I fear we are going to collapse because of out of diskspace...

---

No worries.
Eric made a cleanup not that long ago.

---

With each crime and every kindness we birth our future.

Eric turned off file deletion so when we do get invalid results, we can still download the Wu to run it offline and find out why it was invalid,

Claggy

I fear we are going to collapse because of out of diskspace...

The amount of diskspace this project uses is very small relative to the main project, and the WU or Task pages only reflect BOINC database records rather than actual WU or uploaded Result files. Eric has always tried to delay purging the BOINC database records while the data is still pertinent to whatever is being Beta tested.

                                                                   Joe

I think we may have a problem with deleted files, after all.

I just had a run of downloading errors: Error SETI@home v7 tasks for computer 12316. These are all resends for workunits in the range 3788475 to 3842403, created 19 Jan 2012.

I can't get the input files manually either, and it's led to a lot of bad karma in the BOINC v7.0.28 message log:

22/06/2012 10:07:10 | SETI@home Beta Test | Started download of 05oc11al.25211.14419.7.14.189
22/06/2012 10:07:14 | SETI@home Beta Test | Giving up on download of 05oc11al.25211.14419.7.14.189: permanent HTTP error
22/06/2012 10:07:18 | SETI@home Beta Test | [sched_op] Deferring communication for 3 min 12 sec
22/06/2012 10:07:18 | SETI@home Beta Test | [sched_op] Reason: Unrecoverable error for task 05oc11al.25211.14419.7.14.189_2 (WU download error: couldn't get input files:<file_xfer_error> <file_name>05oc11al.25211.14419.7.14.189</file_name> <error_code>-224</error_code> <error_message>permanent HTTP error</error_message></file_xfer_error>)
22/06/2012 10:10:42 | SETI@home Beta Test | Message from server: Your app_info.xml file doesn't have a usable version of SETI@home v7.
22/06/2012 10:10:42 | SETI@home Beta Test | This computer has finished a daily quota of 29 tasks

Oh yes my app_info does have a usable version - what does it think crunched the pending and valid tasks in the same run?

I think we may have a problem with deleted files, after all.

I just had a run of downloading errors: Error SETI@home v7 tasks for computer 12316. These are all resends for workunits in the range 3788475 to 3842403, created 19 Jan 2012.
...

I just had a run of those on 10490 too, with creation dates of both 19 and 20 January. But curiously in the middle of those I also got one which downloaded OK for WU 3849146. I don't see any obvious clues why that one worked, it's sort of in the middle of the ones which failed.

Well, having the basic quota at 33 at least limits the number each host will "error" on. Like the -221 server aborts, this -224 error caused by the server's HTTP 404 really ought not be counted as a client error, though Dr. Anderson is so set against liberalizing anything a user might use for cherry picking that it surely won't change.

                                                                  Joe

06/23/12 12:12:26 | SETI@home Beta Test | Started download of 05oc11al.6268.14010.3.14.103
06/23/12 12:12:30 | SETI@home Beta Test | Giving up on download of 05oc11al.6268.14010.3.14.103: permanent HTTP error
06/23/12 12:12:31 | SETI@home Beta Test | Started download of 05oc11al.6268.13192.3.14.153
06/23/12 12:12:38 | SETI@home Beta Test | Giving up on download of 05oc11al.6268.13192.3.14.153: permanent HTTP error
...
06/23/12 12:47:56 | SETI@home Beta Test | This computer has finished a daily quota of 1 tasks

Also got these ugly "permanent errors" on my host, hundreds! The effect the quota message shows.
This made me check my network hardware first, before i concluded that the problem was not located on my side.

Luckily i have enough "good" workunits on that host, so the quota will be no problem.

@Josef W. Segur:

...ought not be counted as a client error, though Dr. Anderson is so set against liberalizing anything a user might use for cherry picking that it surely won't change.

Thanks for adding the reason why this approach has been choosen.

---

_\|/_
U r s

I've had my fair share of these in the last few days on my rig 43303, currently six in total. These are resends for work units between 3771876 and 3787699, all created on 19 January 2012.

I also have several in my pending list that have been resent to others, with the same -224 error:

Created on 19 January 2012:
3780589, 3780605, 3846780

Created on 20 January 2012:
3847692, 3853225

I've got a couple more from 20 January 2012 that haven't been resent yet:
3860624, 3860635

Can this be fixed?

---

I thought I would also report the following in case it might help in Beta testing:

as Joe so nicely explained in Message 43012

.....

There's no good evidence why those WU files cannot be found, but my guess is they went missing when the WU storage was migrated to a new server.

As noted in the No deletion of old files for a while thread, the issue appears to only be on tasks which were split in January. That batch of reissues seems to have all been sent, the current S@H v7 tasks being assigned are original replication tasks created March 22 which are OK.

Joe

I'm using: BOINC 7.0.28 Windows XP

computer #1 my computer was using: v7 v6.97 and marked as: valid

computer #2 was using: v7 v6.97 and marked as: invalid

computer #3 was using: Anonymous platform and marked as: valid ????????

is that correct ??

http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=3851336

• Created --- 30 Mar 2012 | 0:32:04 UTC
  
• Sent ------- 22 Jul 2012 | 8:46:30 UTC
  
• Received -- 22 Jul 2012 | 20:44:14 UTC
  

the following wuid's were Created Jan 2012 or before:

WU download error: couldn't get input files:

http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=3710455

http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=3846907

http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=3802561