Bug 855703

Summary:	job logs should be stored on the lab controller, even if no archive server is configured (remove CACHE setting, remove support for sending logs to server)
Product:	[Retired] Beaker	Reporter:	Raymond Mancy <rmancy>
Component:	lab controller	Assignee:	Dan Callaghan <dcallagh>
Status:	CLOSED CURRENTRELEASE	QA Contact:	Amit Saha <asaha>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	0.9	CC:	asaha, bpeck, dcallagh, ebaak, kbaker, llim, rmancy
Target Milestone:	0.12
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:	LogStorage
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2013-04-11 04:57:29 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 Raymond Mancy 2012-09-10 04:26:29 UTC

In labcontroller.conf we specify the following;

  # Set CACHE = True only if you define ARCHIVE_* variables correctly!

I was wondering if there's any reason why we have to tie storing
logs on the lab controller (as opposoed to the server) to the running of beaker transfer?

In the most general sense, is storing the files on the lab controller not a legitimate scenario ?

Comment 1 Dan Callaghan 2012-09-10 06:59:58 UTC

(In reply to comment #0)
> In the most general sense, is storing the files on the lab controller not a
> legitimate scenario ?

This makes sense to me. If we removed support for sending logs to the Beaker server (meaning, you either leave them permanently on the lab controller, or enable beaker-transfer to ship them elsewhere) that would eliminate some code paths and help simplify the log uploading code.

Comment 2 Bill Peck 2012-09-10 14:40:49 UTC

(In reply to comment #1)
> (In reply to comment #0)
> > In the most general sense, is storing the files on the lab controller not a
> > legitimate scenario ?
> 
> This makes sense to me. If we removed support for sending logs to the Beaker
> server (meaning, you either leave them permanently on the lab controller, or
> enable beaker-transfer to ship them elsewhere) that would eliminate some
> code paths and help simplify the log uploading code.

Agree, and for most users who run the scheduler and lab controller on the same machine it still wouldn't change anything either, just makes the code simpler as you mention.

Comment 3 Nick Coghlan 2012-10-17 04:40:13 UTC

Bulk reassignment of issues as Bill has moved to another team.

Comment 4 Dan Callaghan 2013-02-25 05:49:03 UTC

I wrote a patch for this while working on the harness API:

http://gerrit.beaker-project.org/1701

Comment 8 Dan Callaghan 2013-04-11 04:57:29 UTC

Beaker 0.12 has been released.