I think I got my 'client error' problem sorted, finally.
In my case, the culprit was ATI X1950 graphic card (MSI). When it was brand new, only Einstein@Home client exhibited signs of trouble, but over time (half a year) things got much worse.
Additional problems showed up - graphic artefacts, all USB devices disappearing (as if disconnected) suddenly, failing to boot and crashing/hanging. In its final stage, the system needed rebooting several times per week.
Replaced ATI with Gigabyte nVidia GeForce 8800GT and so far - no more client errors.
Actually, this was to be expected as several (same model and make) ATI cards had already failed before on other computers. I'm happy to see it go, it didn't work well on my Linux box anyway.
So far, so "good", what drives me crazy is the fact that this error occures repeatedly (est. every second workunit) but the client shouldn't be the problem, why?
- no overclocking, nothing but the recommended standard apps
- machine >24 hrs prime95 stable (tested with various settings)
- machine >24 hrs memtest stable
- no instabilities or problems at this system generally,
btw. i had statistics at college and i know that a system never can be prooved as 'error free', but it seems to me that it should be unlikely a hardware problem with theese results
- never uses screensaver mode --> no driver problems should occure (as the error-message in the worker-dump implicates)
Further details:
- os is Windows XP SP2 with a AMD Sempron 3300+ (as shown in the result link)
- Happened and happens with the S5R3_4.26 app and the S5R4_6.04 app
- I suspected the version change from the recommended BOINC 5.10.45 to recommended 6.2.14 - which i unfortunately processed on this machine - as the "guilty". After the trouble began: client detached, all boinc software uninstalled (completely, manually controlled), reinstalled 5.10.45, attached again and same things showed up again :-(
And in the results which 'running through', sometimes a 'WARNING: Fixing yUpper (15922093 -> 53) [HoughMap.c 776]' can be found in the chechmark-rows of the stderr.txt. But after all, the result ends sucessfully. The link i've posted is different, it seems to crashed with theese warning. Normally it warns without a crash or it crashes without this warning but exactly the same access violation in the same address. Now the confusion is complete, at least for me :-)
Btw. Bernd wrote
Quote:
we are currently hunting this.
But that was in august 2007. Any news?
To put it in a nutshell, I'm completely running out of ideas. Last option i see is to stop calculate workunits on this system (due to energy save and science protection). But as you can imagine, i would prefer an other solution, so any suggestions are welcome. thx in advance, bye Henk.
---
To prevent question like this:
Quote:
But it also seem to appear on overcloked systems, or some with CPU voltage to low, or PSU that isn't delivering stable enough Voltages...
no, no, most stable voltage line ever seen on an enermax psu, no defect E-CAPs nowhere. As i mentioned before, no nothing hardware none ;-)
Quote:
Did you try to disable graphics ?
No use.
Quote:
Did you try to update your graphic drivers ?
No possibility, using Radeon9000, latest WHQL-Driver is >2 yrs. old, currently using this one since 2 yrs without any problems.
Quote:
Did you run a memtest86 to ensure RAM stability ?
More than once, more than one programm, usually at least 12 hrs. Result? no errors.
Quote:
Is you system overclocked in any ways ?
No over- or under- clocking or -volting. Nowhere.
---
Hi, joined Einstein a couple off month ago and havibg difficulties, with this type: 14-9-2008 13:18:34|Einstein@Home|Resuming task h1_0818.10_S5R4__559_S5R4a_0 using einstein_S5R4 version 604 . This task ended in ERROR @ 96% completion, no GRAPHICS problem, I also run SETI with an optimized app.'s64BIT(AK_WINXP64_SSSE3) with BOINC 6.2.18, the E@H app is 32 BIT. The host is a QX9650 @ 3.4GHz (10x340)(5/6:DDR2=408MHz.;OCZ REAPER PC6500)ASUS P5E MoBo, PSU=650Watt.
Not a clue,where these faults are coming from, I don't suspect my ATI EAH 2400Pro, anylonger, at happens at random. I, never noticed a fault or a crash, after using the graphics, I don't the 'screensaver' either, thoughg.
Linking to the tasks you have problems with is more useful than having us do a search, you know? I see you two error code 99s and two exit code -1073741819 (0xc0000005)s around the 9th of September.
I am constantly running into these general access violation errors (-1073741819 (0xC0000005)) with one of my machines. For example see this result file 112043057. For the moment I will not accept any new work units for this machine and let the cache run dry.
