PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise
1) Message boards : Problems and Help : 321_sr2sieve : missing parameter file (Message 21377)
Posted 4232 days ago by Profile Ananas
Oh, now I understand ... this "factors file" is a result file, not something required for the startup?

In this case it might be caused by the heartbeat thingie of course.

RNA-World has eliminated the heartbeat problem with a minor modification in the wrapper, all wrapper apps with no checkpoint should do that.


If Yoyo did what I recommended, he modified boinc_api.C :

if (standalone) { options.check_heartbeat = false; }


commented out like this :

// if (standalone) { options.check_heartbeat = false; // }



He might have modified wrapper.c instead though, setting options.check_heartbeat = false; there, not sure which method he has choosen.
2) Message boards : Problems and Help : 321_sr2sieve : missing parameter file (Message 21376)
Posted 4232 days ago by Profile Ananas
The heartbeat error is a result of the missing file, it isn't the cause for it. Especially that box has a bit trouble with HDD speed.

2 situations trigger that BOINC heartbeat bug :

- unpacking a CPDN task, i.e. when a new one starts

- several 321_sr2sieve in a row (each having "Factors file not found" in stdout)

In both cases the core client becomes extremely unresponsive.

The problem with CPDN is most likely caused by a quite slow HDD, unpacking the huge amount of stuff required to run those models needs longer than the delay between two heartbeats.

The problem with 321_sr2sieve is new, it started not too long ago.

The heartbeat bug (well, some - or was that "one" ? - stupid people call it a feature) can only be fixed on project application side (i.e. in the wrapper).

p.s.: What I can do of course ... I will set PG to NNW, wait for the current stuff to finish, reset the project and then allow new work. Maybe that fixes it somehow.
3) Message boards : Problems and Help : 321_sr2sieve : missing parameter file (Message 21360)
Posted 4233 days ago by Profile Ananas
bump

Not a single 321_sr2sieve goes through at the moment, they all have the same error.
4) Message boards : Problems and Help : 321_sr2sieve : missing parameter file (Message 21280)
Posted 4236 days ago by Profile Ananas
Factors file not found

in several results, e.g. :

http://www.primegrid.com/workunit.php?wuid=106579338

Not a problem on my side, others have it too.
5) Message boards : Problems and Help : Server can't open database (Message 21075)
Posted 4241 days ago by Profile Ananas
On scheduler contact :

PrimeGrid 16.02.2010 07:53:30 Message from server: Server can't open database
6) Message boards : Number crunching : Compute error Exit status -529697949 (0xffffffffe06d7363) (Message 16848)
Posted 4457 days ago by Profile Ananas
If this is still an issue ... I get it now and then on a machine equipped with 12GB (XP64) with 8GB free (error rate today ~2%)

error list

psp_sr2sieve results, the problem is quite fresh (but so is this box) but it did run a few days without trouble, just today it started to dislike a few results.

No OC'ing on this machine.

Is it possible that the 32bit wrapper limits the memory usage for the task tree or it has trouble when it has to address RAM beyond 4GB? I'm aware that it is mapped for the 32bit application - but does that apply to the shared memory too?

With still 8GB of free physical RAM, it should not run out of memory so easily. The debug output looks as if the wrapper crashed, not the application.
7) Message boards : Number crunching : No Credit for many WUs (Message 9171)
Posted 4865 days ago by Profile Ananas
http://www.primegrid.com/result.php?resultid=45754303

I already had trouble with the llr application months ago, on different boxes (strange enough, one of them is listed under "anonymous" now - but I recognize it from a little core client patch that is always visible ... I'm not using this patch anymore now).

I'm very sure that llr does have a bug, maybe an array that isn't initialized over the full size, an integer that is used before initialized or something like that. Maybe even a side effect of UPX.

If it was only one box that had this problem, I would say it is the computer but the way those failures occur looks much more like a problem in the source code.

There is nothing common between all the failed results, not specific AMD or Intel, not Win2k or XP. Just Linux seems not to be affected (I haven't seen a failed Linux one but that doesn't mean it doesn't exist of course). It hits long running results as well as quickies.
8) Message boards : Sieving : SQL error (Message 9151)
Posted 4867 days ago by Profile Ananas
Warning: mysql_fetch_object(): supplied argument is not a valid MySQL result resource in /home/boinc/proj/html/project/project.inc on line 289

I see this on gcwsieve_3993405 but not on psp_sr2sieve_4747606, so it's probably subproject related
9) Message boards : Number crunching : Viewing completed WU\'s (Message 5513)
Posted 5266 days ago by Profile Ananas
I think I found the reason, one _old_ file in the "slots" directory has been locked on file level, BOINC complained about not beeing able to delete it. Maybe I should reset all projects on that box, who knows what chaos is in the client_state file now :-/

Clearly a problem on my side, no need to take action. Sorry if I caused confusion, thanks :-)
10) Message boards : Number crunching : Viewing completed WU\'s (Message 5512)
Posted 5266 days ago by Profile Ananas
I know :-) But as it's the beta application, I thought I'd report it, in case it's a validator thing or an OS or CPU problem.


Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2021 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 4.03, 3.89, 3.50
Generated 28 Sep 2021 | 4:30:47 UTC