU. When I tried to manually update to report the already uploaded WU the machine started getting blocked.

I then C+A+D and noticed that the SETI client had been closed and the boinc_gui was already not responding. I killed it and the PC returned to its normal state.

Hope it helps.

I went to CPDN messageboards page to see if they have alike problem, but no.
May it be taht we're forcing S@H 4.03 problem and not BOINC ?


Regards

I have a BOINC issue with WinXp Prof. SP2, that is maybe related to some of this problems.

The BOINC client closes while requesting work from the seti server.

The errormessage in stderr.txt is:

***UNHANDLED EXCEPTION****
Reason: Access Violation (0xc0000005) at address 0x00406F7B read attempt to address 0x00000208

After restarting BOINC I usually get this error:
"Unrecoverable error for result ..."

See also: http://setiweb.ssl.berkeley.edu/forum_thread.php?id=3463

correction, the retry now works fine, and all my wu's uploaded successfully. It is the update that bogs the machine.

> > but i think THE question is why computers hangs under ver 4.05 of BOINC
> > and not under ver 3.20???
> > what are the difference beetween those 2 versions???
> my wild guess is that Win98 just isn't that great for running multithreaded
> & multiproc apps, and WinME was little better. Stuff like W2K & XP,
> which are based from NT, seem to be more stable. Since BOINC 4 adds a lot of
> preemptive task switching etc, that probably isn't as "friendly" for 98/ME.
> If you look at the Windows API help library, it seems like half the "fun
> stuff" (i.e. process/thread calls) isn't available for 98/ME.

Maybe a silly request, I'm not so up-to-date with everything. I think the full source code for BOINC 4 is available right? If above is true why not compile a version without multiprocessor/-threading and what else more that's not needed for older PCs and/or 98/98SE/ME? If the guys developing BOINC set a few checks in the sourcecode for compile-inclusion or not of these features, so anyone can easily create a 'Vanilla EXE file'. I mean, wouldn't that make more sense rather than finding a bug that's probably due to using these advanced features in the first place? Fixing this kind of bug will slow down computers where the bug doesn't take place. Also without the extra 'features' it would increase efficiency on older OSes and hardware with the Vanilla version too. Everyone happy. That is, if that is the true reason for all the crashes...

I've got 6 98SE PCs on BOINC (not doing much right now) and 20 more 98SE PCs waiting for BOINC to be installed when and only when all starts to work well (i.e. stable release versions of both BOINC and SETI). Changing the OS on any of these PCs is very unlikely, as 98SE is by far the fastest and still full-functional Windows around. I upgraded BOINC on 1 of those 6 PCs, and had to uninstall it because the PC froze up every time BOINC kicked in. I don't mind helping/testing in any way I can, as far as its possible for me.

jacQues

Okay, I found two knowledge base articles that describe problems which would lead to some of the symptoms we are seeing here.

Could somebody give this patch a try and tell me the results?

http://support.microsoft.com/default.aspx?scid=kb;en-us;242161

Here is an article that spells out the differences in the platforms.

http://support.microsoft.com/default.aspx?scid=kb;en-us;125713

I found one memory leak and I'm in the process of fixing it.

----- Rom
BOINC Development Team, U.C. Berkeley

Running v 4.05 on Win98 SE.

Problems that I have been having include: lockups, slow performance and a work unit that was downloaded after installing 4.05 - the status remains in the "downloading" state.

Still running classic. Upgraded to 4.05 because that was the only way I could get a new work unit. Did not uninstall 3.xx version before installing 4.05.

