<Rude words> 2608

Snapshot · 10 Jul 2007 at 20:05

Just had two 2608 WUs on this PC hang at 23% immediately after an attempt to upload any queued WUs. Restarting them resulted in MISSING_WORK_FILES and a new WU being downloaded. :mad:

Mattus · 10 Jul 2007 at 22:59

I feel your pain. I lost a completed 1523-pointer earlier. Just died at 'will end MPI now' and restarting it gave the dreaded MISSING_WORK_FILES.

theheyes · 10 Jul 2007 at 23:30

I had a 1385 vanish at 7% yesterday but it downloaded a 1780 in its place so I wasnt that bothered.

SiriusB · 10 Jul 2007 at 23:37

I'm doing quite well with 1760 pointers. Had 3 in a row, then got a 1523, then a 1385, now just finished another 1760.

I know the pain though. I lost two 1760pointers a while back in the space of 2 days. Was not a happy bunny.

Snapshot · 11 Jul 2007 at 06:44

Gah, now my borged server has chucked a 2604 (1760 points) in the bin at 94%. No wonder I'm not getting away from Neil as I should be.

shadowscotland · 11 Jul 2007 at 07:59

This seams to be happening more oftern, I guess they are tweeking the beta.
Or it might have been a problem for ages but due to the recient server issue it's just showing itself now

nfoulger · 11 Jul 2007 at 10:07

Not getting away

I've been in torture over the new SMP beta changeover on Intel Mac OS X. I'm not sure about the history here, but the first "Installer client" download that became available seemed to be withdrawn so there was just an empty space in the F@H download link. However, I managed to download and install something before it was withdrawn. (May be it was faulty - that would explain why it was leaving my machines standing idle.) Then I stumbled upon the revised version which downloaded OK. Up to that point only one machine seemed to be producing with any degree of reliability - early termination, failing to connect to server, etc. Average ppd has dropped like a stone.
I think it's picking back up again now, but 2K is not the same as 3K ppd.

Neil

Snapshot · 11 Jul 2007 at 10:39

shadowscotland said:
This seams to be happening more oftern, I guess they are tweeking the beta.
Or it might have been a problem for ages but due to the recient server issue it's just showing itself now

I think it's since we got the new client after 1st July.

shadowscotland · 11 Jul 2007 at 11:08

I'd like to agree but i think its just coincidence, as the mix of wu's has also changed reciently (to add yet another viable)

Has anyone posted in the official forum regarding this?

Cob · 11 Jul 2007 at 13:19

I've had another

[0]0:Return code = 0, signaled with Segmentation fault
[0]1:Return code = 0, signaled with Segmentation fault
[0]2:Return code = 0, signaled with Segmentation fault
[0]3:Return code = 0, signaled with Segmentation fault
[07:49:39] CoreStatus = 0 (0)
[07:49:39] Client-core communications error: ERROR 0x0
[07:49:39] Deleting current work unit & continuing...

during the night. That's atleast 4 since the new clients came out.

Snapshot · 12 Jul 2007 at 07:00

This PC (E6600) has just had a third 2608 fail at exactly the same place as the other two but the E6300 has crept through to 55% so far. So it looks like it's somehow PC related. I'll have a clear out of old cores and work files when the current WU ends.

shadowscotland · 12 Jul 2007 at 07:53

Snapshot said:
This PC (E6600) has just had a third 2608 fail at exactly the same place as the other two but the E6300 has crept through to 55% so far. So it looks like it's somehow PC related. I'll have a clear out of old cores and work files when the current WU ends.

Is it the identical wu crashing each time (run clone gen) did happen to me a while back - thats what the -delete ## (insert number) flag is for

Snapshot · 12 Jul 2007 at 20:51

Well spotted, Sherlock. They were all 1 487 21, even the two that were consecutive. If I ever meet the assignment server I'll give it a good kicking.