ATI GPU: No appropriate work server was available

[bapriebe]

Code: Select all

[13:02:07] Connecting to http://assign-GPU.stanford.edu:8080/
[13:02:07] Posted data.
[13:02:07] Initial: 0000; + No appropriate work server was available; will try again in a bit.
[13:02:07] + Couldn't get work instructions.
[13:02:07] - Attempt #9  to get work failed, and no other work to do.  Waiting before retry.

171.64.65.102 is shown in REJECT mode.
171.64.65.103 is shown in ACCEPTING mode with WU's available. Yet I'm not getting any...

[toTOW]

On next update, Server Status shows both server working ... but yes, 103 is low on work

[bapriebe]

It might be low but it wasn't completely out. Why would it refuse to assign to it ,low or not?

[bruce]

It might be low but it wasn't completely out. Why would it refuse to assign to it ,low or not?

There is a threshold that it doesn't go below. I can only guess, but I suspect it has to do with the coordination if information between the Assignment Server and the various Work Servers and whatever associated timing issues might exist.

It takes about 7 minutes to generate the data we see on serverstat.html and it is posted about every 10 minutes. As completed WUs are returned, new WUs are generated, so the number changes frequently between the updates that we can see.

[bapriebe]

Are 171.64.65.102 and 171.64.65.103 the only ATI GPU work servers? We're back in the same boat we were earlier with 102 in REJECT and 103 critically low on work. One GPU here idled for 1.5hrs and another about to be...

[EvilAlchemist]

It seems I am having this issue as well.

Code: Select all

[20:23:35] + Attempting to send results [June 19 20:23:35 UTC]
[20:23:35] - Reading file work/wuresults_09.dat from core
[20:23:35]   (Read 177313 bytes from disk)
[20:23:35] Connecting to http://171.67.108.17:8080/
[20:23:35] - Couldn't send HTTP request to server
[20:23:35]   (Got status 503)
[20:23:35] + Could not connect to Work Server (results)
[20:23:35]     (171.67.108.17:8080)
[20:23:35] + Retrying using alternative port
[20:23:35] Connecting to http://171.67.108.17:80/
[20:23:35] - Couldn't send HTTP request to server
[20:23:35]   (Got status 503)
[20:23:35] + Could not connect to Work Server (results)
[20:23:35]     (171.67.108.17:80)
[20:23:35]   Could not transmit unit 09 to Collection server; keeping in queue.
[20:23:35] + Sent 0 of 1 completed units to the server
[20:23:35] - Autosend completed
[20:49:00] + Attempting to get work packet
[20:49:00] - Will indicate memory of 4027 MB
[20:49:00] - Connecting to assignment server
[20:49:00] Connecting to http://assign-GPU.stanford.edu:8080/
[20:49:00] Posted data.
[20:49:00] Initial: 0000; + No appropriate work server was available; will try again in a bit.
[20:49:00] + Couldn't get work instructions.
[20:49:00] - Attempt #11  to get work failed, and no other work to do.
Waiting before retry.
[21:37:14] + Attempting to get work packet
[21:37:14] - Will indicate memory of 4027 MB
[21:37:14] - Connecting to assignment server
[21:37:14] Connecting to http://assign-GPU.stanford.edu:8080/
[21:37:14] Posted data.
[21:37:14] Initial: 0000; + No appropriate work server was available; will try again in a bit.
[21:37:14] + Couldn't get work instructions.
[21:37:14] - Attempt #12  to get work failed, and no other work to do.
Waiting before retry.

[bapriebe]

It seems I am having this issue as well.

102 just dispensed a WU here...

[toTOW]

Are 171.64.65.102 and 171.64.65.103 the only ATI GPU work servers?

Yes