inconsistent logging between different providers

Bug #893379 reported by Mark Mims on 2011-11-21
10
This bug affects 2 people
Affects Status Importance Assigned to Milestone
pyjuju
Low
Unassigned

Bug Description

lxc instances have no /var/lib/juju/units/<unit-name>/charm.log... all charm logging looks like it's going to the unit agent log in /var/log/juju/unit-<unit-name>.log

ec2 instances still have the charm.log separate from the unit-agent log.

Mark Mims (mark-mims) on 2011-11-21
description: updated
Clint Byrum (clint-fewbar) wrote :

I was quite confused by this as well at first. Seems like a by-product of the 1 machine -> multiple units paradigm that only exists in the local provider right now. Maybe if LXC is used everywhere (sans networking?) then the logging will end up consistent?

Changed in juju:
status: New → Confirmed
importance: Undecided → Medium

Excerpts from Clint Byrum's message of Tue Nov 22 00:39:47 UTC 2011:
> I was quite confused by this as well at first. Seems like a by-product
> of the 1 machine -> multiple units paradigm that only exists in the
> local provider right now. Maybe if LXC is used everywhere (sans
> networking?) then the logging will end up consistent?
>
> ** Changed in: juju
> Status: New => Confirmed
>
> ** Changed in: juju
> Importance: Undecided => Medium
>

I'm fine with either, consistency is important, afaics per fhs we should do the
/var/log/juju/unit-name-escaped.log

Changed in juju:
milestone: none → florence
Changed in juju:
milestone: florence → galapagos
Changed in juju:
milestone: galapagos → honolulu
Changed in juju:
milestone: 0.6 → none
Curtis Hovey (sinzui) on 2013-10-12
Changed in juju:
status: Confirmed → Triaged
Curtis Hovey (sinzui) on 2013-10-15
Changed in juju:
importance: Medium → Low
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers