Message boards :
Number crunching :
BOINC 4 and Win98
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 . . . 19 · Next
Author | Message |
---|---|
![]() Send message Joined: 5 Feb 02 Posts: 14 Credit: 1,725,644 RAC: 4 ![]() |
Just in case this information helps: I had installed 4.05 on both of my Win98SE machines yesterday (Sunday), one is a Sony Vaio Laptop, memory 256MB, 500MHz P3, Win98SE (build 4.10.2222) and it had been running fine. This morning (Monday) it was locked up, and I rebooted it because it has locked up on rare occasions in the past, and I didn't think anything of it. Right when I was reading the post on the main page about Win98 problems I went to the laptop and the mouse was very jumpy and it took a long time to move it around. I did C-A-D and saw that BOINC was "Not Responding" and then killed it. I restarted it and found that it had just finished a WU and was trying to upload it. SETI is the only project I am running on it since CPDN won't run on the Win98 machines. I have not seen the problem on the other machine which is one I built a couple of years ago, but will keep an eye on it. |
bjacke ![]() Send message Joined: 14 Apr 02 Posts: 346 Credit: 13,761 RAC: 0 ![]() |
Hi @all, I'm running BOINC under Windows 2000, so I'm not sure wheather it is only my problem. Boinc runs with system start. After start it close it self, if I click in the menu on transfer, but if I wait a moment it works, so I'm realy confused! Projects: S@h + CPDN The whole is more then the sum of its particles. Aristoteles Best wishes from Berlin(52°35'N,13°23'O), Basti |
![]() Send message Joined: 22 Jun 03 Posts: 49 Credit: 447,066 RAC: 0 ![]() |
Just for clarification - when you say "Windows 98" do you mean "Windows 98", "Windows 98SE" or "Windows 98 and 98SE"? What's all this talk of winsock? AFAIK that's only of relevance to dial-up users, and who still has dial-up these days? :/ I'm leaving BOINC alone methinks - just have 3 Predictor WUs left to finish. SETI is too unstable to run it as my sole project. I need Predictor as a backup and there's no knowing when it will convert to v4. The fact that it (a) doesn't work on Windows 98 machines; and (b) could trash those machines has pretty much sealed its fate :) Good luck! -- john |
![]() Send message Joined: 5 Feb 02 Posts: 14 Credit: 1,725,644 RAC: 4 ![]() |
Hi John, I mean Windows 98 Second Edition, thats all I have on two machines. This is an update to my earlier post, with info that may or may not help. I just checked the my home built Win98SE system and BOINC was just sitting there not processing any workunits. It had stopped processing the one in progress in the middle of the WU, but indicated in the "Status" column as "Running". Started sysmon and the CPU was at 4 or 5% usage. I exited BOINC and restarted it, and it resumed executing the WU. BOINC is the only project I have running on that machine. Don |
![]() Send message Joined: 18 Mar 00 Posts: 23 Credit: 3,083,330 RAC: 0 ![]() |
> Oh I got the point, I just have limited power in my apartment. I had to move > one computer, and printer into my bedroom because my living room breaker was > tapped out. I also have a computer in my dining room area. LOL. you got to move to a better breakered house !!! I have 10 computers, a p133 firewall, two monitors, the cable modem, a tv, vcr, and stereo all running off the one wall socket. it's not reset once in the year I've been here. |
![]() Send message Joined: 26 Dec 00 Posts: 5 Credit: 52,388 RAC: 0 ![]() |
I am also having this problem. I am running both S@H and Climate Predictor, and since upgrading to the new BOINC, have noticed that once open, my computer slows down to the point where it takes 30 seconds just to close a window. At this point I have been able to close BOINC through pressing ctrl+alt+del (in one case it was already listed as not responding, in the other, not), and my computer is back to normal again. I am running win98SE. |
bjacke ![]() Send message Joined: 14 Apr 02 Posts: 346 Credit: 13,761 RAC: 0 ![]() |
> Hi @all, > I'm running BOINC under Windows 2000, so I'm not sure wheather it is only my > problem. > Boinc runs with system start. After start it close it self, if I click in the > menu on transfer, but if I wait a moment it works, so I'm realy confused! > Projects: S@h + CPDN > [url=www.adastrawithseti.de] > The whole is more then the sum of its particles. > Aristoteles > Best wishes from Berlin(52°35'N,13°23'O), Basti Hi, don't know why but after complete new installation of BOINC everything is working fine! (only CPDN ?!) |
