There is now plenty of space on the server so no one should be having upload/download problems. Of course, please let us know if any of you still has problems. We still have work to do to ensure this doesn't happen again, but that's for us to worry about.

Thanx to all the hard work you all did to sort it out !!!

Well - it's Monday afternoon in Spain and still no reaction. Clearly no-one is minding the shop...

Yeah I am out of work and have units to send back, I set my pc's to no new tasks and attached to a different project.

There no reason to set No New Tasks. Just let BOINC's transmit-retry-back-off do its job until GPUGrid can complete the transfers, and make sure you are attached to backup-GPU projects.

I did that to avoid problems with the project that IS sending me work and its own deadlines. I don't like units hanging out in my cache just sitting there so keep it pretty short, about a day total. I have now finished the 'other' projects units and am back here again. I KNOW...I am a micro-manager, but it is who I am.

Uploads slowed to a crawl a few hours ago and are getting slower and slower. Other projects are uploading normally. Checked my connection with Ookla Speedtest and the speed is normal. WU uploads are even starting to stall and lose connection:

2422	GPUGRID	10-07-14  22:39	Started upload of 2x2162-NOELIA_5MG-2-3-RND6808_0_9	
2423	GPUGRID	10-07-14  22:39	[error] Error reported by file upload server: [2x2162-NOELIA_5MG-2-3-RND6808_0_9] locked by file_upload_handler PID=22012	
2424	GPUGRID	10-07-14  22:39	Temporarily failed upload of 2x2162-NOELIA_5MG-2-3-RND6808_0_9: transient upload error	
2425	GPUGRID	10-07-14  22:39	Backing off 00:02:50 on upload of 2x2162-NOELIA_5MG-2-3-RND6808_0_9	
2426	GPUGRID	10-07-14  22:39	Started upload of 2x2162-NOELIA_5MG-2-3-RND6808_0_9	
2427	GPUGRID	10-07-14  22:40	Temporarily failed upload of 2x2162-NOELIA_5MG-2-3-RND6808_0_9: transient HTTP error	
2428	GPUGRID	10-07-14  22:40	Backing off 00:06:50 on upload of 2x2162-NOELIA_5MG-2-3-RND6808_0_9	
2429			10-07-14  22:40	Project communication failed: attempting access to reference site	
2430			10-07-14  22:40	Internet access OK - project servers may be temporarily down.	

Things seem to be normal again. Thanks!

Yo Guys/Gals. Is there a problem with the server again?
Uploads are maxing at 8 KBps and then stalling.

my log show this error
20/11/2014 6:40:32 PM | GPUGRID | Temporarily failed upload of 20mgx465-NOELIA_20MG2-25-50-RND2603_0_9: transient HTTP error

No problems with the internet from this machine, and WCG tasks have uploaded at 50-80 KBps while GPUGrid has been stuck.

I have 2 tasks struggling to upload at the moment.

HELP!!!

Been like this for the past hour or so.

1/6/2015 9:45:10 PM | GPUGRID | Backing off 05:36:00 on upload of 1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9
1/6/2015 9:46:42 PM | GPUGRID | Started upload of 1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9
1/6/2015 9:46:44 PM | GPUGRID | [error] Error reported by file upload server: [1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9] locked by file_upload_handler PID=17978
1/6/2015 9:46:44 PM | GPUGRID | Temporarily failed upload of 1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9: transient upload error
1/6/2015 9:46:44 PM | GPUGRID | Backing off 04:24:32 on upload of 1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9
1/6/2015 9:58:25 PM | GPUGRID | Started upload of 1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9
1/6/2015 9:58:27 PM | GPUGRID | [error] Error reported by file upload server: [1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9] locked by file_upload_handler PID=17978
1/6/2015 9:58:27 PM | GPUGRID | Temporarily failed upload of 1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9: transient upload error
1/6/2015 9:58:27 PM | GPUGRID | Backing off 04:27:53 on upload of 1x81-GERARD_CXCL12_LIG10-4-5-RND5536_0_9

Sounds similar to this communication/server problem described by mundayweb,
(03. Error code -121 to -130 explained).

ERR_UPLOAD_TRANSIENT -127

First an explanation what transient means. Transient refers to a module that, once loaded into main memory, is expected to remain in memory for a short time.