Bruce

> Okay, I found two knowledge base articles that describe problems which would
> lead to some of the symptoms we are seeing here.
>
> Could somebody give this patch a try and tell me the results?
>
> http://support.microsoft.com/default.aspx?scid=kb;en-us;242161
>
> Here is an article that spells out the differences in the platforms.
>
> http://support.microsoft.com/default.aspx?scid=kb;en-us;125713
>
> I found one memory leak and I'm in the process of fixing it.
>
> ----- Rom
> BOINC Development Team, U.C. Berkeley
> <a> href="http://www.boinc.dk/index.php?page=user_statistics&project=sah&userid=85465">

> Rom, I did a file search and can't find a vmm.vxd on my system. I do find
> vmm32.vxd. The MS knowledge base article states that:
>
>
> The English-language version of this fix should have the following file
> attributes or later:
> 9/21/99 11:07am 4.10.1999 472,564 Vmm.vxd Windows 98
> 9/21/99 10:33am 4.10.2223 472,564 Vmm.vxd Windows 98 Second Edition
>
> Could that be my problem or would you like me to apply the fix and see?

I would check the file sizes first. Are they roughly the same?

Does the patch contain a file named Vmm32.vxd, maybe the KB article is just wrong on the filename.

----- Rom
BOINC Development Team, U.C. Berkeley

> I would check the file sizes first. Are they roughly the same?
>
> Does the patch contain a file named Vmm32.vxd, maybe the KB article is just
> wrong on the filename.
>
> ----- Rom
> BOINC Development Team, U.C. Berkeley
> <a> href="http://www.boinc.dk/index.php?page=user_statistics&project=sah&userid=85465">

Hi All,
Rom and fellow volunteers, I have been going through the posts here and it seems to be talking just about WIN 98. I have Win ME and am experiencing the exact same problem as been quoted in these posts.. Everything is fine until I hit the update key. Then it is Hung Solid. I have had to exit. It is running right now with the Network Disabled. I hope we can find a "fix" for ME to.
My computer is as follows:
Microsoft Windows Millennium , (04.90.3000.00)

Warm Regards All,

Rocky

I have WinMe after deblocking and a new install, I got the following two uploading WU with missing files? (I deleted the client too before reinstall)

--- - 2004-08-31 21:50:29 - Starting BOINC client version 4.05 for windows_intelx86
SETI@home - 2004-08-31 21:50:29 - Project prefs: using your defaults
SETI@home - 2004-08-31 21:50:29 - Host ID is 7648
--- - 2004-08-31 21:50:29 - General prefs: from SETI@home (last modified 2004-07-15 07:28:46)
--- - 2004-08-31 21:50:29 - General prefs: using your defaults
SETI@home - 2004-08-31 21:50:31 - Unrecoverable error for result 04my04aa.2277.12321.573572.36_4 (One or more missing files)
SETI@home - 2004-08-31 21:50:31 - Started upload of 25ap04aa.23553.12000.990916.169_5_0
SETI@home - 2004-08-31 21:50:31 - Started download of setiathome_4.03_windows_intelx86.exe
--- - 2004-08-31 21:50:31 - May run out of work in 2.00 days; requesting more
SETI@home - 2004-08-31 21:50:31 - Requesting 117949 seconds of work
SETI@home - 2004-08-31 21:50:31 - Sending request to scheduler: http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
SETI@home - 2004-08-31 21:50:33 - Finished upload of 25ap04aa.23553.12000.990916.169_5_0
SETI@home - 2004-08-31 21:50:33 - Throughput 27616 bytes/sec
SETI@home - 2004-08-31 21:50:33 - Finished download of setiathome_4.03_windows_intelx86.exe
SETI@home - 2004-08-31 21:50:33 - Throughput 263518 bytes/sec
SETI@home - 2004-08-31 21:50:34 - Started download of better_banner.jpg
SETI@home - 2004-08-31 21:50:34 - Scheduler RPC to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded
SETI@home - 2004-08-31 21:50:34 - Project prefs: using your defaults
SETI@home - 2004-08-31 21:50:34 - Unrecoverable error for result 04my04aa.2277.15569.1040906.84_1 (One or more missing files)
SETI@home - 2004-08-31 21:50:34 - Started download of 04my04aa.25870.21536.447140.144
SETI@home - 2004-08-31 21:50:34 - Finished download of better_banner.jpg
SETI@home - 2004-08-31 21:50:34 - Throughput 14892 bytes/sec
SETI@home - 2004-08-31 21:50:34 - Started download of 04my04aa.25870.21536.447140.101
SETI@home - 2004-08-31 21:50:38 - Finished download of 04my04aa.25870.21536.447140.144
SETI@home - 2004-08-31 21:50:38 - Throughput 74753 bytes/sec
SETI@home - 2004-08-31 21:50:38 - Started download of 04my04aa.25870.21536.447140.98
SETI@home - 2004-08-31 21:50:39 - Finished download of 04my04aa.25870.21536.447140.101
SETI@home - 2004-08-31 21:50:39 - Throughput 75692 bytes/sec
SETI@home - 2004-08-31 21:50:39 - Started download of setiathome_4.03_windows_intelx86.pdb
SETI@home - 2004-08-31 21:50:43 - Finished download of 04my04aa.25870.21536.447140.98
SETI@home - 2004-08-31 21:50:43 - Throughput 82416 bytes/sec
SETI@home - 2004-08-31 21:50:51 - Finished download of setiathome_4.03_windows_intelx86.pdb
SETI@home - 2004-08-31 21:50:51 - Throughput 399003 bytes/sec
SETI@home - 2004-08-31 21:50:51 - Starting result 04my04aa.2277.15569.1040906.84_1 using setiathome version 4.03
--- - 2004-08-31 21:51:25 - Suspending network activity - user request
--- - 2004-08-31 21:51:41 - Resuming network activity
--- - 2004-08-31 21:52:18 - Suspending network activity - user request


