No GPU WU's?

[Zerix01]

Has anyone else been having issues getting GPU WU's? Before the weekend I was getting nothing but these messages

10:15:40] + Attempting to get work packet
[10:15:40] - Connecting to assignment server
[10:15:40] + No appropriate work server was available; will try again in a bit.
[10:15:40] + Couldn't get work instructions.
[10:15:40] - Attempt #6 to get work failed, and no other work to do.
Waiting before retry.

Then I lost power in that big ice storm that happened in New England, now I'm just getting my system back up to speed but I'm still getting no WU's. My SMP client seems to be fine. The WU it had was past the due date so it trashed it then grabbed a new one immediately. So things seem fine on my end, I know the SMP and GPU servers are separate but I have not seen anyone else complaining about this.

[torin3]

Has anyone else been having issues getting GPU WU's? Before the weekend I was getting nothing but these messages



Then I lost power in that big ice storm that happened in New England, now I'm just getting my system back up to speed but I'm still getting no WU's. My SMP client seems to be fine. The WU it had was past the due date so it trashed it then grabbed a new one immediately. So things seem fine on my end, I know the SMP and GPU servers are separate but I have not seen anyone else complaining about this.

I've got my full compliment of GPU2 clients running WUs now. There is one that will be finishing in 5 minutes. I'll report back if it gets a new one.

Edit: Picked it up normally. I'd say check your configuration on the client and see if it has something exclusionary checked. Like not allowing units over 10MB in size.

[Zerix01]

I'd say check your configuration on the client and see if it has something exclusionary checked. Like not allowing units over 10MB in size.

I'll double check that when I get home. Normaly all my clients are set to the highest limits on everything and allowing experimental WU's. RAM would be the one variable but with 4GB's of RAM I upped the limit to 2GB's on that client I think. No WU's should need that for some time to come.

[Zerix01]

Well I checked on the client. Big work units are set, advanced methods is on, memory set to 2048 MB's. I did end up looking through the log and saw it finished two WU's, it took more than 10 retries between each. They were different projects too so it is not being picky about that, 4737 and 4744. SMP client still cranking along.

I remember seeing a Wine update last week before the storm. I wonder if something on that end is messing up the client. I would expect it to be more of a solid issue though, like never getting any WU's. Anyone know of a way to ping the F@H servers through Wine?

[Zerix01]

Another Wine update was pushed and I did a distro upgrade from Kubuntu 8.04 to 8.10 and still no changes in this client. Any suggestions?

[Zerix01]

Well seems fixed. I killed the FAHcore file, work folder was already empty. Then the client started downloading and trying to run CAL (AMD) based WU's!? I am using the -forcegpu nvidia_g80 flag but the client didn't seem to care much. I tried this twice with the same results until I just killed the whole folder and re-installed the client. All better now :santa: I just hope it stays fixed after it finishes this unit.