Relationship Graph
View Issue Details
ID | Project | Category | View Status | Date Submitted | Last Update |
---|---|---|---|---|---|
0002372 | SOGo | Backend General | public | 2013-07-20 05:45 | 2013-10-15 12:44 |
Reporter | thomasg | Assigned To | |||
Priority | normal | Severity | minor | Reproducibility | always |
Status | resolved | Resolution | fixed | ||
Product Version | 2.0.6 | ||||
Fixed in Version | 2.1.0 | ||||
Summary | 0002372: /etc/cron.daily/sogo exited with return code 1 with Ubuntu package | ||||
Description | Getting the error "/etc/cron.daily/sogo exited with return code 1" from cron once a day. Anthony Milan posted a fix in 2009 http://permalink.gmane.org/gmane.comp.cms.opengroupware.sogo/3842 Seems that there is a "/" missing in the line SOGOSPOOL=/var/spool/sogo/ | ||||
Tags | No tags attached. | ||||
I see the error too. (debian 7, x64) However: adding the missing trailing "/" doesn't solve it here. It does correct the issue on your server? |
|
Yes getting still the error. The fix didn't correct the error. |
|
What do you get, when you start this script as root on the command line? |
|
root@mail:/etc/cron.daily# ./sogo |
|
Could you change the find command, to not drop error messages? |
|
root@mail:/etc/cron.daily# ./sogo and then interestingly, running it again: |
|
So the source of this error code is indeed the find command. Looks like a race condition to me. I have seen such things on remote file systems. Or are there any other scripts removing these directories? |
|
Nope: they're on a normal, local xfs filesystem, and nothing else (as far as I know) is accessing them. It seems to me it gives the error message, but it succeeds to actually removes the file. (I created an empty test directory, and it disappeared, even though the error message came up) |
|
Sorry for the late response. I can confirm the things heupink wrote. The folder is on a normal ext4 and also nothing else is accessing them. |
|
I have the same problem. Could anyone find a solution for that? I wonder why the issue string is incomplete?! |
|
For reference, the current command: It is broken since find will try to openat() the directory once rmdir returns, yielding the error message that you see. To fix it, we need to tell find to delete the directory on its own: And that cron job scares me, it should check for the owner before deleting anything: Does this work for you guys? |
|
Works for me on CentOS 5. |
|
Worked here too, on debian 7 x64. Thanks! |
|
Shoudl be fixed now: https://github.com/inverse-inc/sogo/commit/dceead3997d00db403e0803f33ba14b9cc061095 |
|
Date Modified | Username | Field | Change |
---|---|---|---|
2013-07-20 05:45 | thomasg | New Issue | |
2013-07-23 07:38 | heupink | Note Added: 0005765 | |
2013-07-30 18:45 | thomasg | Note Added: 0005776 | |
2013-07-31 09:00 | Christian Mack | Note Added: 0005779 | |
2013-07-31 09:09 | heupink | Note Added: 0005780 | |
2013-07-31 11:13 | Christian Mack | Note Added: 0005781 | |
2013-07-31 11:46 | heupink | Note Added: 0005782 | |
2013-07-31 13:22 | Christian Mack | Note Added: 0005783 | |
2013-07-31 14:31 | heupink | Note Added: 0005784 | |
2013-07-31 15:36 | thomasg | Note Added: 0005785 | |
2013-08-30 12:08 | masoodmx | Note Added: 0005934 | |
2013-08-30 15:58 |
|
Note Added: 0005935 | |
2013-09-03 08:40 | Christian Mack | Note Added: 0005936 | |
2013-09-03 09:21 | heupink | Note Added: 0005937 | |
2013-09-03 14:05 |
|
Note Added: 0005938 | |
2013-09-03 14:05 |
|
Assigned To | => jraby |
2013-09-03 14:05 |
|
Status | new => resolved |
2013-09-03 14:05 |
|
Resolution | open => fixed |
2013-10-15 12:43 | francis | Fixed in Version | => 2.1.0 |
2013-10-15 12:44 | francis | Relationship added | has duplicate 0002450 |