![]() ![]() Send message Joined: 20 Apr 00 Posts: 141 Credit: 13,912 RAC: 0 ![]() |
Rom, Rather than repost the original message here, please refer to the thread at http://setiweb.ssl.berkeley.edu/forum_thread.php?id=3372 for detailed information on my problem with the new version of SETI/BOINC, including the register and memory dump information from my machine. The memory seems to hold the same block of code as the other dumps posted in this thread. I had to photograph the screen to be able to post this information, since the machine seemed locked up in a tight abend loop. Digital cameras are nice for this sort of thing. When the SETI/BOINC application abended, I could not Cancel out of the error, nor could I Ctrl-Alt-Del to cancel the application or reboot the machine. The only way out was to do a hard reset or power cycle the machine. The SETI@home project (SETI@home 4.03) is the only BOINC 4.05 application being run. Version 3.20 worked without problems. I had just installed BOINC, BOINC had just installed SETI, SETI had just installed it's files, and they had downloaded 13 WUs. The BOINC_GUI showed that SETI had processed about 6 hours and 40 minutes (96+%) of the first WU. Since it still had 12 untouched WUs to process, I don't think it was updating to/from the server. I am constantly connected to the Internet through my university's LAN. My system is an AMD XP 3100+, VIA chipset, 512 MB RAM, one 40GB EIDE HD and one 16MB SCSI HD, and Win98SE with all current security patches and IE6 installed. I run my monitor at 1024 by 768, in True color (32-bit) color mode. I run AVG anti-virus, and trial versions of Trojan Remover and Outpost Pro firewall. The file system is FAT32 on both drives, and I have run Scandisk-standard after terminating SETI/BOINC. Scandisk reported no errors on either drive. Until you get this thing working correctly, I've gone back to SETI Classic, thank you. If I can answer any other questions or be of any help, please ask. Stephen <P>"They better be out there, I think S@H needs their help" (GRIN)</P><a> |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13884 Credit: 208,696,464 RAC: 304 ![]() ![]() |
Just to add to the count. Win98SE, IE6 (fully patched), KT600 chipset, Athlon XP2600+, GeForce4 MX440 video card. Everything appeared to be OK after i upgraded from 3.X to 4.X (shutdown 3.X to install new version). At that stage it wasn't able to get anything- schedulers not responding. Contiunued surfing the net for several hours & went to bed, system still responding normally. Went to check my email the next morning, system hung with the BOINIC_GUI page fault message the others have mentioned (mine being 0177:BFF88396). Couldn't clear the error message, had to reset. After resetting all appeared to be working OK, it had downloaded a couple of dozen work units, there was one ready to be returned. Went to work, came back to find system locked up again. Went to check forums (but the web site was having a fit at the time). When i finally was able to get in here & saw the thread about the sluggish system. I restarted BOINIC & after a few minutes it started to crawl. The 3 fingner salute brought up the task manager & BOINC was listed as Not Responding. The only other software i run on this system is FreeWX 2.03 (a weather programme for my weather station). Good luck Ron. |
Walt Gribben Send message Joined: 16 May 99 Posts: 353 Credit: 304,016 RAC: 0 ![]() |
This problem starts because the BOINC client sits at a HLT instruction. Apparently from either a corrupted return address or bad branch. At least thats what it shows very early in the process. The "HLT" instruction is part of a call, compare, call sequence in shmem.c: >0045179A call dword ptr ds:[54163Ch] >004517A0 cmp esi,esp >004517A2 call 0047FB20 The first call is to CreateFileMapping, which should return to the next instruction at 0x004517a0. Instead its sitting at 0x004517a1, the second byte of that compare instruction, which just happens to be a HLT. Just a guess here, but it probably goes thru this code when starting a new workunit, and for an "update". |
CyberGoyle ![]() Send message Joined: 2 Jun 99 Posts: 160 Credit: 3,622,756 RAC: 26 ![]() ![]() |
