Bug 1319517

Summary:	Openshift auto-logouts user if command takes too long/RAM/CPU
Product:	OpenShift Online	Reporter:	lucas0033 <lucas0033>
Component:	oc	Assignee:	Abhishek Gupta <abhgupta>
Status:	CLOSED NOTABUG	QA Contact:	Wei Sun <wsun>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	2.x	CC:	aos-bugs, jokerman, mmccomas, vdinh
Target Milestone:	---
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2016-04-12 17:25:26 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:

Description lucas0033@gmail.com 2016-03-20 20:35:43 UTC

Description of problem:
If I run a command on OpenShift Online that takes too long or requires too much RAM/CPU, OpenShift Online disconnects me.

I noticed it for example with package django-geoip:
https://django-geoip.readthedocs.org/en/latest/installation.html
on command `python manage.py geoip_update`.

Or if my database (Django app) had 470MB in session data, command did not complete:

$ ./manage.py clearsessions
[app-lucas03.rhcloud.com bet]\> timed out waiting for input: auto-logout
Connection to app-lucas03.rhcloud.com closed.

Version-Release number of selected component (if applicable): OpenShift Online

Actual results: Ssh disconnects


Expected results: Command finishes successfully, even if it takes long time.

Comment 1 Vu Dinh 2016-03-22 17:52:17 UTC

Hi Lucas,

The timeout and out-of-memory ssh connection closed is rather expected behavior unfortunately. For the timeout issue, you can try to get around:

unset TMOUT
export TMOUT=2880000

Remember that you will do those commands every time you log in.

For the memory constrain, unfortunately, there is no way (as I know of) to work around. The RAM for small gear is rather limited (512MB). I highly recommend you to upgrade to medium gear at least to avoid that issue. This memory constrain also happens if you install some big packages as well. The medium or larger gears are better fit for those.

I hope this helps.

Thanks,
Viu