-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviewboard-fedoraserver.rhcloud.com/r/90/#review294
-----------------------------------------------------------
This RFE was filed before I had figured out that we could signal failure from the systemd
target support. We can't really treat "inactive" as being a rolekit error.
We should treat it as "ready-to-start". The error case should probably only show
up if we receive a failure message from the targets. We need to handle the
NotifyUnitFailed method on the role manager interface for that.
src/rolekit/server/rolebase.py
<
http://reviewboard-fedoraserver.rhcloud.com/r/90/#comment169>
I think if it's in the "inactive" state, we should treat it as
'ready-to-start', since we can probably assume it was stopped by some other
process.
However, the 'failure' state should put us in Error.
- Stephen Gallagher
On Oct. 11, 2014, 12:36 p.m., Thomas Woerner wrote:
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviewboard-fedoraserver.rhcloud.com/r/90/
-----------------------------------------------------------
(Updated Oct. 11, 2014, 12:36 p.m.)
Review request for RoleKit Mailing List, Miloslav Trmac, Stephen Gallagher, and Thomas
Woerner.
Repository: rolekit
Description
-------
If the role was running and the systemd target unit is inactive at roled wakeup, the
instance will be moved into error state.
Diffs
-----
src/rolekit/dbus_utils.py 018e5fb4c878c7d61e89fff2559d42e4bc3c984c
src/rolekit/server/rolebase.py f9596e0bf15840d21869d434ae45ffa00dc53632
Diff:
http://reviewboard-fedoraserver.rhcloud.com/r/90/diff/
Testing
-------
Thanks,
Thomas Woerner