[Pacemaker] "No respawn" for cluster resources

[Stallmann, Andreas]

Hi!


Von: Andrew Beekhof [mailto:andrew at beekhof.net]

> on-fail=block for the monitor action will prevent automated respawn
> migration-threshold=1

Thanks Andrew. It even is in the book i was reading "Linux Hochverfügbarkeit" by Oliver Liebel), but I somehow missed this.

Now something else is going wrong: When tomcat m/s resource (intentionally) fails (by killing it hard) on node1 (and is reported "unmanaged failed"), but is never moved away from the node. :-(

Are there any other possible parameteres to "on-fail" besides "restart" and "block". Perhaps I have to try something different. More likely though, something is still wrong in my resource script.

Any further ideas welcome.

Cheers,

Andreas


-----------------------------
CONET Solutions GmbH, Theodor-Heuss-Allee 19, 53773 Hennef.
Registergericht/Registration Court: Amtsgericht Siegburg (HRB Nr. 9136)
Geschäftsführer/Managing Directors: Anke Höfer
-----------------------------