Machine is not overclocked and while computing I have not screen saver active, so no access to the graphics occured.
I am constantly running into these general access violation errors (-1073741819 (0xC0000005)) with one of my machines. For example see this result file 112043057. For the moment I will not accept any new work units for this machine and let the cache run dry.
Machine is not overclocked and while computing I have not screen saver active, so no access to the graphics occured.
Regards
Uba36
This could be a bug. If it is, the new Beta App that's out right now is supposed to have fixed it. I had a run in with that error myself here, but it does seem to be affecting you much more than it did me, so you might want to also look into some of the other advice given in the thread.
I am constantly running into these general access violation errors (-1073741819 (0xC0000005)) with one of my machines. For example see this result file 112043057. For the moment I will not accept any new work units for this machine and let the cache run dry.
Machine is not overclocked and while computing I have not screen saver active, so no access to the graphics occured.
Regards
Uba36
This could be a bug. If it is, the new Beta App that's out right now is supposed to have fixed it. I had a run in with that error myself here, but it does seem to be affecting you much more than it did me, so you might want to also look into some of the other advice given in the thread.
I gave the new beta a try. I have 2 units ( one at 75%, the other at 79% 9 in prgress. Lets see what happens.
first error i have seen running this project. http://einsteinathome.org/task/113664604
h1_0955.95_S5R4__647_S5R4a_2
Exit status -1 (0xffffffffffffffff)
CPU time 0
stderr out
I just had something like 18 tasks die with the same error message as shown in my earlier post
the boinc manager shows this:
1/10/2009 10:46:42 PM|Einstein@Home|Starting task h1_0955.95_S5R4__636_S5R4a_0 using einstein_S5R4 version 610
1/10/2009 10:46:58 PM|Einstein@Home|Finished download of l1_0956.35_S5R4
1/10/2009 10:46:59 PM|Einstein@Home|Finished download of h1_0956.35_S5R4
1/10/2009 10:48:43 PM|Einstein@Home|Computation for task h1_0955.95_S5R4__636_S5R4a_0 finished
1/10/2009 10:48:43 PM|Einstein@Home|Output file h1_0955.95_S5R4__636_S5R4a_0_0 for task h1_0955.95_S5R4__636_S5R4a_0 absent
1/10/2009 10:48:43 PM|Einstein@Home|Starting h1_0955.95_S5R4__635_S5R4a_0
1/10/2009 10:48:43 PM|Einstein@Home|Starting task h1_0955.95_S5R4__635_S5R4a_0 using einstein_S5R4 version 610
1/10/2009 10:50:43 PM|Einstein@Home|Computation for task h1_0955.95_S5R4__635_S5R4a_0 finished
1/10/2009 10:50:43 PM|Einstein@Home|Output file h1_0955.95_S5R4__635_S5R4a_0_0 for task h1_0955.95_S5R4__635_S5R4a_0 absent
then it goes to this:
1/10/2009 11:09:51 PM|Einstein@Home|Fetching scheduler list
1/10/2009 11:09:56 PM|Einstein@Home|Master file download succeeded
1/10/2009 11:10:47 PM|Einstein@Home|Computation for task h1_0956.00_S5R4__621_S5R4a_0 finished
1/10/2009 11:10:47 PM|Einstein@Home|Output file h1_0956.00_S5R4__621_S5R4a_0_0 for task h1_0956.00_S5R4__621_S5R4a_0 absent
1/10/2009 11:10:47 PM|rosetta@home|Resuming task t075_1_NMRREF_1_t075_1_id_model_04_coreIGNORE_THE_REST_idl_6205_7348_0 using rosetta_beta version 598
1/10/2009 11:26:52 PM|Einstein@Home|Sending scheduler request: To fetch work. Requesting 259201 seconds of work, reporting 12 completed tasks
1/10/2009 11:26:57 PM|Einstein@Home|Scheduler request completed: got 0 new tasks
1/10/2009 11:26:57 PM|Einstein@Home|Message from server: No work sent
1/10/2009 11:26:57 PM|Einstein@Home|Message from server: (reached daily quota of 2 results)
and now it goes with:
1/11/2009 1:34:32 AM|Einstein@Home|Sending scheduler request: To fetch work. Requesting 174371 seconds of work, reporting 0 completed tasks
1/11/2009 1:34:37 AM|Einstein@Home|Scheduler request completed: got 0 new tasks
1/11/2009 1:34:37 AM|Einstein@Home|Message from server: No work sent
1/11/2009 1:34:37 AM|Einstein@Home|Message from server: (reached daily quota of 2 results)
I think I got my 'client
)
I think I got my 'client error' problem sorted, finally.
