|
On 1999-07-07T10:09:00,
John Connett <jrc@xxxxxxxxxxxxxx> said:
> I have produced an xyz.monitor (using an xyz.ping) which appears to have
> the desired behaviour but I am uncertain about suitable time values for
> "alertevery", "upalertafter", and similar. I would like the "alert" to
> delete the failing server and the "upalert" to add the restarted server
> to the configuration using ipvsadm. Once I have a working example I
> will be able to tune it for my particular service.
Thats not perfectly true.
You are right: The alert will delete the failed server. However, you only get
an upalert if _the whole hostgroup_ comes up again, so you will not get a
single server added back in if one is still down.
I am currently working with Jim to fix those issues. Main problem is that mon
keeps a per hostgroup status instead of a per host status. Doing this right is
a perfect time to fix some other things within mon, so don't expect something
too soon. ;-)
Sincerely,
Lars Marowsky-Brée
--
Lars Marowsky-Brée
Network Management
teuto.net Netzdienste GmbH - DPN Verbund-Partner
|
