Uploading linuxSMP WU's

  • Thread starter Thread starter Cob
  • Start date Start date

Cob

Cob

Cob

Soldato
Joined
30 Jul 2006
Posts
18,488
Location
Antrim town
Has anybody elses finished linuxSMP (an maybe the others) WU's stopped uploading with the client not carrying on and downloading a new WU?

Got up this morning to find three clients had stopped at the 'uploading results' stage during the night. No data seems to have been sent.

Tho what worries me is that after stopping and restarting the client it didn't start to upload the finished WU's :/

This is one of them-

Code: 
[23:50:16] Completed 500000 out of 500000 steps  (100 percent)
[23:50:16] Writing final coordinates.
[23:50:18] Past main M.D. loop
[23:50:20] Will end MPI now
[23:51:20] 
[23:51:20] Finished Work Unit:
[23:51:20] - Reading up to 3718704 from "work/wudata_01.arc": Read 3718704
[23:51:20] - Reading up to 1772160 from "work/wudata_01.xtc": Read 1772160
[23:51:21] goefile size: 0
[23:51:21] logfile size: 21467
[23:51:21] Leaving Run
[23:51:22] - Writing 5516731 bytes of core data to disk...
[23:51:22]   ... Done.
[23:51:24] - Shutting down core
[23:51:24] 
[23:51:24] Folding@home Core Shutdown: FINISHED_UNIT
[23:51:32] CoreStatus = 64 (100)
[23:51:32] Sending work to server


[23:51:32] + Attempting to send results


--- Opening Log file [October 9 05:38:40] 


# SMP Client ##################################################################
###############################################################################

                       Folding@Home Client Version 6.00beta1

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: /home/davy/foldingathome/CPU2
Executable: ./fah6
Arguments: -smp 

[05:38:40] - Ask before connecting: No
[05:38:40] - User name: Cob (Team 10)
[05:38:40] - User ID: 683FB8CF6F4E8C1A
[05:38:40] - Machine ID: 4
[05:38:40] 
[05:38:40] Loaded queue successfully.
[05:38:40] - Preparing to get new work unit...
[05:38:40] + Attempting to get work packet
[05:38:40] - Connecting to assignment server


[05:38:40] + Attempting to send results
[05:38:40] - Successful: assigned to (171.64.65.64).
[05:38:41] + News From Folding@Home: Welcome to Folding@Home
[05:38:41] Loaded queue successfully.
[05:38:52] + Closed connections
[05:38:52] 
[05:38:52] + Processing work unit
[05:38:52] Core required: FahCore_a1.exe
[05:38:52] Core found.
[05:38:52] Working on Unit 02 [October 9 05:38:52]
[05:38:52] + Working ...
[05:38:52] 
[05:38:52] *------------------------------*
[05:38:52] Folding@Home Gromacs SMP Core
[05:38:52] Version 1.73 (November 27, 2006)
[05:38:52] 
[05:38:52] Preparing to commence simulation
[05:38:52] - Ensuring status. Please wait.
[05:38:54] - Starting from initial work packet
[05:38:55] 
[05:38:55] Project: 2653 (Run 21, Clone 34, Gen 34)
[05:38:55] 
[05:38:56] Assembly optimizations on if available.
[05:38:56] Entering M.D.
[05:39:14]  percent)
[05:39:18] - Sta
[05:39:18] Project: 2653 (Run 21, Clone 3
[05:39:18] Project: 2653 (Run 21, Clone 34, Gen 34)
[05:39:18] 
[05:39:19] Entering M.D.
[05:39:29] Protein: Protein in POPC
[05:39:31] Writing local files
[05:39:33] Extra SSE boost OK.
 
Having trouble returning results myself although the client has picked up a new wu.

Hope I can get in before the deadline expires.
 
Same here. One client failed to upload the finished WU but did download a new one. The other 2 clients are due to finish in acouple of hours.
 
Well atleast it's not me. The first happened y-day evening and as of 6.30am there's no sign that it was uploaded.
 
Looks like the home rig didn't upload either, got another due around 7pm - if that thread above if correct should be all sorted by tomorrow morning at latest
 
Last edited:
Yip I've come home to find that my 4th client died at 12.30pm too.

I wonder why mine are stopping rather than adding the finished WU to the queue and then downloading another.
 
My C2D rigs arent uploading results but my quads are so it looks like the quad specific WUs are being returned.
When this happened earlier this year not being able to upload meant not being able to download a new WU, the same as you are experiancing Cob, though this time my rigs are able to download and continue working. Im using native Ubuntu installs if you are using VMware it may be the problem.
 
Yip running VMWare, tho I'm not sure that is the cause.

Ok this is the first time the servers have been down with v6 of the client, but I've never had this issue before.

But I'll not worry about it as long as -send x works once the servers are back online.
 
Thanks for the update Cob.
I get a bit concerned when this happens with the SMP Client, not sure how usefull the data would be to Stanford if the WUs were delayed beyond their deadline.
Im sure they would still credit us the points but if they couldnt use the data it would hurt their research, good job theyve got it accepting again.
 
As cob said - my home rig just uploaded a WU (on 9 retry since this morning)
But unlike cob - I only lost a hour of crunching as I'm now 50% through the currrent WU :D

sculptor said:
Im sure they would still credit us the points but if they couldnt use the data it would hurt their research, good job theyve got it accepting again.
Click to expand...

They didn't last time - I guess It doesn't pay to run systems that only 'just' meet the deadlines
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom