Recovering from power outage.

Is this why I got a whole slew of 8/26-8/28 expiration date jobs?
Incidentally, running multiple GPU jobs at once doesn't work all that well on my machine. I've found 3 work okay, 7+ and one or more of them, (usually all of them,) return a calculation error. I'm manually throttling the jobs to keep them from running more than 3 at a time until the rush jobs are all done and I can get back to processing one 6.10 at a time alongside the 6.03s.

I hope all of the data will be recovered- I had three computers with no more tasks because they were all ready to transmit.

Anyone know if this was just an annoying delay (I had 6 Astro's and they take 30+hours each), or did we lose credits?

Ben

care about credits, back online! Lets go find some alien civilization!

Thanks to all those hard working people behind the scenes, maintaining the Seti@home network, for us, the crunchers, so we can continue our efforts to process the W.U's. Your efforts are greatly appreciated by many.

And thanks goes out to Speedy, for the info on where to find the directory listing the downed servers at Berkeley. This is usefull info to have during these trying times.

Once again, thanks for the efforts ev1 is giving, it is appreciated.

Cheers!
____________

Oh, that's the reason for all the tasks in my manager.
Hoping for soon restart of the uploads and good state of the servers.
____________
Only a boy with responsabilities of an old man

I was wondering what was up well here's to a speedy recovery guys and see you back when you get back. I also notice something about a Nivida gpu problem I am running a gtx520 they were saying it was the older nivida gpus however what is the definition of an OLD nivida gpu??
____________

Have you tried using SpinRite on the drives of the server? That ought to fix any performance issues resulting from drive damage.

http://www.spinrite.com

"The drives of the server" are fairly new (months) and run in RAID
If one fails they replace it with new empty drive which is automatically resynced (filled with the same 'old' data that was on the broken HDD)

(and 'spinrite' can in fact induce final death to an old aging HDD if you have one)


____________



- ALF - "Find out what you don't do well ..... then don't do it!" :)

Well even new drives can get damaged during a power outage. Say one drive is damaged. Seek errors begin to occur that creates a performance hit while the drive tries to retrieve the data.

Even in RAID a 100% healthy drive is better than a slightly damaged one. If the drive is totally dead, they'll replace it of course. However, if it keeps working they may not investigate further.

The original post was about performance issues after a power outage. Damaged components, hard drives in particular, might be the cause.

And if your drive is in such bad shape it might die from Spinrite, then you should back it up. Spinrite even warns you about that if SMART-data is available.

Hi, I was wondering why doesn't SETI send out certificates like Einstein does? After a lot of years I've finally reached the million mark, though I've downloaded the certificate thing that is offered. How about sprucing the certificates appearance up at least?

Sincerely,
Arkynut