After reading this thread, this seems a more puzzling problem than I thought. I have BOINC 4.05 running on three Win98 PC's and two Win95 PC's since Sunday - and not one has page faulted, blue-screened, locked, or slowed down any more than you would expect these old PC's to normally be. Why are my five systems immune from this bug? All are crunching and producing results, none have been rebooted since Sunday. Perhaps 4.05 is conflicting with some other system component that isn't present on my PC's. <a> |
Digitus Send message Joined: 6 Aug 99 Posts: 2 Credit: 4,551,340 RAC: 0 ![]() |
Hi Rom I have the same problems since version 4.05 (also with 4.03) on my systems with win98se (without all updates and also with all updates, without .net framework extension and within). In the last hours I have tried to find some rules, and here they come: Only if there is a new result (the first time it appears, not the results waiting in the upload-queue for a retry) then the upload starts. After this, the WIN32-HANDLES rapidly grows up in som seconds (about 6000 handles per second) from 1400 to over 45000, and Boinc takes nearly 100% of cpu-time. In this time (handles grows up) the system becomes slow, but it is possible to terminate boinc with a task-manager (and a second problem apear - see later). With too many handles (on my system between 44000 and 50000) Boinc crashes with an error message an there is no chance for a soft-reboot - but it is possible to ignore the alert-box and use other programs (shut them down for example). Trys to terminate boinc in this state will freeze the system. (See messages from other users too). The second problem: Terminate boinc by a task-manager In this situation, the actually wu terminates after some seconds with 100% done. This is a new result for uploading (see above) and with the next start of boinc this result will be sent the first time - gotcha. Hope, this will help a little bit. |
![]() ![]() Send message Joined: 5 Apr 03 Posts: 155 Credit: 213,891 RAC: 0 ![]() |
I guess i'll add my 2 cents... The new client has been running quite fine for me. It's when the server is accessed that causes the problem (as discovered before). Somehow, previously i got a big load of WU's, so it has the ability the crunch away nicely. To solve the server accessing problem, I disabled network access on BOINC. no problems for 9 hours so far. (win98SE, all updates etcetc, 333 celery, 64 RAM) I also offer my services as a guinea pig. this is an old junk machine anyways, i don't care too much if she dies. Just tell me what to do, and i'll get on it ASAP! -Fox (e-mail just in case: pyrofox at. sympatico dott. ca) ![]() |
![]() Send message Joined: 13 Mar 03 Posts: 112 Credit: 497,631 RAC: 0 ![]() |
OK, i have posted this message before in an other thread, but i think it's better to post it again in the "official" win9x / ME-CC-hanging-thread! (edit at the end) [quote] Same Problem! :-/ try to connect to the scheduler...it hangs...reboot...try to connect the scheduler...it hangs...reboot... So I try the following. I disconnect the computer from the network before i start the CC, but the win98SE-machine is getting slow too. Then, at this morning, the machine is running and running. but this afternoon, it hangs again! (after the serverproblems at berkeley are over) Sometimes the following error message is occured (german): BOINC_GUI verursachte einen Fehler durch eine ungültige Seite in Modul KERNEL32.DLL bei 017f:bff88396. Register: EAX=c00309c4 CS=017f EIP=bff88396 EFLGS=00010212 EBX=006effec SS=0187 ESP=005efe98 EBP=005f0010 ECX=00000000 DS=0187 ESI=00000000 FS=2177 EDX=bff76855 ES=0187 EDI=bff79060 GS=0000 Bytes bei CS:EIP: 53 56 57 8b 75 10 8b 38 33 db 85 f6 75 2d 8d b5 Stapelwerte: Sometimes when i will "kill" the client by pressing ctrl/lt/del nothing happens. Only a reset can reactivate the PC. But later, when i try it again (after a reset/reboot) the client hangs, but i can "kill" it by ctrl/alt/del. *confused* Different versions of this bug, but ever the same pithy systemhanging! I just wanna crunch! ;-) sorry for this buggy english [edit] I've tested a little on this machine again. So when I "killed" the CC _without_ a reboot, ANY connection to this win98-PC is "crashing" the PC from where I want to connect. The boincclient try to connect to the scheduler immdiately by starting the it. So I think there is a correlation between the schedulerconnect and the crashing. When the CC after a reboot don't start at this machine, there's NO problem with any connection from/to this machine over LAN! Maybe this one can helping in find the bug! _AGAIN_ sorry for the poor (sometimes i mean you can call it "bad" :D ) english! |
