Server - 171.64.65.106

[patonb]

Having issues on this server afterreassigned from 171.67.108.25

[Oldhat]

bruce wrote
I hope that the file attribute on the servers will be modified to fix this "problem" but I believe that when you get the response that FF wants to download a file, you can assume it is the OK message. If the server cannot be contacted, FF is much clearer than IE.

Thanks for the clarification on that. I can only imagine that this is a modification to FF version 3, as I'm fairly sure FF version 2 displayed the "OK" without the additional file download request.

Either that, or old age is finally catching up with me.

[Teddy]

GPU servers in meltdown again, can't send or receive work for any of my GPU clients. SMP is fine, standard client is fine.
We need more servers for GPU clients & where is this new server code they keep promising? It seems like it has been in testing for a long time.

Teddy

& No I am not getting notification emails either, no matter will check in the morning.

[statesidecoma]

same here as well, can't send work nor get work.

[Jester]

From the logs my rigs have been fine most of the day.... until now,
Third evening running, around the same time,

[11:06:30] Folding@home Core Shutdown: FINISHED_UNIT
[11:06:34] CoreStatus = 64 (100)
[11:06:34] Sending work to server
[11:06:34] Project: 5771 (Run 6, Clone 171, Gen 450)
[11:06:34] - Read packet limit of 540015616... Set to 524286976.


[11:06:34] + Attempting to send results [May 21 11:06:34 UTC]
[11:06:35] - Couldn't send HTTP request to server
[11:06:35] + Could not connect to Work Server (results)
[11:06:35] (171.67.108.11:8080)
[11:06:35] + Retrying using alternative port
[11:06:36] - Couldn't send HTTP request to server
[11:06:36] + Could not connect to Work Server (results)
[11:06:36] (171.67.108.11:80)
[11:06:36] - Error: Could not transmit unit 07 (completed May 21) to work server.
[11:06:36] Keeping unit 07 in queue.
[11:06:36] Project: 5771 (Run 6, Clone 171, Gen 450)
[11:06:36] - Read packet limit of 540015616... Set to 524286976.


[11:06:36] + Attempting to send results [May 21 11:06:36 UTC]
[11:06:38] - Couldn't send HTTP request to server
[11:06:38] + Could not connect to Work Server (results)
[11:06:38] (171.67.108.11:8080)
[11:06:38] + Retrying using alternative port
[11:06:39] - Couldn't send HTTP request to server
[11:06:39] + Could not connect to Work Server (results)
[11:06:39] (171.67.108.11:80)
[11:06:39] - Error: Could not transmit unit 07 (completed May 21) to work server.
[11:06:39] - Read packet limit of 540015616... Set to 524286976.


[11:06:39] + Attempting to send results [May 21 11:06:39 UTC]
[11:06:40] - Couldn't send HTTP request to server
[11:06:40] (Got status 503)
[11:06:40] + Could not connect to Work Server (results)
[11:06:40] (171.67.108.25:8080)
[11:06:40] + Retrying using alternative port
[11:06:40] - Couldn't send HTTP request to server
[11:06:40] (Got status 503)
[11:06:40] + Could not connect to Work Server (results)
[11:06:40] (171.67.108.25:80)
[11:06:40] Could not transmit unit 07 to Collection server; keeping in queue.
[11:06:40] - Preparing to get new work unit...
[11:06:40] + Attempting to get work packet
[11:06:40] - Connecting to assignment server
[11:06:41] + No appropriate work server was available; will try again in a bit.
[11:06:41] + Couldn't get work instructions.
[11:06:41] - Attempt #1 to get work failed, and no other work to do.
Waiting before retry.
[11:06:52] + Attempting to get work packet
[11:06:52] - Connecting to assignment server
[11:06:53] + No appropriate work server was available; will try again in a bit.
[11:06:53] + Couldn't get work instructions.
[11:06:53] - Attempt #2 to get work failed, and no other work to do.
Waiting before retry.
[11:07:14] + Attempting to get work packet
[11:07:14] - Connecting to assignment server
[11:07:15] + No appropriate work server was available; will try again in a bit.
[11:07:15] + Couldn't get work instructions.
[11:07:15] - Attempt #3 to get work failed, and no other work to do.
Waiting before retry.
[11:07:49] + Attempting to get work packet
[11:07:49] - Connecting to assignment server
[11:07:50] + No appropriate work server was available; will try again in a bit.
[11:07:50] + Couldn't get work instructions.
[11:07:50] - Attempt #4 to get work failed, and no other work to do.
Waiting before retry.
[11:08:34] + Attempting to get work packet
[11:08:34] - Connecting to assignment server
[11:08:35] + No appropriate work server was available; will try again in a bit.
[11:08:35] + Couldn't get work instructions.
[11:08:35] - Attempt #5 to get work failed, and no other work to do.
Waiting before retry.
[11:09:55] + Attempting to get work packet
[11:09:55] - Connecting to assignment server
[11:09:56] - Successful: assigned to (171.64.65.106).
[11:09:56] + News From Folding@Home: Welcome to Folding@Home
[11:09:56] Loaded queue successfully.
[11:09:57] + Could not connect to Work Server
[11:09:57] - Attempt #6 to get work failed, and no other work to do.
Waiting before retry.
[11:12:46] + Attempting to get work packet
[11:12:46] - Connecting to assignment server
[11:12:47] - Successful: assigned to (171.64.65.106).
[11:12:47] + News From Folding@Home: Welcome to Folding@Home
[11:12:47] Loaded queue successfully.
[11:12:48] + Could not connect to Work Server
[11:12:48] - Attempt #7 to get work failed, and no other work to do.
Waiting before retry.
[11:18:16] + Attempting to get work packet
[11:18:16] - Connecting to assignment server
[11:18:17] + No appropriate work server was available; will try again in a bit.
[11:18:17] + Couldn't get work instructions.
[11:18:17] - Attempt #8 to get work failed, and no other work to do.
Waiting before retry.

