Quite a few computation errors recently - more than one machine

Message boards : Bug reports : Quite a few computation errors recently - more than one machine

To post messages, you must log in.

AuthorMessage
Profile Networkman

Send message
Joined: 22 Nov 07
Posts: 3
Credit: 128,847
RAC: 0
Message 373 - Posted: 20 Dec 2007, 22:11:29 UTC

Been noticing problems yesterday and now today with computation errors on packets that are "Enigma 0.76 5.17" of three seperate machines now. These machines are all standard, no frills, no overclocking type machines that have been working 100% solid since I began the project.

I've not seen any such errors on the newer "Enigma-0.67-Test 5.19" packets.
ID: 373 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 70,470,112
RAC: 388,578
Message 374 - Posted: 20 Dec 2007, 23:21:40 UTC - in response to Message 373.  
Last modified: 20 Dec 2007, 23:29:25 UTC

Been noticing problems yesterday and now today with computation errors on packets that are "Enigma 0.76 5.17" of three seperate machines now. These machines are all standard, no frills, no overclocking type machines that have been working 100% solid since I began the project.


Nothing was changed on the project side recently, could you post links to the failed results ? Maybe it's the weird 'http 404 bug' which I'm trying to trace for last few days. It causes some downloads to fail with 404 error for unknown reason.


I've not seen any such errors on the newer "Enigma-0.67-Test 5.19" packets.


These workunits are very short, so it's unlikely to see them failing with computation errors.

EDIT: I went through the list of your machines and found this:


5.8.16

Maximum disk usage exceeded

]]>


could you check what's inside slot folders ? I've set workunits disk space limit at 8M, it's high enough because the only file that's growing is the result.txt, but I've never seen it larger than 50kBytes.



M4 Project homepage
M4 Project wiki
ID: 374 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Networkman

Send message
Joined: 22 Nov 07
Posts: 3
Credit: 128,847
RAC: 0
Message 376 - Posted: 21 Dec 2007, 1:19:30 UTC

Okay, I should be able to check my home machines by 10pm EST. My sole machine at my desk has 8 folders under "Slots" listed as 0 thru 7; slots 0, 2 & 3 are at about 145k, while slot 1 is up over 9.25 meg.

This particular machine is(was) sharing some time with Rosetta, but I've detached from that project to troubleshoot this issue.


ID: 376 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Networkman

Send message
Joined: 22 Nov 07
Posts: 3
Credit: 128,847
RAC: 0
Message 377 - Posted: 21 Dec 2007, 1:21:55 UTC - in response to Message 376.  

Okay, I should be able to check my home machines by 10pm EST. My sole machine at my desk has 8 folders under "Slots" listed as 0 thru 7; slots 0, 2 & 3 are at about 145k, while slot 1 is up over 9.25 meg.

This particular machine is(was) sharing some time with Rosetta, but I've detached from that project to troubleshoot this issue.


I suppose it would help to know which machine that is - it's ID# 3238 - "heimdall.kdl.net"
ID: 377 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 70,470,112
RAC: 388,578
Message 379 - Posted: 22 Dec 2007, 9:39:34 UTC - in response to Message 377.  

Each result fails with the same error message: -177 ERR_RSC_LIMIT_EXCEEDED. Perhaps some additional (debug?) info is written to result.txt or stderr.txt, and when it gets too large the manager aborts workunit. Could you periodically check what's inside slot dirs used by Enigma ? Usually it looks like this:



results.txt is growing slowly (each time when a result with score higher than previous top score is found, it's added there), but if it's larger than 20-30 kBytes then probably something is wrong.

Stderr usually looks like this:


5.10.20

2007-12-21 23:02:35 enigma: working on range ...
2007-12-22 01:47:35 enigma: finished range


]]>



M4 Project homepage
M4 Project wiki
ID: 379 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : Bug reports : Quite a few computation errors recently - more than one machine




Copyright © 2017 TJM