Just started getting this message:

13/08/2009 11:37:42 GPUGRID Started upload of 164-GIANNI_BINDX119-41-100-RND1295_2_4
13/08/2009 11:37:43 GPUGRID [error] Error reported by file upload server: Server is out of disk space
13/08/2009 11:37:43 GPUGRID Temporarily failed upload of 164-GIANNI_BINDX119-41-100-RND1295_2_4: transient upload error
13/08/2009 11:37:43 GPUGRID Backing off 2 min 3 sec on upload of 164-GIANNI_BINDX119-41-100-RND1295_2_4
13/08/2009 11:37:45 GPUGRID Started upload of 164-GIANNI_BINDX119-41-100-RND1295_2_1
13/08/2009 11:37:46 GPUGRID [error] Error reported by file upload server: Server is out of disk space
13/08/2009 11:37:46 GPUGRID Temporarily failed upload of 164-GIANNI_BINDX119-41-100-RND1295_2_1: transient upload error
13/08/2009 11:37:46 GPUGRID Backing off 29 min 34 sec on upload of 164-GIANNI_BINDX119-41-100-RND1295_2_1

Regards
Zy

Other symptoms - Both files will upload to 100% as reported by the progress bar in the transfer tab, but does not finish off, going to transient error. I saw on the server page that the db_purge process has stopped.

Regards
Zy

yeah... I am getting a repeated "no space left on server" error message.. which is a bummer, because it wont allow another WU to download until that one uploads..

I have a second WU running, but I hope this one upload before THAT one finishes..or my GPU will get really cold.....
____________
64 Ghz of home grown cruncherism...running 24/7

Proud member of BOINC@Australia

yeah... I am getting a repeated "no space left on server" error message.. which is a bummer, because it wont allow another WU to download until that one uploads..

I have a second WU running, but I hope this one upload before THAT one finishes..or my GPU will get really cold.....
____________
64 Ghz of home grown cruncherism...running 24/7

Proud member of BOINC@Australia

I am seeing upload size of 12mb (megabytes). That seems unusually large. Is that the normal size for gpugrid uploads?



I was unaware the upload sizes were that large. My average upload speed is about 32kb and I am paying for only 128kb download on dsl.

Yes, that's the norm for many WUs.

Got very simular messages:

13/08/2009 20:30:26|GPUGRID|Temporarily failed upload of 68-GIANNI_CLOZ-0-50-RND0734_1_4: transient upload error
13/08/2009 20:30:26|GPUGRID|Backing off 1 hr 27 min 29 sec on upload of 68-GIANNI_CLOZ-0-50-RND0734_1_4
13/08/2009 20:30:29|GPUGRID|Started upload of 68-GIANNI_CLOZ-0-50-RND0734_1_4
13/08/2009 20:30:30|GPUGRID|[error] Error reported by file upload server: can't open file
13/08/2009 20:30:30|GPUGRID|Temporarily failed upload of 68-GIANNI_CLOZ-0-50-RND0734_1_4: transient upload error
13/08/2009 20:30:30|GPUGRID|Backing off 3 hr 4 min 50 sec on upload of 68-GIANNI_CLOZ-0-50-RND0734_1_4

Really don't know, what to do in such case, the rest of almost processed result files was uploaded automatically.
It would be really a pity, if I can't upload the results file.
Any advice ?
Thanks in advance.

Just leave it to the BOINC client - it will keep track of it and upload when its back in. The Client will slowly increase the time between retrys, so keep your eye on that - it maybe at "retry in 24 hrs (etc)" when the server is back and will need a manual update to get past that.

It will not matter if you need to reboot/close down (whatever) the Client will track it for you, so you will not lose anything.

Regards
Zy

Many thanks, Zydor, for advice, will keep a eye on its behavior.
Regards,
Jet

Looks like things are on the move - server is "all green" at the moment, so will not be long, upload still not working at the moment but that can happen if there is a large queue to sort out.

Regards
Zy

Anyone out there managed to upload anything yet? Servers been all green for 5 hours now .... I get instant retry when pressing "retry now" button in transfer tab, which tends to indicate uploads still disabled ... just checking what the latest is

Regards
Zy

I have been able to upload work, but when I go to download new work (having worked thru my GPUGrid cache) I am getting:

8/13/2009 4:54:07 PM|GPUGRID|Sending scheduler request: To fetch work. Requesting 3224249 seconds of work, reporting 0 completed tasks
8/13/2009 4:54:12 PM|GPUGRID|Scheduler request completed: got 0 new tasks
8/13/2009 4:54:12 PM|GPUGRID|Message from server: No work sent
8/13/2009 4:54:12 PM|GPUGRID|Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer.

I encounter this with both the 186 driver and the newer 190 driver on Windows XP SP3. I have the 9800GT.

It seems with enough retries I sometimes do get a work unit. I was letting my cache drain out before, so as to not have a backlog of two or three work units available. Seems that this approach is not the right way to go at the moment.

It is a bit confusing as the server status indicates there are 100's of work units available.

Anyone out there managed to upload anything yet? Servers been all green for 5 hours now .... I get instant retry when pressing "retry now" button in transfer tab, which tends to indicate uploads still disabled ... just checking what the latest is

Regards
Zy

Okie Doke

Uploads were suspended temporarily (post 11834 above) - it might be they were never re-enabled when the disk issue was solved. No doubt we'll see in the morning.

Regards
Zy

I can upload new work after complete calculation, but I can't upload the older completed WUs from the time, when the server disk was full. I get this message:

14.08.2009 02:52:00 GPUGRID [error] Error reported by file upload server: can't open file
14.08.2009 02:52:00 GPUGRID Backing off 2 hr 12 min 58 sec on upload of 188-GIANNI_CLOZ-0-50-RND8622_1_4

Unfortunely this count also for new work, so I get not enough work for my GTX295. This WUs in upload blocks the download.

14.08.2009 02:55:45 GPUGRID Message from server: No work sent
14.08.2009 02:55:45 GPUGRID Message from server: (reached limit of 4 GPU tasks)

What shall I do? Abort the upload to get new work or wait for a solution on the server site?

____________

Dont abort - leave it to BOINC Client it will look after it. Carry on crunching new models downloaded. When the upload is fixed you might need to "retry" in the transfer Tab of BOINC Manager, but for the moment let them crunch away and stack up in a queue - the Client will look after them. You can still close down/reboot etc in the normal way it will not lose them provided you close BOINC manager properly.

It will resolve itself once they fix the upload problem, hopefully tomorrow - but it will not matter if it takes longer, just crunch away and let them stack up inside BOINC Manager, they will be fine. You will be able to download new WUs in the normal way if you need them.

Regards
Zy

I get the same error message as Kokomiko!

2009/8/14 AM 09:05:58 GPUGRID [error] Error reported by file upload server: can't open file
2009/8/14 AM 09:05:58 GPUGRID Temporarily failed upload of 119-KASHIF_HIVPR_sub_so_ba2-19-100-RND4220_1_4: transient upload error
2009/8/14 AM 09:05:58 GPUGRID Backing off 1 hr 51 min 45 sec on upload of 119-KASHIF_HIVPR_sub_so_ba2-19-100-RND4220_1_4

Any method to solve the problem?

No - its a server side issue, nothing we can do until they solve the earlier Upload problem, just leave them in BOINC Manager, download what you can if new ones needed (may have to increase cache in preferences). You might still hit the upper limit of held WUs, if you do, not a lot you can do until they clear the upload problem

Regards
Zy

I'm getting files uploaded now, but still cant clear the _1 and _2 files from:

http://www.gpugrid.net/result.php?resultid=1099998

This was one that was running/uploading at the time of the Disk Problem, may have got corrupted on the server. When retry is pressed in Transfer tab, they vertually instantly revert to waiting state, and will not upload.

Newly completed WUs appear to upload fine.

Regards
Zy

Just tried it and still can't upload. I have 2 waiting to upload and 1 that will finish in 15 mins. I have reached my limit of 4. Any chance of increasing the limit.

this is what I am seeing;

8/14/2009 3:55:30 AM GPUGRID [error] Error reported by file upload server: can't open file
8/14/2009 3:55:30 AM GPUGRID [error] Error reported by file upload server: can't open file
8/14/2009 3:55:30 AM GPUGRID Backing off 3 hr 16 min 58 sec on upload of 583-GIANNI_BIND001-7-100-RND1578_2_4
8/14/2009 3:55:30 AM GPUGRID Backing off 3 hr 53 min 32 sec on upload of 920-GIANNI_BIND001-7-100-RND4605_2_4

Is the upload problem fixed now?

