Some uploads stuck in limbo
10 Nov 2011 0:20:08 UTC
Topic 196050
(moderation:
I don't usually have any problems with Einstein@Home, but BOINC has been consistently failing to upload several files. Other files will still download and upload just fine, but these six have been failing for days now. I tried enabling some debugging flags in cc_config.xml, and these show the following failure:
Einstein@Home | [file_xfer] URL http://einstein-dl.aei.uni-hannover.de/cgi-bin/file_upload_handler: not found
Should I just cancel these uploads, upload the file somewhere for diagnosis, or is there a real problem on Einstein@Home's end? The files are named as follows:
p2030.20100912.G46.76+00.38.C.b6s0g0.00000_3512_1_6 p2030.20100912.G46.76+00.38.C.b6s0g0.00000_3512_1_7 p2030.20100912.G46.76+00.38.S.b2s0g0.00000_3616_1_6 p2030.20100912.G46.76+00.38.S.b2s0g0.00000_3616_1_7 p2030.20100420.G60.52-00.04.C.b0s0g0.00000_496_1_6 p2030.20100420.G60.52-00.04.C.b0s0g0.00000_496_1_7
Some uploads stuck in limbo
)
Hi
Strange Indeed. I understand the Files are really present in the Boinc project directory for E@H? the error message looks like they could not be found by Boinc.
For now I would consider it a fluke and cancel the upload.
HB
RE: I understand the Files
)
Good question! No, they don't seem to exist in C:\ProgramData\BOINC\projects\einstein.phys.uwm.edu\. The following files do exist, but maybe they're being kept alive by the attempted upload? All my current WUs are in the G49 range.
This looks possibly like a
)
This looks possibly like a problem with BOINC client v6.13.xx
I've just reported a similar problem from another project to the boinc_alpha buglist. In my case, I was running v6.13.10 when the problem started, and it persisted with v6.13.12: I was away for the weekend, so I skipped the v6.13.11 that Ver Greeneyes is running.
My file seems to have been uploaded successfully to the project's server, and deleted from the client data directory at that point (as it should have been). But the BOINC client didn't register the successful upload, and continued to attempt to re-upload the now non-existent file.
Interesting! In that case, I
)
Interesting! In that case, I guess I'll abort these uploads. They're clearly not going to succeed, and all the information is already here. I haven't seen it happen since these 6 and I'm running 6.13.12 now, so the problem was probably fixed at some point.
I seem to be having problems
)
I seem to be having problems uploading my latest files too, it's been trying for a couple of days now and the deadline is later on today. It looks like the server is not responding? ...
etc.
This was a 13 hour chunk of work too, not just the usual 6 or 7 hour ones I usually get. I'll be quite irritated if I've been frying my cpu for the last week to get this through and it won't accept it! :-(
Thanks, whatever someone did
)
Thanks, whatever someone did - 3 hours to go and suddenly it uploaded everything and reported the task completed ok! :-)
Cheers,
Gary.
I have the same problem with
)
I have the same problem with uploads of E@H results too. It has been taking hours for uploading tasks of Binary Radio Pulsar Search and Gravitational Wave GS Search, but it failed, while uploading processes of other projects (S@H, Roseta etc.) pass normally.
Hi same here with Boinc
)
Hi same here with Boinc Version 6.12.34
I have this also with seti since a week.
Greets
Andy
Thanks. Everything have
)
Thanks. Everything have already been uploaded.
RE: This looks possibly
)
Thanks for you help in the other thread, I didn't saw this one, sorry for it.
As I run 6.12.34 under ubuntu, it's not a problem of the new, experimental, client, but of the production one (as you mentioned in the boinc_dev-list already).
In that thread is a imho quite good explanation on how to deal with this, if it happens to anyone else.
Grüße vom Sänger
