Author |
Message |
|
Several times this week I have found this host has not completed some tasks and the task log says that the client detached.
I thought it may be my Boinc version so I moved back to 6.10.43 but it is still occurring.
Any Ideas? Here is what my message log said the last time it happened:
8/12/2010 6:22:05 PM GPUGRID Sending scheduler request: To fetch work.
8/12/2010 6:22:05 PM GPUGRID Requesting new tasks for CPU
8/12/2010 6:22:11 PM GPUGRID Scheduler request completed: got 0 new tasks
8/12/2010 6:22:11 PM GPUGRID Message from server: Result m6-IBUCH_51b_pYEEI_100304-65-80-RND2775_1 is no longer usable
8/12/2010 6:22:11 PM GPUGRID Message from server: Result f192r185-TONI_CAPBINDsp1-68-100-RND7320_1 is no longer usable
8/12/2010 6:22:11 PM GPUGRID Message from server: No work sent
8/12/2010 6:22:11 PM GPUGRID Message from server: ACEMD beta version is not available for your type of computer.
8/12/2010 6:22:39 PM GPUGRID Computation for task m6-IBUCH_51b_pYEEI_100304-65-80-RND2775_1 finished
8/12/2010 6:22:39 PM GPUGRID Computation for task f192r185-TONI_CAPBINDsp1-68-100-RND7320_1 finished
8/12/2010 6:22:44 PM GPUGRID Sending scheduler request: To fetch work.
8/12/2010 6:22:44 PM GPUGRID Reporting 2 completed tasks, requesting new tasks for GPU
8/12/2010 6:22:50 PM GPUGRID Scheduler request completed: got 2 new tasks
|
|
|
skgivenVolunteer moderator Volunteer tester
Send message
Joined: 23 Apr 09 Posts: 3968 Credit: 1,995,359,260 RAC: 0 Level
Scientific publications
|
Did either of these two results actually run, and if so for how long (roughly, if you noticed)? Obviously the server says 0sec.
f192r185-TONI_CAPBINDsp1-68-100-RND7320_1
m6-IBUCH_51b_pYEEI_100304-65-80-RND2775_1
Did the system restart or anything unusual happen during these times, Virus scan found something, driver updates ran, power flickers, network outage?
You could look at the times they ran and check your logs:
Sent 12 Aug 2010 23:16:09 UTC
Received 13 Aug 2010 1:08:37 UTC
Less than 2h time frame to look at.
Control Panel\System and Maintenance\
Administrative Tools
Start with, Windows Logs and Application, then have a look at Security and system.
Might also want to look at Problem Reports and Solutions.
If you dont get anywhere, I would suggest you uninstall Boinc, then install the latest version 6.10.58 (the one you had been using), and try to troubleshoot it from there, rather than using an older version of Boinc. |
|
|
|
No restarts or anything unusual that I know about, One task was at around 90% and the other around 70%. When it happened in the past the tasks had been running for quite awhile also.
I should mention that another reason I downgraded boinc was that this host also runs rosetta and I was seeing issues there at the same time.
On rosetta I was seeing duplicate computer IDs for this host and the last time that occurred it seemed to coincide with the client detachment here.
This is not my account but the same thing was happening to my 6172 system that has happened with Bigtuna's 1055t system. In my account I have merged hosts so you can't see that is has occurred.
Anyway, that is why I was thinking this is most likely a boinc issue. I might try the latest beta and see what happens.
|
|
|
skgivenVolunteer moderator Volunteer tester
Send message
Joined: 23 Apr 09 Posts: 3968 Credit: 1,995,359,260 RAC: 0 Level
Scientific publications
|
I agree it is probably a boinc issue and I think it relates to the duplicate host identities and security.
Can you merge those accounts on Rosetta? Are you using any account management software?
Might be an idea to detach from Rosetta after merging, do a Boinc uninstall, remove old folders, and then update to the latest version, or Beta. I recall that having different accounts can be problematic and the issues depend on the order of attachment and associated ID, with the oldest taking preference. I’m guessing this is Rosetta, so the problems stemmed from your two accounts conflicting.
I would suggest you ask for advice on the Boinc site. They are the real experts when it comes to Boinc problems; they write it.
Might also be an idea to check with Rosetta, just in case they are having server problems?
- Check you are using the same email address for your accounts.
Reading this BoincStats FAQ might help you understand and resolve the problem. |
|
|
Alain MaesSend message
Joined: 8 Sep 08 Posts: 63 Credit: 1,650,833,561 RAC: 2,256,370 Level
Scientific publications
|
Bikermatt
please check your BAM account and make sure that for your affected host the attach box is checked for GPUGRID in your host list-host details-projects list.
If not, BAM will automatically order a detach.
Kind regards
Alain |
|
|
liveonc Send message
Joined: 1 Jan 10 Posts: 292 Credit: 41,567,650 RAC: 0 Level
Scientific publications
|
Are you sure about this Alain Maes? I use BAM!myself & I've NEVER been able to add freehal@home to BAM! but it doesn't detach freehal@home even though I add it manually via BOINC Client.
____________
|
|
|
|
I'm not using any kind of account management software, and I signed up for a BAM account a few months ago but ended up never using it. I never added any of my hosts or information. I just checked again just in case and there is nothing active on my BAM account.
The host has not detached here in a few days but has also not been running 24/7. It did give me a duplicate again over at rosetta today however there were no GPU grid tasks running at the time so I don't know if it would have detached.
When I fire it up tonight I am going to try the fresh install with the beta like skgiven suggested. If that doesn't help I will post something over at boinc. |
|
|
Alain MaesSend message
Joined: 8 Sep 08 Posts: 63 Credit: 1,650,833,561 RAC: 2,256,370 Level
Scientific publications
|
Are you sure about this Alain Maes?
Oh yes, because I had this happening to me a while ago. Checking the attach box solved the issue,
But of course, if for some mysterious reason a project is not in your BAM list it will presumably also not try to detach it...
Kind regards
Alain |
|
|