I have just restarted the web server.

Both files just started uploading. You had me worried for a bit. Thought I was going to run out of work.

Oppps - you fixed it whilst I was typing the above .....

Its working they are going through right now

Many Thanks :)

Regards
Zy

Unfortunately, the problem is not solved yet.
Still have this message in manager:

Temporarily failed upload of 68-GIANNI_CLOZ-0-50-RND0734_1_4: transient upload error

Regards,

Jet

Jet

In the messages screen of BOINC Manager - whats the line above that? It will give a short reason for the failure

Regards
Zy

2009/8/14 PM 04:57:42 GPUGRID Started upload of 119-KASHIF_HIVPR_sub_so_ba2-19-100-RND4220_1_4
2009/8/14 PM 05:01:54 GPUGRID Finished upload of 119-KASHIF_HIVPR_sub_so_ba2-19-100-RND4220_1_4

Upload the cruching data successfully.
Thanks to project administrators.

Ryan

Just looked at manager - this file was uploaded automatically, so, problem was fixed however. Lost + 25% bonus credits for fast processing, but that isn't so important :-), just one WU.

Thanks for the support !
Regards,

Jet

All Wu's uploaded as I slept. :-) Everything is back to normal. Thanks!
____________

Hi all,

i think, the problem still exist?
Now i've two wu's ready-to-upload, but the client can't upload anything ...

17.08.2009 16:27:54 GPUGRID Started upload of 83-GIANNI_BINDX-36-100-RND9183_0_3
17.08.2009 16:28:18 Project communication failed: attempting access to reference site
17.08.2009 16:28:18 GPUGRID Temporarily failed upload of 83-GIANNI_BINDX-36-100-RND9183_0_3: HTTP error
17.08.2009 16:28:18 GPUGRID Backing off 35 min 24 sec on upload of 83-GIANNI_BINDX-36-100-RND9183_0_3
17.08.2009 16:28:19 Internet access OK - project servers may be temporarily down.

The status-page looks very green, so what's going on? Do i something wrong ?-/
(Now the gtx is in idle-mode. I think, i must play some windows-cardgames, to prevent the gtx cooling down to much... ;->)

17.08.2009 16:33:15 GPUGRID Message from server: (reached limit of 2 GPU tasks)

____________

Arrrrgh, Murphy is my best friend ... X-)

There is another problem on my side, it hits the cable-hardware (thx to KabelDeutschland for excessive port-blocking, -filtering, etc. during daytime)
My DNS-cache was wrong.

"ipconfig /flushdns" solve the prob. Sorry. :-/
____________

Perhaps related to a resurfacing of one of the problems over there (disk full or something else).

Just installed a new 9800GT -- 186 driver, 6.4.5 client, and am getting a resurfacing of this message - even though, according to the server status page, there are hundreds of workunits available. This has happened before when other server problems were in the midst of being resolved or diagnosed. I figure to report it to see if other users have seen this today.

8/17/2009 4:26:22 PM|GPUGRID|Sending scheduler request: To fetch work. Requesting 2189748 seconds of work, reporting 0 completed tasks
8/17/2009 4:26:28 PM|GPUGRID|Scheduler request completed: got 0 new tasks
8/17/2009 4:26:28 PM|GPUGRID|Message from server: No work sent
8/17/2009 4:26:28 PM|GPUGRID|Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer.

It seems this is one of those intermittents -- that is, if one is persistent over a period of say an hour looking to get additional work, one will eventually succeed. Since work downloads are (at the moment) a hit and miss sort of thing, the temptation is to build up a work unit cache instead of adding work just as a work unit is completed.

Perhaps related to a resurfacing of one of the problems over there (disk full or something else).

Just installed a new 9800GT -- 186 driver, 6.4.5 client, and am getting a resurfacing of this message - even though, according to the server status page, there are hundreds of workunits available. This has happened before when other server problems were in the midst of being resolved or diagnosed. I figure to report it to see if other users have seen this today.

8/17/2009 4:26:22 PM|GPUGRID|Sending scheduler request: To fetch work. Requesting 2189748 seconds of work, reporting 0 completed tasks
8/17/2009 4:26:28 PM|GPUGRID|Scheduler request completed: got 0 new tasks
8/17/2009 4:26:28 PM|GPUGRID|Message from server: No work sent
8/17/2009 4:26:28 PM|GPUGRID|Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer.