Bug 973893

Summary: docs do not have a general (non-version-specific) upgrade guide
Product: [Retired] Beaker Reporter: Dan Callaghan <dcallagh>
Component: DocAssignee: Dan Callaghan <dcallagh>
Status: CLOSED CURRENTRELEASE QA Contact: wangjing <jingwang>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 0.12CC: asaha, dcallagh, jingwang, llim, qwan, rglasz, rmancy, xjia
Target Milestone: 0.13.xKeywords: Documentation
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-07-11 02:44:57 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:
Embargoed:

Description Dan Callaghan 2013-06-13 05:00:59 UTC
The Administration Guide explains how to do a fresh Beaker installation but there is no mention of how generally to upgrade from one version to another (excluding the version-specific release notes).

It might just be a matter of adding a new section "Upgrading" after "Installing" explaining the steps:

    yum upgrade ...
    service httpd reload
    beaker-repo-update ...

Comment 1 Dan Callaghan 2013-06-21 07:08:17 UTC
On Gerrit: http://gerrit.beaker-project.org/2052

Comment 3 Nick Coghlan 2013-06-26 00:58:34 UTC
Given the misbehaviour noted in bug 977949, it's probably worth making the steps for an upgrade that involves database changes (and hence a full service outage) even more explicit:

1. Shut down Beaker daemons on the lab controllers
2. Shut down Apache and the scheduler daemon on the main server
3. Do the database and RPM updates
4. Restart Apache and the scheduler daemon on the main server
5. Restart the Beaker daemons on the lab controllers

We may also want to explicitly note that such an outage *may* cause currently running jobs to fail. Jobs still running during the outage may end up reporting failures attempting to register interim results, upload log files, etc. If a recipe terminates completely during the outage window, this will be detected as an external watchdog timeout after the system is restored to operation.

Comment 4 Dan Callaghan 2013-06-26 03:17:55 UTC
On Gerrit: http://gerrit.beaker-project.org/2060

Comment 5 Nick Coghlan 2013-06-27 07:38:52 UTC
This is now available as a new section "Upgrading an existing installation" in the Beaker Administrator's Guide:

http://beaker-project.org/docs-release-0.13/admin-guide/
http://beaker-project.org/docs-release-0.13/admin-guide/upgrading.html

Comment 6 xjia 2013-06-28 03:19:48 UTC
Verify:
It makes sense.

Comment 7 Amit Saha 2013-07-11 02:44:57 UTC
Beaker 0.13.2 has been released. (http://beaker-project.org/docs/whats-new/release-0.13.html#beaker-0-13-2).