Exit code -1073741819 (0xC0000005)
22 Nov 2008 1:54:07 UTC
Topic 194047
(moderation:
OK, this task out of the blue decided it would error out. I noticed it was sitting there at 100% with the cpu time still going and stutus showed running. I put BOINC on snooze, exited, then restarted BOINC and it ran for a few more seconds after startup and then straight to the error.
Not seeing anything in it about graphics, so I'm not sure what it might be, though I don't use the graphics and the card is about as up to date as on board graphic card can be from the nvidia site. Don't see any Dump of the Graphics thread, Dump of the Worker thread with houghmap.c in it either.
Haven't been any events that I know of and the only thing that happened at the time of the error was it switching from an AP of SETI back to the WU for Einstein.
I do have a new opti app of AP running for the first time on this pc, but I don't think that has anything to do with it. Nothing strange I can see in the log either other than when the error happened with a read problem. Just thought I'd report it in case there's somethign else I can do to track down the problem.
[Edit} Started looking at my settings and I don't know why this would have been changed, but Leave at least X Gigabytes disk spance free for some odd reason was set to 0.00. Not sure if that could impact this though. I did change it to 1 Gigabyte now[Edit]
Exit code -1073741819 (0xC0000005)
)
I don't know if this is relevant or not but one of the guys had a similar issue over on Seti and it turned out his disk was full. You might want to check if you actually have 1Gb of free disk space.
BOINC blog
RE: OK, this task out of
)
Don't what error exactly is, but the "Leave at least X gigabytes of space free" is fine at 0. Just means that BOINC doesn't really care about how much space it uses, although it's not usually many gigabytes worth.
As for the error code, a quick Google search reveals that it's an "Access Violation." I couldn't tell you what caused it though. Sounds like a machine specific error however.
EDIT: Yeah, a full disk could possibly cause it.
RE: RE: [Edit} Started
)
93 Gig free space on a 144 Gig drive, don't think free space is an issue here. I didn't think the setting Leave at least X Gigabytes disk spance free was the issue either since there was so much free drive space.
Something made it burp right at the end of the task when it should have gone to the next step and contacted Boinc. According to the log it was finished with the task, but couldn't get past that. My guess is in the gathering of the info to get it ready to upload was where it got stuck.
The real funny thing about that is I noticed it when it went from the AP to the Einstein the cycle before, there was still about an hours worth of time left till completion on the WU. In the past I've had similar ones stop and tranfer over to the other project with just a few minutes left, so it ended up waiting an hour before it would start again and finish the last few minutes.
Is it possible that by some fluke BOINC would have gotten the cue it's time to go to the next project becuase it's been an hour at the exact moment the Einstien WU was finishing? I would think that would be built into the software not to happen and just extend it for a bit, but to be honest when I looked at it before it transfered over that's what it looked like had been the case.
Did a search over at Seit but couldn't find anything in a general search of any thread with the error Exit code -1073741819 (0xC0000005)
I've suspended Seti for the moment until I can try and finish the last two tasks I have in my cache. At this point the one I have running now seems to be fine but won't really know until it finishes and gets validated as well as the next one if it's going to be a problem that keeps coming up.
I did notice my wingman did complete the task, so I'm pretty sure it's not the WU.
Exit code -1073741819
)
Exit code -1073741819 (0xC0000005), is actually a windows ver 5.x (2000 and XP) error. More details can be found by searching the MS database but you will probably not find out very much more than it is an access violation. The last three digits 005, should point you to where the violation occurred but may not and there will probably be very little more info.
RE: Exit code -1073741819
)
Thanks for the info. It does seem to be a Windows thing from what I'm seeing, so I'll write it off to a Windows burp as long as things seem to be going ok on the system. If it starts doing it again I'll look into the OS and may put it back on again to attempt to straighten it out.
Thanks again.
RE: I do have a new opti
)
Optimised applications in general put more stress on the CPU than stock apps, and some overclockers found with the initial opti AP app that they had to allow a little more thermal headroom than before. The new opti AP app is even more highly optimised, so will put even more thermal stress on the CPU.
Additionally, Einstein applications have a very high memory throughput when they first start up - I had an Intel motherboard which often suffered a Windows crash during the first few seconds of a new Einstein task. I transferred all components (CPU, RAM, video) to a better-quality motherboard, and it's been running 24/7 ever since with no errors.
So I suggest some attention to temperatures, cooling and overclocking (if any) might be in order.
RE: RE: I do have a new
)
Thanks Richard, I wasn't sure if the AP could have anything to do with it.
I don't OC with the opti apps, I like to run them without having to go thru the hassle of wondering if that's the problem if/when I have problems or dealing with heat issues that may cause. Also just feel it's better for the sys not to.
Been checking the temps and they are what they've always been around 52C which I guess could be lower but is well within operating specs.
Dual ch mem running in dual mode but I haven't tested them recently, it will be one of the things I do if I see this again though...Could always use a better board, but as to wether that was the problem I guess time will tell. No visual problems on it though. Have np putting in a new one if that's what it takes, but I've never lost a MB yet, and knock on wood this isn't the start.
Just cleaned out the pc of dust a week or so ago, wasn't much to clean though. Heat doesn't seem to be an issue, I kind of wish it was I could fix that.
It is the second error task I've had in several months, the first one was the cpu throtle issue and never had that one again after I changed the setting (wasn't on the same pc). The current einstein task I have running is doing fine so far, of course the last one didn't error till the very end once it read 100% so I guess anything could still happen. I've got the task switching back and forth between Einstein and Seti again, just to see if I can make the error happen again. No use to me if I can't switch tasks like before, so I might as well find out now.
This could be a bug, I
)
This could be a bug, I forwarded this to the devs.
Thanks for reporting this.
CU
Bikeman
RE: This could be a bug, I
)
Fished the last two task I had with no error. Will be doing the new AP opti on this pc for a while, but will have it back online to Einstein after I've had a chance to run some of them on it.
This definitely is a bug,
)
This definitely is a bug, thanks for reporting.
The case that the App is interrupted between writing the last checkpoint and writing the final result file is rather rare, but I guess that it crashes every time this happens. I'm trying to fix this.
BM
BM