This is a server error.
The file you are trying to upload is locked on the server. The file_upload_handler put an advisory lock on the file, to prevent other file upload handlers to write to the file.

This can only be fixed by the project.

Extra messages in 5.8 branch of BOINC.
can't open file - Advisory file locking is not guaranteed reliable when used with stream buffered IO.
can't lock file - File Upload Handler can't put an advisory lock on the file to prevent it being overwritten by other FUHs.
Maintenance underway: file uploads are temporarily disabled. - You can't upload as the server is down for maintenance.

http://boincfaq.mundayweb.com/index.php?viewCat=3

I can upload and report work, as can others.
Is this specific to one task and one system?

---

FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

It was and it cleared about 2 hours later.

It appears this problem has resurfaced. Seems the AMD uploads are processing fine, but the Nvidea ones are not:

This so far is on two different systems

2/7/2015 10:23:11 PM | GPUGRID | Started upload of e2s134_e1s85f79-GERARD_BENTRYP_GAAMPCGEN2-0-1-RND0868_0_9
2/7/2015 10:23:12 PM | GPUGRID | [error] Error reported by file upload server: [e2s134_e1s85f79-GERARD_BENTRYP_GAAMPCGEN2-0-1-RND0868_0_9] locked by file_upload_handler PID=27570
2/7/2015 10:23:12 PM | GPUGRID | Temporarily failed upload of e2s134_e1s85f79-GERARD_BENTRYP_GAAMPCGEN2-0-1-RND0868_0_9: transient upload error
2/7/2015 10:23:12 PM | GPUGRID | Backing off 00:21:54 on upload of e2s134_e1s85f79-GERARD_BENTRYP_GAAMPCGEN2-0-1-RND0868_0_9


When I've seen the transient upload error on other projects it seems it is a case of a lack of disk space..

It was and it cleared about 2 hours later.

When I've seen the transient upload error on other projects it seems it is a case of a lack of disk space.

When the problem is a lack of disk space, it says so in the event log message. We saw both of these at LHC recently:

Server is out of disk space
can't write file ... No space left on server

There are plenty of other failure modes which aren't reported explicitly, but the cause can be discovered by enabling <http_debug> logging for the duration.

In your case, the single file you wanted to upload was "locked by file_upload_handler PID=27570" - probably a comms glitch hadn't cleared properly. As you found, the process times out after a while, the lock is released, and the upload can proceed normally. It was an individual problem with that one single file, not an indication of server trouble in general.

Richard -- thanks for the reply -- the problem uploads on both computers did clear last night.

I did find it curious that I encountered with two different computers at around the same time.

Upload issue.

3/18/2015 8:12:17 PM | GPUGRID | Started upload of e10s60_e8s18f109-NOELIA_27x3-1-2-RND6719_0_0
3/18/2015 8:12:19 PM | GPUGRID | Temporarily failed upload of e10s60_e8s18f109-NOELIA_27x3-1-2-RND6719_0_0: transient HTTP error
3/18/2015 8:12:19 PM | GPUGRID | Backing off 04:37:43 on upload of e10s60_e8s18f109-NOELIA_27x3-1-2-RND6719_0_0
3/18/2015 8:12:22 PM | | Project communication failed: attempting access to reference site
3/18/2015 8:12:23 PM | | Internet access OK - project servers may be temporarily down.

Just Me?

I got that on a couple of systems -- not sure. For me to resolve it I went to the newest version of BOINC -- for some reason that resolved it on both systems.

THANX Barry. That did work. Not sure why but was time to upgrade anyway.

Had the same problem and when I upgraded from 7.0.28 to 7.4.36 everything worked again.

When I was TS did a project update and noticed in the log something about an error in the certif. maybe related to update to HTTPS ?

I'm also having upload problems, am on the latest boinc, the lagest file gets to various percentages and then stops.

rebooted pc and managed to upload the large file but still got problems with the small ones.

19/03/2015 07:12:04 | GPUGRID | Temporarily failed upload of e16s55_e1s204f48-NOELIA_1mgx1-1-4-RND3607_0_1: transient HTTP error

Yea same here the log message I recieved says...

29741 GPUGRID 19/03/2015 6:30:15 PM Scheduler request failed: Peer certificate cannot be authenticated with given CA certificates

are you running windows or Linux jugnut, sound like you have a possible problem with your ca-bundle.crt certificate. are you running the latest version?