Comment 2 for bug 1060089

Revision history for this message
Kapil Thangavelu (hazmat) wrote : Re: [Bug 1060089] Re: failure during destroy-service

Its not material per se, there is no orderly/graceful shutdown of a unit in
pyjuju, its forcibly terminated by its machine agent atm, and these
tracebacks are possible. Go juju does address this.. this is effectively
juju should execute stop hooks. It would take a dev week to work on this.

On Tue, Oct 23, 2012 at 6:21 AM, Clint Byrum <email address hidden> wrote:

> Seems legit, we should be checking the unit data's return type, or
> perhaps there is a deeper error allowing empty unit data to get there in
> the first place.
>
> ** Changed in: juju
> Status: New => Triaged
>
> ** Changed in: juju
> Importance: Undecided => Medium
>
> --
> You received this bug notification because you are subscribed to juju.
> https://bugs.launchpad.net/bugs/1060089
>
> Title:
> failure during destroy-service
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/juju/+bug/1060089/+subscriptions
>