Bug 917314 - Disk quota
Summary: Disk quota
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: OKD
Classification: Red Hat
Component: Pod
Version: 1.x
Hardware: All
OS: All
unspecified
high
Target Milestone: ---
: ---
Assignee: Anderson Silva
QA Contact: libra bugs
URL:
Whiteboard:
: 917313 (view as bug list)
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2013-03-03 02:26 UTC by Johan Erixon
Modified: 2013-03-15 12:53 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: Enhancement
Doc Text:
Clone Of:
Environment:
Last Closed: 2013-03-15 12:53:40 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
deployment logfile (14.77 KB, application/octet-stream)
2013-03-13 22:23 UTC, Johan Erixon
no flags Details

Description Johan Erixon 2013-03-03 02:26:31 UTC
Description of problem:
When I try to push a change to my Git on OpenShift I get this message:
error: file write error (Disk quota exceeded)
fatal: unable to write sha1 file
error: unpack failed: unpack-objects abnormal exit


If I ssh my application and check quota I get:

Filesystem  blocks   quota   limit   grace   files   quota   limit   grace
/dev/mapper/EBSStore01-user_home01
                1048576*      0 1048576            1055       0   40000


Right now I can not even ssh my application. I get this message (translated from swedish):
bash: can not create temp file for here-document: Disk quota exceeded


The total size of my repository i 1.1 MB so that should not be the problem.

Comment 1 Johan Erixon 2013-03-03 02:34:45 UTC
*** Bug 917313 has been marked as a duplicate of this bug. ***

Comment 2 Johan Erixon 2013-03-03 09:28:47 UTC
When I try to restart my application I get this message:

We appear to be having technical difficulties
...
Reference #d1e1c9144966a0cbb9a1e82e9d471265

Comment 3 Johan Erixon 2013-03-03 20:00:55 UTC
I have now read on the forum that you can run rhc app tidy to clear out tmp files and rotate logs. My problem now is that I can not even connect by ssh because of the quota problem so I can not run tidy. 
It would be nice if it was possible to run tidy from the web and if you get a warrning before the quota get this full.

Comment 4 Anderson Silva 2013-03-06 18:12:47 UTC
Johan,

What's your app url?

AS

Comment 5 Johan Erixon 2013-03-06 22:02:22 UTC
http://parkera-lambdas.rhcloud.com

Comment 6 Anderson Silva 2013-03-06 22:12:45 UTC
Johan,

Are you still getting a problem with your quota? because it looks like your quota has been bumped to 2 GB of storage, but in your zend gear there is close to 1 GB of logs, which is what caused your quota kick. Clean up your log directory, and you should be fine. You should be able to ssh into the directory:

994M	./zend-5.6/var/log


I hope this helps. Let me know.

Thanks,

AS

Comment 7 Johan Erixon 2013-03-07 07:15:41 UTC
I have clean up my logs but I can't seem to start my application

[parkera-lambdas.rhcloud.com ~]\> ctl_all restart
Stopping services
zdd is not running
monitor is not running
httpd (no pid file) not running
jqd is not running
lighttpd is not running
Waiting for stop to finish
Warning: MySQL process exists without a pid file.  Use force-stop to kill.
Done
Starting services
Starting Deployment [FAILED]
[07.03.2013 02:09:56 SYSTEM] watchdog for zdd is running. 
[07.03.2013 02:09:56 SYSTEM] zdd is not running. 
Failed to start zend-5.6

Comment 8 Anderson Silva 2013-03-07 11:24:11 UTC
did you try to force stop it.

Comment 9 Anderson Silva 2013-03-13 12:53:33 UTC
closing ticket. feel free to re-open if the issue still there.

Comment 10 Johan Erixon 2013-03-13 22:23:32 UTC
Created attachment 709823 [details]
deployment logfile

There seem to be some problem with the deployment.
I do not really understand the log-file but it seems like some of the files needed to create a deployment package got damaged when I ran out of disc space.

Comment 11 Anderson Silva 2013-03-13 23:34:37 UTC
have you tried recreating your gear?

AS

Comment 12 Anderson Silva 2013-03-15 12:53:40 UTC
I am closing ticket again. Feel free to hop into #openshift on freenode to follow up with problem.


Note You need to log in before you can comment on or make changes to this bug.