Problem stops when connection has been switched off...

Greetings from Belgium ;-))

Actually, I think we are dealing with three seperate issues.

But, until I can rule out at least one of them, it'll continue to look like the same issue.

----- Rom
BOINC Development Team, U.C. Berkeley

Rom, the Vmm.vxd file is 910K and dated 9/2/98

I've been trying to get the fix. can't find it to download it. trying help via email and when they try to connect me to https://webresponse.one.microsoft.com/
the page won't load. Tried HP.com, without result. I'll keep trying

Well, my PIII finished its wu, it got to 100% and began to bog, after ten minutes, ctr alt del -boinc(not responding) I end task, restart boinc, and it uploads the finished wu and starts on crunching the next one. The celery is still crunching fine as well.

Not getting any gui faults on either machine anymore, or at least for the time being.

The "ready to reports" are stacking up, but if I hit update, I know its going to crash again. I do wonder whats going to happen when my cache gets low and it starts looking for the scheduler again...

rom, I've sent in a request to Support@HP they say they'll respond in an hour. we'll see.

Hello all. My P III with win 98 is behaving the same way as described in this thread. I had trouble after rebooting but ran scan disk and defragged and now running classic untill fix is worked out.
I will make this machine available if dev wants to run a beta test with debug code ect.
AS an aside The 4 xp machines on bionic that I operate are running very well with the new ver. of software.

Just a note on download issues (with Win2k this time): I was downloading ClimatePredictor and got a corrupt download (on the EXE for it) for some reason (Checksum didn't match and the file size was 400k too small...probably download stopped prematurely). However, the problem occured when BOINC tried to delete it to redownload.

BOINC running - said "Unable to delete (filename.exe)" (or something similar).
I "reset ClimatePredictor" (only the EXE had downloaded at that point) hoping it would clear the Unable to delete problem. First thing it said "Unable to delete (filename.exe)"
So I went to the ClimatePredictor directory and tried to manually delete the file and got "File sharing violation - unable to delete"
I exited BOINC, and then I could delete it.

Is there some reason BOINC would be unable to delete a file that had an error in download?

(File IO problem? Might give a clue as to some of the Win9x problems.)