Bug 1011305 - RFE: Epic - Openshift support for PressGang
Summary: RFE: Epic - Openshift support for PressGang
Keywords:
Status: NEW
Alias: None
Product: PressGang CCMS
Classification: Community
Component: CCMS-Core
Version: 1.1
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
: ---
Assignee: Nobody
QA Contact:
URL:
Whiteboard:
Depends On: 1029231
Blocks:
TreeView+ depends on / blocked
 
Reported: 2013-09-24 04:16 UTC by Misty Stanley-Jones
Modified: 2023-02-21 23:19 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed:
Embargoed:


Attachments (Terms of Use)

Description Misty Stanley-Jones 2013-09-24 04:16:18 UTC
PressGang needs to be completely compatible with and deployable within Openshift. This includes the ability to share information between PressGang instances. It also implies the need for ACLs to protect not-yet-released or embargoed topics and content specs, the ability for PressGang servers / services to authenticate to each other, and a role-based authentication / authorization system (preferrably with SSO).

Obviously this is a large unit of work and will need to be broken up into several sub-tasks.

Comment 1 Lee Newson 2013-09-24 04:25:55 UTC
Just want to note that the free version of OpenShift Online that Red Hat provides, isn't really usable with PressGang because the max storage you can get for a gear is 1GB. Currently our Production DB is about 10-15x that size (not including any backups) just to give an example of how much room you'd need.

(I know you can get larger gears, but I'm just using the free account as an example. The paid still only gives you 6GB which would still be about 2x too small for our prod db)

Comment 2 Stephen Gordon 2013-09-24 14:56:12 UTC
Probably two comments here:

1) Would it be possible at least in terms of "getting stuff working" to use the internal OpenShift instance managed by IT where more gear customization may be available/possible?

2) A key change of thinking needed for moving to cloud computing is to scale out, not up, using elastic resources. That is arguably one of the reasons for the limited (from a traditional view of compute resources) resources of gears even in the paid offering. Rather than trying to continue with the traditional architecture of fitting the entire DB into one instance perhaps it might be time to look at sharding the database out over multiple gears (sharding of a MySQL setup on OpenShift is probably an interesting test case in and of itself)?

Comment 3 Misty Stanley-Jones 2013-09-25 00:55:58 UTC
Thanks for both of your thoughts. We will be fleshing out this requirement over the next little bit, as it is part of a larger requirement. The bug is here as a placeholder for the moment.


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