![]() Send message Joined: 10 Jun 99 Posts: 9 Credit: 122,638 RAC: 0 ![]() |
> > Yeah, I could see that, except I would have expected the 3.x clients to > have > > run into a problem too. > > > > > Personally, I was having the same problem with the 3.x client. Though it > wasn't an always repeatable occurrence. Same here. I have a Celeron 1.2Ghz with 98SE on a KVM with my main machine (for old non-XP-compatible games - and SETI) and it used to run SETI Classic for days, weeks even, no problems. Since going to BOINC several times I've switched the KVM to the 98 machine and sometimes a blank screen and sometimes a "BOINC has caused an error (or fault, I didn't note the wording) box and clicking it did nothing. I had to hard-reset every time. I haven't installed 4.x on it, I basically gave up on BOINC on 98. It seemed to occur about every second day, or maybe more often and that's just when I noticed/checked it. I just leave it running. Forgot to mention, this only started happening after upgrading from 3.19 to 3.20. |
grumpy Send message Joined: 2 Jun 99 Posts: 209 Credit: 152,987 RAC: 0 ![]() |
Some problem on my xp machine too. The Seti program quits when it fails to go on the internet via my dail-up . ( It don't let it connect auto ,I don't save the password) The gui is running ok . 2004-08-31 08:51:09 [SETI@home] scheduler init_op_project to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi failed, error -1 2004-08-31 08:51:10 [SETI@home] Deferring communication with project for 59 seconds 2004-08-31 08:52:51 [---] Can't resolve hostname setiboinc.ssl.berkeley.edu (valid name, no data record of requested type) 2004-08-31 08:52:51 [SETI@home] scheduler init_op_project to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi failed, error -113 2004-08-31 08:52:51 [SETI@home] Deferring communication with project for 1 minutes and 0 seconds I have to close the gui and restart. |
Gandalf Send message Joined: 15 Jul 01 Posts: 15 Credit: 8,153,890 RAC: 18 ![]() ![]() |
> Great to hear that is being dealt with seriosly, cause my 2 win98se pc's at > home are now idle. > > I do have some input (hopefully usefull) > > It happens with both the cli and gui proggies. > > It happens EXACTLY when a WU have finished and a new 1 is started. > > I do not see this behavior myself. The system begins slow down as soon as it starts to process a workunit and dies within a minute. It does not occur when a WU has finished as it has not been able to finish even the first one downloaded. This is on a FRESH install of win98SE on a P3-400 Celeron |
![]() Send message Joined: 9 Jul 99 Posts: 7 Credit: 139,055 RAC: 0 ![]() |
Guess I'll throw this in to the mix as well. I'm currently running BOINC 4.05/seti 4.03 on a P133. This machine has 48 m ram and is running under win98se. After update it D/L ed 2 wu's It's been happly crunching the first for about 40 hours now with no problem. There is no MOUSE/KEYBOARD or MONITOR attached to the computer and I'm using desktop sharing in netmeeting to monitor it. As best as I can remember there are no updates applied to the install of SE. This machine is just a cruncher so ROM if you need it for anything let or would like any stats on it let me know. rwatson1@hotmail.com is the best place to get me as I don't get to check these boards to often. <p align="center"> ![]() |
![]() Send message Joined: 1 Jul 03 Posts: 1 Credit: 5,684 RAC: 0 ![]() |
> Same problem on my xp machine too. > The Seti program quits when it fails to go on the internet via my dail-up . > ( It don't let it connect auto ,I don't save the password) > > The gui is running ok . > > 2004-08-31 08:51:09 [SETI@home] scheduler init_op_project to > http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi failed, error -1 > 2004-08-31 08:51:10 [SETI@home] Deferring communication with project for 59 > seconds > 2004-08-31 08:52:51 [---] Can't resolve hostname setiboinc.ssl.berkeley.edu > (valid name, no data record of requested type) > 2004-08-31 08:52:51 [SETI@home] scheduler init_op_project to > http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi failed, error -113 > 2004-08-31 08:52:51 [SETI@home] Deferring communication with project for 1 > minutes and 0 seconds > > I have to close the gui and restart. Ditto mate, hope they get a fix on it... |
![]() ![]() Send message Joined: 29 May 99 Posts: 175 Credit: 67,332 RAC: 0 ![]() |
|
©2025 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.