FAH gpu servers down.

vertica

vertica

vertica

Associate
Joined
21 Apr 2008
Posts
1,088
Location
By the river
Just incase anyone was wondering:

It appears as per this .

The gpu assignment servers are currently down. And given its not yet 10am in stanford, we could be waiting a few hours for WUs to available.

For those of you too lazy to look at the link, the work means the stats databse site is down; stats updates have been suspended and the PS3 assignment server may be down.
 
Last edited:
Bump!

One of the main Nvidia Work Servers is out of work, and the Assignment server isn't directing clients to the other servers with work.

2am in stanford so this time we really could be in for a wait before it is sorted.

Mod has posted this:
"Server is up and running, but has no WU to go and I don't know why the AS doesn't send us to 171.64.65.102 which has plenty of work ... "
Click to expand...
 
Last edited:
my poor card doesn't know what to do! its not used to only having to render the desktop!:p
Ah well full power to the smp clients it is
 
Well it seems they are working on it, Mr Pande's blog has finally acknowlegeded the issue and said

May 21, 2009
Issues with GPU AS migration
We have had some issues with the GPU assignment server (AS) migration to new hardware. Several issues have been resolved, including

- issues with ATI GPU client work assignments
- some work servers without work


But there are two remaining issues which we have filed tickets with Stanford IT, including
- port 80 forwarding for the GPU AS
- allowing assignments to 171.64.122.70


Of these remaining issues, the port 80 issue is a big issue for those who have port 8080 blocked and we've been pushing IT to get this resolved ASAP. The second issue (going to 171.64.122.70) is less time critical (that server can still receive completed WUs from clients), but we would like to get more GPU work servers on line in general.
Click to expand...

EDIT: can't send or receive again!
 
Last edited:
by VijayPande on Thu May 21, 2009 11:11 pm

Looking at the AS and work servers, there seems to be lots of normal assigns and jobs, but we will continue to look on our side as well to see what's up.

The server migration did prove to be much messier than we originally had thought. For now, all the known issues have been resolved, but there may be something else that we haven't caught yet.
Click to expand...

lets see what tomorrow brings then.
 
for the last 4 days (and i'm assuming today will be the same) our team ppd is down a bit due to this, but not by much compared to some.

Our graph here

Following the official forums, they really dont seem to know what is going on, or when or why. Judging by the few responses theat have been given by Pande group, i dont think they have realised how much this is affecting their work.

As a comparison take a look at folding@EVGA's ppd: down from an average of 5.25M to 3.75M yesterday, thats nearly a 30% drop in output! linky
 
Didn't have any issues over night, just checked logs and got two units shortly after you posted at 2:40am

Have had no issues since thursday...runs away to find some wood
 
You must log in or register to reply here.
Back
Top Bottom