In my case, the culprit was ATI X1950 graphic card (MSI). When it was brand new, only Einstein@Home client exhibited signs of trouble, but over time (half a year) things got much worse.
Additional problems showed up - graphic artefacts, all USB devices disappearing (as if disconnected) suddenly, failing to boot and crashing/hanging. In its final stage, the system needed rebooting several times per week.
Replaced ATI with Gigabyte nVidia GeForce 8800GT and so far - no more client errors.
Actually, this was to be expected as several (same model and make) ATI cards had already failed before on other computers. I'm happy to see it go, it didn't work well on my Linux box anyway.
Last 'client error' log: http://einsteinathome.org/task/99847503
I'm not sure if i do right to
)
I'm not sure if i do right to post on a 'sticky-thread' but as other people did before i hope it was a right decision not to open on a new thread...
The trouble is...
Exactly what Bernd described as
and it appears in the
with the
example: Result 104319847
So far, so "good", what drives me crazy is the fact that this error occures repeatedly (est. every second workunit) but the client shouldn't be the problem, why?
- no overclocking, nothing but the recommended standard apps
- machine >24 hrs prime95 stable (tested with various settings)
- machine >24 hrs memtest stable
- no instabilities or problems at this system generally,
btw. i had statistics at college and i know that a system never can be prooved as 'error free', but it seems to me that it should be unlikely a hardware problem with theese results
- never uses screensaver mode --> no driver problems should occure (as the error-message in the worker-dump implicates)
Further details:
- os is Windows XP SP2 with a AMD Sempron 3300+ (as shown in the result link)
- Happened and happens with the S5R3_4.26 app and the S5R4_6.04 app
- I suspected the version change from the recommended BOINC 5.10.45 to recommended 6.2.14 - which i unfortunately processed on this machine - as the "guilty". After the trouble began: client detached, all boinc software uninstalled (completely, manually controlled), reinstalled 5.10.45, attached again and same things showed up again :-(
And in the results which 'running through', sometimes a 'WARNING: Fixing yUpper (15922093 -> 53) [HoughMap.c 776]' can be found in the chechmark-rows of the stderr.txt. But after all, the result ends sucessfully. The link i've posted is different, it seems to crashed with theese warning. Normally it warns without a crash or it crashes without this warning but exactly the same access violation in the same address. Now the confusion is complete, at least for me :-)
Btw. Bernd wrote
But that was in august 2007. Any news?
To put it in a nutshell, I'm completely running out of ideas. Last option i see is to stop calculate workunits on this system (due to energy save and science protection). But as you can imagine, i would prefer an other solution, so any suggestions are welcome. thx in advance, bye Henk.
---
To prevent question like this:
no, no, most stable voltage line ever seen on an enermax psu, no defect E-CAPs nowhere. As i mentioned before, no nothing hardware none ;-)
No use.
No possibility, using Radeon9000, latest WHQL-Driver is >2 yrs. old, currently using this one since 2 yrs without any problems.
More than once, more than one programm, usually at least 12 hrs. Result? no errors.
No over- or under- clocking or -volting. Nowhere.
---
Hi, joined Einstein a couple
)
Hi, joined Einstein a couple off month ago and havibg difficulties, with this type: 14-9-2008 13:18:34|Einstein@Home|Resuming task h1_0818.10_S5R4__559_S5R4a_0 using einstein_S5R4 version 604 .
This task ended in ERROR @ 96% completion, no GRAPHICS problem, I also run SETI with an optimized app.'s64BIT(AK_WINXP64_SSSE3) with BOINC 6.2.18, the E@H app is 32 BIT.
The host is a QX9650 @ 3.4GHz (10x340)(5/6:DDR2=408MHz.;OCZ REAPER PC6500)ASUS P5E MoBo, PSU=650Watt.
Not a clue,where these faults are coming from, I don't suspect my ATI EAH 2400Pro, anylonger, at happens at random. I, never noticed a fault or a crash, after using the graphics, I don't the 'screensaver' either, thoughg.
Linking to the tasks you have
)
Linking to the tasks you have problems with is more useful than having us do a search, you know? I see you two error code 99s and two exit code -1073741819 (0xc0000005)s around the 9th of September.
