[Linux-HA] Active node gives up resources when passive fails -> not good

[Nikita Michalko]

Am Dienstag, 16. September 2008 10:36 schrieb Wayne Gemmell:
Hi Wayne,

just a dumb question:

> Hi all
>
> I'm running 2.1.3 on Ubuntu hardy. My ha.cf looks like below. This is the
> first time I'm using unicast connections and so far there is wierdness
> going on. shutdown by ipmi because of high ambient temp. They've been
> having aircon problems in our hosting center and when the ambient temp gets
> to around 38 deg C my passive server shuts down (No I don't know how to
> stop this from happening.) 
- >  is this maybe so configured in BIOS ?

> The active server then follows suite and gives 
> up all its resources. This happened again at 2am this morning and I really
> need to get to the bottom of this. I'm trying broadcast coms to see if it
> makes a difference but I though I'd post here for ideas.
>
> udpport 694
> ucast eth5 10.40.0.1
> ucast eth5 10.40.0.2
> debug 1
> use_logd on
> keepalive 2
> warntime 15
> deadtime 30
> initdead 45
> auto_failback off
> node nikki
> node asia
> crm off

HTH

Nikita Michalko