Down Load Failures
11 Jul 2010 1:12:15 UTC
Topic 195195
(moderation:
Just experienced the following:
7/11/2010 9:06:58 AM Einstein@Home Requesting new tasks for CPU
7/11/2010 9:07:05 AM Einstein@Home Scheduler request completed: got 1 new tasks
7/11/2010 9:07:07 AM Einstein@Home Started download of skygrid_1040Hz_S5GC1.dat
7/11/2010 9:07:07 AM Einstein@Home Started download of l1_1034.65_S5R7
7/11/2010 9:07:09 AM Einstein@Home Incomplete read of 43.000000 < 5KB for l1_1034.65_S5R7 - truncating
7/11/2010 9:07:09 AM Einstein@Home Finished download of l1_1034.65_S5R7
7/11/2010 9:07:09 AM Einstein@Home [error] File l1_1034.65_S5R7 has wrong size: expected 3345408, got 0
7/11/2010 9:07:09 AM Einstein@Home [error] Checksum or signature error for l1_1034.65_S5R7
7/11/2010 9:07:26 AM Einstein@Home Finished download of skygrid_1040Hz_S5GC1.dat
(Download Failed)
Is this a "one of" experience or indicative of a larger problem?
Best regards,
Stacey
Down Load Failures
)
I think it's a temporary download failure.
Meanwhile your BOINC client should have retried the download, hopefully with more success. :-)
Gruß,
Gundolf
Computer sind nicht alles im Leben. (Kleiner Scherz)
Continuing download
)
Continuing download failures
7/13/2010 4:31:19 PM Einstein@Home Sending scheduler request: To fetch work.
7/13/2010 4:31:19 PM Einstein@Home Reporting 1 completed tasks, requesting new tasks for CPU
7/13/2010 4:31:26 PM Einstein@Home Scheduler request completed: got 1 new tasks
7/13/2010 4:31:28 PM Einstein@Home Started download of l1_1034.65_S5R7
7/13/2010 4:31:29 PM Einstein@Home Incomplete read of 43.000000 < 5KB for l1_1034.65_S5R7 - truncating
7/13/2010 4:31:29 PM Einstein@Home Finished download of l1_1034.65_S5R7
7/13/2010 4:31:29 PM Einstein@Home [error] File l1_1034.65_S5R7 has wrong size: expected 3345408, got 0
7/13/2010 4:31:29 PM Einstein@Home [error] Checksum or signature error for l1_1034.65_S5R7
7/13/2010 4:32:37 PM Einstein@Home Sending scheduler request: To fetch work.
7/13/2010 4:32:37 PM Einstein@Home Reporting 1 completed tasks, requesting new tasks for CPU
7/13/2010 4:32:42 PM Einstein@Home Scheduler request completed: got 0 new tasks
7/13/2010 4:32:42 PM Einstein@Home Message from server: No work sent
7/13/2010 4:32:42 PM Einstein@Home Message from server: No work is available for Global Correlations S5 search #1
7/13/2010 4:32:42 PM Einstein@Home Message from server: No work is available for Global Correlations S5 Engineering
7/13/2010 4:32:42 PM Einstein@Home Message from server: No work is available for Hierarchical S5 all-sky GW search #6
7/13/2010 4:32:42 PM Einstein@Home Message from server: No work available for the applications you have selected. Please check your preferences on the web site.
7/13/2010 4:33:43 PM Einstein@Home Sending scheduler request: To fetch work.
7/13/2010 4:33:43 PM Einstein@Home Requesting new tasks for CPU
7/13/2010 4:33:50 PM Einstein@Home Scheduler request completed: got 2 new tasks
7/13/2010 4:33:53 PM Einstein@Home Started download of skygrid_1040Hz_S5GC1.dat
7/13/2010 4:33:53 PM Einstein@Home Started download of l1_1034.65_S5R7
7/13/2010 4:33:54 PM Einstein@Home Incomplete read of 43.000000 < 5KB for l1_1034.65_S5R7 - truncating
7/13/2010 4:33:54 PM Einstein@Home Finished download of l1_1034.65_S5R7
7/13/2010 4:33:54 PM Einstein@Home [error] File l1_1034.65_S5R7 has wrong size: expected 3345408, got 0
7/13/2010 4:33:54 PM Einstein@Home [error] Checksum or signature error for l1_1034.65_S5R7
7/13/2010 4:34:14 PM Einstein@Home Finished download of skygrid_1040Hz_S5GC1.dat
Stacey Baird
I have forgotten how to
)
I have forgotten how to determine the directory name for the download URL, so I can't check if I could download that file manually.
Are you using a cc_config.xml file? By setting the option, you could determine the URL yourself.
Gruß,
Gundolf
Computer sind nicht alles im Leben. (Kleiner Scherz)
RE: I have forgotten how to
)
In the OP's various snippets, each attempt at downloading the large data file l1_1034.65_S5R7 seems to be failing for some reason. The easiest way to see if this is a problem with a particular mirror from which the download attempt is being made is to try different mirrors.
The state file (client_state.xml) contains entries for every large data file that needs to be present on the host. So just open the state file with a simple text editor like notepad and search for the block for the filename l1_1034.65_S5R7. That name might exist as part of different entries throughout the state file so use the string l1_1034.65_S5R7 to get to the correct block. Here is an example (obviously a completely different file) of a complete entry from one of my machines.
l1_1139.95_S5R7 3345408.000000 0.000000 0b91aaac69a763b46908e32f0ab35706 1 http://einstein.ligo.caltech.edu/download/279/l1_1139.95_S5R7 http://einstein-dl2.phys.uwm.edu/download/279/l1_1139.95_S5R7 http://einstein-dl.phys.uwm.edu/download/279/l1_1139.95_S5R7 http://einstein-dl3.phys.uwm.edu/download/279/l1_1139.95_S5R7 http://einstein-dl4.phys.uwm.edu/download/279/l1_1139.95_S5R7 http://einstein.aei.mpg.de/download/279/l1_1139.95_S5R7 http://einstein.ligo.caltech.edu/download/279/l1_1139.95_S5R7 http://einstein-dl2.phys.uwm.edu/download/279/l1_1139.95_S5R7 http://einstein-dl.phys.uwm.edu/download/279/l1_1139.95_S5R7 http://einstein-dl3.phys.uwm.edu/download/279/l1_1139.95_S5R7 http://einstein-dl4.phys.uwm.edu/download/279/l1_1139.95_S5R7 http://einstein.aei.mpg.de/download/279/l1_1139.95_S5R7If you ignore the harmless bug where the block of 6 URLs has been inserted in the state file twice, you have the actual location of the file on the six different mirrors from where you could attempt to download the file manually.
So the OP could extract his own set of URLs and try different ones to see if the file can be retrieved. If successful (at least check the no. of bytes downloaded () as a bare minimum - preferably also the MD5 checksum using a freeware MD5 checksum generator) then simply place the file in the E@H project directory, overwriting any bad file that is already there.
It would be interesting if the OP could try all six mirrors to see if one is indeed bad. If so that information needs to go to the Devs - reporting the outcome here would be appreciated, thank you.
Cheers,
Gary.