OFFICIAL "SERVER PROBLEMS" THREAD

[Madpoo]

Originally Posted by Prime95

In trying to repair the recent huge CPU credit problem, I managed to destroy all the ECM progress data.
I've emailed Aaron to restore from a backup.

Sorry.

It wasn't too bad to roll things back on that table. That's why we do backups (and fortunately had a test server to use).

[axn]

My results from last night is not reflected in the totals, so I submitted them again.

[LaurV]

I just found out that one of my computers that I don't check so often is trying to send proof file since July 2nd (like over 50 days). It always get back an "unauthorized" error about the fact that I didn't send a result first, but the result was sent on 1st of July, as the exponent status shows. The result file was purged long time ago, I don't have it anymore, and even if I would have it, probably I would get an error that this computer already sent this result
Proof file I still have. Maybe I can send it somewhere, and save the future user the time for a PRP re-run?

[Thu Jun 30 21:18:47 2022]
For proof 111618799-8.proof, no entries in need list: 
[Sat Jul  2 22:32:34 2022]
Unexpected error during 111618811-8.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
[Sat Jul  2 23:43:42 2022]
Unexpected error during 111618811-8.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
[Sun Jul  3 00:48:45 2022]
Unexpected error during 111618811-8.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
[Sun Jul  3 01:53:48 2022]

<... snip ...>


[Mon Aug 22 08:48:10 2022]
Unexpected error during 111618811-8.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
[Mon Aug 22 09:53:43 2022]
Unexpected error during 111618811-8.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
[Mon Aug 22 10:58:50 2022]
Unexpected error during 111618811-8.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
[Mon Aug 22 11:49:37 2022]
Unexpected error during 111618811-8.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}

[Prime95]

Send me a link to the proof (or email it to me). I'll make sure it gets uploaded. I don't know what is going wrong as your result did make it to the server and was properly recorded.

EDIT: I've temporarily changed the server to allow uploads of proof files that have an MD5 mismatch. This will "unblock" any prime95's stuck on proof file uploads. If you put your file back in the prime95 folder, it should allow you to upload the proof. The MD5 mismatch may mean that your proof file is corrupt.

[LaurV]

Thanks George. I am still getting the same error, so if you now accept any MD5 (which actually is not good, people will start DoS-ing you with megabytes of crap, haha, you should switch back to how it was), then it means that the issue is not related to MD5. Hopefully the file is good, otherwise somebody will need to re-do the PRP. No big deal anyhow.

I will see where I can put it (probably gugu drive) and send you a link. Tell me if is of any use and is so, what the issue was. If not, scrap it.

[Dobri]

Please consider cancelling all assignments of users with closed accounts like tuluc.alexandru, etc., their GIMPS accounts seem to be closed but their assignments are still active.
For instance, I would like to reserve for a DC test the lowest unverified palindromic exponent, M100707001 if the current assignment is cancelled.

[Prime95]

I manually unreserved 100707001 for you. Feel free to poach an obviously abandoned assignments that are way ahead of the wavefront.

[Dobri]

Originally Posted by Prime95

I manually unreserved 100707001 for you. Feel free to poach an obviously abandoned assignments that are way ahead of the wavefront.

Awesome! Thanks, Prime95.

[ixfd64]

Doesn't seem to be anything serious, but the "recent cleared" list is displaying this message:

Warning: Null value is eliminated by an aggregate or other SET operation.

[Uncwilly]

This exponent https://www.mersenne.org/report_expo...exp_hi=&full=1 is showing up in the query for exponents with 1 LL and 1 PRP that are not verified. But it is verified. Is it a condition problem with the custom query or something else?

[kriesel]

Reporting a P-1 NF result needed for a PRP assignment still incorrectly expires the PRP manual assignment.
See for example https://www.mersenne.org/report_expo...5827987&full=1
(This issue has been previously reported.)

And a manual reassignment can not be obtained.