yahoo-eng-team team mailing list archive

Thread
Date

[Bug 1240045] [NEW] Weak service state probe algorithm

To: yahoo-eng-team@xxxxxxxxxxxxxxxxxxx
From: Launchpad Bug Tracker <1240045@xxxxxxxxxxxxxxxxxx>
Date: Tue, 15 Oct 2013 12:42:05 -0000
Reply-to: Bug 1240045 <1240045@xxxxxxxxxxxxxxxxxx>
Sender: bounces@xxxxxxxxxxxxx

You have been subscribed to a public bug:

I've faced with problem when service state of compute node got from
nova-manage service list command was XXX most time and some times it was
ok, despite in fact compute was ok all time. I found that it coused by
time shift between controller and compute node. I think that is the bug
to treat node as down basing only on update time, because if we try to
ping, probe tcp port or any other real ways to conversate with node we
will found it ok.

** Affects: nova
     Importance: Undecided
         Status: New

-- 
Weak service state probe algorithm
https://bugs.launchpad.net/bugs/1240045
You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Compute (nova).