If it's anything like the previous two days the "problem" will just suddenly go away (after a couple of hours ) and all rigs run fine.

[dempaSD]

Third period in three days - "unable to get work packet". Affecting all gpu's finishing WU's now. Happened yesterday, last night and now in the morning again. When will the Core_14 servers have more data to crunch?

Thanks!

[road-runner]

Someone please fill the gpu servers up or give them a kick start or something, this has been going on a couple days on and off that I have seen here with multiple GPUs...

[shdbcamping]

bruce wrote
I hope that the file attribute on the servers will be modified to fix this "problem" but I believe that when you get the response that FF wants to download a file, you can assume it is the OK message. If the server cannot be contacted, FF is much clearer than IE.

Thanks for the clarification on that. I can only imagine that this is a modification to FF version 3, as I'm fairly sure FF version 2 displayed the "OK" without the additional file download request.

Either that, or old age is finally catching up with me.

Neat to know info . But it does not get the work flowing by knowing that it can't . Pande needs to put something about this issue in the Announcements Section (with follow-ups) so we don't keep getting posts flooding on this KNOWN issue

[MichaelO]

Its not assigning work again!

Code: Select all

[19:11:15] + Attempting to get work packet
[19:11:15] - Connecting to assignment server
[19:11:16] - Successful: assigned to (171.64.65.106).
[19:11:16] + News From Folding@Home: Welcome to Folding@Home
[19:11:16] Loaded queue successfully.
[19:11:16] + Could not connect to Work Server
[19:11:16] - Attempt #7  to get work failed, and no other work to do.
Waiting before retry.

[Teddy]

Still getting assigned to 106 & it wont give out work...

[DocJonz]

Tried restarting a few of the clients - still getting assigned to 171.64.65.106 - have six GPU's currently in lmbo ....

[bruce]

Tried restarting a few of the clients - still getting assigned to 171.64.65.106 - have six GPU's currently in lmbo ....

There are two servers that have NVidia work and both are quite busy. I suppose that has something to do with the backlog of clients that are all trying to get work at the same time. If that's true, your best approach is to just let it keep trying.

[patonb]

All good again since yesterday afternoon (may 21)

[shdbcamping]

Down again. I'd post the servers, but i have 8 clients hanging on work. Pande by now should be aware of the problem.

@Bruce,
Can we get an ETA from someone with authrity at Pande on how long the "Backlog" will take to get cleared up? My GPU's drink up energy even at idle you know. All GPU's do. Some folks have a pretty large number of them and it's expensive to idle. All I'd like to know is an estimate on when we can expect a different explanation as the problem seems cyclical but not constant.

[MichaelO]

My Nvidia clients have been without work for almost 2 hours now, but my ATI client just completed and got a new WU. So it would appear the issue is with the Nvidia clients. Is any progress being made to correct this??