Fri, Jul 26, 2013 at 12:59:41PM CEST, olichtne(a)redhat.com wrote:
From: Ondrej Lichtner <olichtne(a)redhat.com>
This patch series extends the amount of cleanup that the slave does by itself.
In the past cleanup was only executed when the controller requested it through
rpc, which was not the best way since both the controller or the slave can stop
before that happens and leave the slave machine in an inconsistent state.
Now the slave detects when the controller disconnects and cleans any leftover
configuration by itself, the same happens when the slave is terminated.
Finally I added an exception handler to the connection handler since it could
cause the slave to crash without cleanup.
There should be a followup patch that also notifies the controller about the
fact that the slave is terminating, but that will have to wait a bit when Radek
stabilizes his Controller code ans submits it.
Ondrej Lichtner (4):
NetConfig: store references to NetConfigDevice objects
NetTestSlave: cleanup on controller disconnect
NetTestSlave: add cleanup on slave finish
ConnectionHandler: add handling of select exceptions
lnst/Common/ConnectionHandler.py | 5 +++-
lnst/Slave/NetConfig.py | 26 +++++++++++---------
lnst/Slave/NetTestSlave.py | 53 ++++++++++++++++++++--------------------
3 files changed, 45 insertions(+), 39 deletions(-)
--
1.8.3.1
_______________________________________________
LNST-developers mailing list
LNST-developers(a)lists.fedorahosted.org
https://lists.fedorahosted.org/mailman/listinfo/lnst-developers
Applied, thanks.