Has anybody else had problems with work units not being processed recently?
So far I've chalked up three bad days in a row (two zeroes, and working on another one) even though I've watched big, chunky Tinker work units complete and send several times. The log consistently shows, for each WU, that the client had trouble submitting the unit several times, but then eventually succeeded.
Goose-eggs?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
- Patron of SPCR
- Posts: 2674
- Joined: Sun Jul 04, 2004 6:07 am
- Location: Houten, The Netherlands, Europe
As of version 5 of the client, F@H uses a collection server (CS). If the client cannot sent the finished work unit (WU) back to the work server (WS), then it will be uploaded to the CS. That way the WU will not expire while sitting in the queue on your computer. Your log will look something like below.
Last weekend at least two servers got into trouble. You can check that here. Getting a server back to full service usually takes one or two workdays.
Once you upload a WU to the CS your WU count will increase, but your score won't change. The CS has no means to check whether the WU you uploaded is a valid WU. Your WU will be stored on the CS until the WS comes back online. Once the WS is back online it will test your WU and its points will appear in your score.FAHlog.txt wrote:[02:12:00] Folding@home Core Shutdown: FINISHED_UNIT
[02:12:03] CoreStatus = 64 (100)
[02:12:03] Unit 9 finished with 94 percent of time to deadline remaining.
[02:12:03] Updated performance fraction: 0.945915
[02:12:03] Sending work to server
[02:12:03] + Attempting to send results
[02:12:03] - Reading file work/wuresults_09.dat from core
[02:12:03] (Read 4354576 bytes from disk)
[02:12:05] Couldn't send HTTP request to server (wininet)
[02:12:05] + Could not connect to Work Server (results)
[02:12:05] (171.67.89.154:8080)
[02:12:05] - Error: Could not transmit unit 09 (completed October 15) to work server.
[02:12:05] - 1 failed uploads of this unit.
[02:12:05] Keeping unit 09 in queue.
[02:12:05] Trying to send all finished work units
[02:12:05] + Attempting to send results
[02:12:05] - Reading file work/wuresults_09.dat from core
[02:12:05] (Read 4354576 bytes from disk)
[02:12:06] Couldn't send HTTP request to server (wininet)
[02:12:06] + Could not connect to Work Server (results)
[02:12:06] (171.67.89.154:8080)
[02:12:06] - Error: Could not transmit unit 09 (completed October 15) to work server.
[02:12:06] - 2 failed uploads of this unit.
[02:12:06] + Attempting to send results
[02:12:06] - Reading file work/wuresults_09.dat from core
[02:12:06] (Read 4354576 bytes from disk)
[02:14:04] - Uploaded at ~36 kB/s
[02:14:04] - Averaged speed for that direction ~32 kB/s
[02:14:04] + Results successfully sent
[02:14:04] Thank you for your contribution to Folding@Home.
[02:14:04] + Number of Units Completed: 17
[02:14:04] Successfully sent unit 09 to Collection server.
[02:14:04] + Sent 1 of 1 completed units to the server
Last weekend at least two servers got into trouble. You can check that here. Getting a server back to full service usually takes one or two workdays.
-
- Patron of SPCR
- Posts: 2674
- Joined: Sun Jul 04, 2004 6:07 am
- Location: Houten, The Netherlands, Europe
You're welcome.
Have you seen the graph for the team
Two low point days when there where server troubles. Then the big spike yesterday.
Have you seen the graph for the team
Two low point days when there where server troubles. Then the big spike yesterday.