Which one(s) do you mean?
I am constantly running into
)
I am constantly running into these general access violation errors (-1073741819 (0xC0000005)) with one of my machines. For example see this result file 112043057. For the moment I will not accept any new work units for this machine and let the cache run dry.
Machine is not overclocked and while computing I have not screen saver active, so no access to the graphics occured.
Regards
Uba36
RE: I am constantly running
)
This could be a bug. If it is, the new Beta App that's out right now is supposed to have fixed it. I had a run in with that error myself here, but it does seem to be affecting you much more than it did me, so you might want to also look into some of the other advice given in the thread.
RE: RE: I am constantly
)
I gave the new beta a try. I have 2 units ( one at 75%, the other at 79% 9 in prgress. Lets see what happens.
uba36
first error i have seen
)
first error i have seen running this project.
http://einsteinathome.org/task/113664604
h1_0955.95_S5R4__647_S5R4a_2
Exit status -1 (0xffffffffffffffff)
CPU time 0
stderr out
6.4.5
- exit code -1 (0xffffffff)
]]>
Validate state Invalid
I just had something like 18
)
I just had something like 18 tasks die with the same error message as shown in my earlier post
the boinc manager shows this:
1/10/2009 10:46:42 PM|Einstein@Home|Starting task h1_0955.95_S5R4__636_S5R4a_0 using einstein_S5R4 version 610
1/10/2009 10:46:58 PM|Einstein@Home|Finished download of l1_0956.35_S5R4
1/10/2009 10:46:59 PM|Einstein@Home|Finished download of h1_0956.35_S5R4
1/10/2009 10:48:43 PM|Einstein@Home|Computation for task h1_0955.95_S5R4__636_S5R4a_0 finished
1/10/2009 10:48:43 PM|Einstein@Home|Output file h1_0955.95_S5R4__636_S5R4a_0_0 for task h1_0955.95_S5R4__636_S5R4a_0 absent
1/10/2009 10:48:43 PM|Einstein@Home|Starting h1_0955.95_S5R4__635_S5R4a_0
1/10/2009 10:48:43 PM|Einstein@Home|Starting task h1_0955.95_S5R4__635_S5R4a_0 using einstein_S5R4 version 610
1/10/2009 10:50:43 PM|Einstein@Home|Computation for task h1_0955.95_S5R4__635_S5R4a_0 finished
1/10/2009 10:50:43 PM|Einstein@Home|Output file h1_0955.95_S5R4__635_S5R4a_0_0 for task h1_0955.95_S5R4__635_S5R4a_0 absent
then it goes to this:
1/10/2009 11:09:51 PM|Einstein@Home|Fetching scheduler list
1/10/2009 11:09:56 PM|Einstein@Home|Master file download succeeded
1/10/2009 11:10:47 PM|Einstein@Home|Computation for task h1_0956.00_S5R4__621_S5R4a_0 finished
1/10/2009 11:10:47 PM|Einstein@Home|Output file h1_0956.00_S5R4__621_S5R4a_0_0 for task h1_0956.00_S5R4__621_S5R4a_0 absent
1/10/2009 11:10:47 PM|rosetta@home|Resuming task t075_1_NMRREF_1_t075_1_id_model_04_coreIGNORE_THE_REST_idl_6205_7348_0 using rosetta_beta version 598
1/10/2009 11:26:52 PM|Einstein@Home|Sending scheduler request: To fetch work. Requesting 259201 seconds of work, reporting 12 completed tasks
1/10/2009 11:26:57 PM|Einstein@Home|Scheduler request completed: got 0 new tasks
1/10/2009 11:26:57 PM|Einstein@Home|Message from server: No work sent
1/10/2009 11:26:57 PM|Einstein@Home|Message from server: (reached daily quota of 2 results)
and now it goes with:
1/11/2009 1:34:32 AM|Einstein@Home|Sending scheduler request: To fetch work. Requesting 174371 seconds of work, reporting 0 completed tasks
1/11/2009 1:34:37 AM|Einstein@Home|Scheduler request completed: got 0 new tasks
1/11/2009 1:34:37 AM|Einstein@Home|Message from server: No work sent
1/11/2009 1:34:37 AM|Einstein@Home|Message from server: (reached daily quota of 2 results)
One possible cause and
)
One possible cause and solution to this error as found here on Einstein.