mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   PrimeNet (https://www.mersenneforum.org/forumdisplay.php?f=11)
-   -   OFFICIAL "SERVER PROBLEMS" THREAD (https://www.mersenneforum.org/showthread.php?t=5758)

Syntony 2018-10-23 00:33

End of day chron job stalled?
 
I notice that it's well after midnight UTC on 23rd October, but the end of day chron tasks for 22nd don't appear to have been run yet (e.g. assignment time-outs and category boundaries not updated)... :confused2:

Syntony 2018-10-24 23:15

Well, things worked as normal 24 hours later, so chron jobs ran OK between 23rd & 24th October

chalsall 2018-10-24 23:35

[QUOTE=Syntony;498699]Well, things worked as normal 24 hours later, so chron jobs ran OK between 23rd & 24th October[/QUOTE]

It's cron.

ramgeis 2018-12-19 11:12

I got this message after submitting my latest LL DC result:
[QUOTE]Notice: Undefined index: timing in C:\inetpub\v5\v5server\gimps\0.95_cpu_credit.inc.php on line 72 [/QUOTE]
It has been treated as a LL test result in terms of credits given. Something seems broken...

axn 2018-12-19 11:32

[QUOTE=ramgeis;503318]I got this message after submitting my latest LL DC result:

It has been treated as a LL test result in terms of credits given. Something seems broken...[/QUOTE]

Is that the giant test? I bet the server code doesn't have timing information for that non-standard huge FFT size.

ramgeis 2018-12-19 11:54

[QUOTE=axn;503319]Is that the giant test? I bet the server code doesn't have timing information for that non-standard huge FFT size.[/QUOTE]
When submitting the even bigger one some months ago I didn't get this message. My wild guess is that the server doesn't expect a DC result in that range. (?)

axn 2018-12-19 13:14

[QUOTE=ramgeis;503322]When submitting the even bigger one some months ago I didn't get this message. My wild guess is that the server doesn't expect a DC result in that range. (?)[/QUOTE]

Possibly. Although... there is no credit difference between LL & DC. Only the exponent (i.e. # iterations) and FFT size used are relevant parameters for the calculation.

ramgeis 2018-12-19 13:52

1 Attachment(s)
[QUOTE=axn;503327]Possibly. Although... there is no credit difference between LL & DC. Only the exponent (i.e. # iterations) and FFT size used are relevant parameters for the calculation.[/QUOTE]
Good point. Maybe it helps if I attach a screenshot of the submission.
The server clearly seems to recognize the result as a DC, nevertheless the credit has been assigned to my LL stats.
Then in the table the "submitted" field doesn't contain a value, a situation which I didn't had before.

GP2 2018-12-19 14:48

In cases where the first LL test has a scary error code (namely 02001500 in this case) and has [B]Status[/B] = "Suspect", the double check assignment is usually treated as a first-time test by the server.

For example, [M]M98111053[/M] is "Suspect". If you manually assigned yourself the double-check, I think the link in the previous sentence would display [B]Work Type[/B] = "LL first test".

Prime95 2018-12-20 02:49

[QUOTE=ramgeis;503318]I got this message after submitting my latest LL DC result:

It has been treated as a LL test result in terms of credits given. Something seems broken...[/QUOTE]

Fixed.

ramgeis 2018-12-22 16:44

[QUOTE=GP2;503336]For example, [M]M98111053[/M] is "Suspect". [...][/QUOTE]
Is there a page where those results marked as "Suspect" are listed or can be queried?


All times are UTC. The time now is 23:07.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.