Hi,

I see it the other way:
What are the reasons to use V2-Clusters?

- Resource Group interdependencies can't be modelled in v1
- Monitoring Resources is not possible with pure v1

So put all together I still use V1-Clusters in General where the following 
criteria are met:

- Straight sequential resource dependencies only within each RG
- Monitoring can easyly be done by using cron-jobs

So up to now there is only one cluster left, which should be v2 - but I did not 
have time to test it's dependencies and the various error-constellations and 
cluster reactions. So that cluster ist being operated manually...

My last try was with 2.0.7 - where I hit a bug regarding RG-interdependencies 
(which was propably fixed with 2.0.8).

Kind regards, Nils
 

> -----Ursprüngliche Nachricht-----
> Von: [EMAIL PROTECTED] 
> [mailto:[EMAIL PROTECTED] Im Auftrag von 
> Martin Bene
> Gesendet: Dienstag, 28. August 2007 09:51
> An: General Linux-HA mailing list
> Betreff: [Linux-HA] Any reason not to keep using v1 style 
> configurations?
> 
> Hi,
> 
> I'm a longtime user of heartbeat, and have quite a few 
> clusters (usualy two node drbd/web/database) running in v1 
> style configurations.
> 
> Setting up my latest system, I actually tried to get such a 
> system configured using crm - and am now back at a v1 style config.
> 
> Is there any compelling reason NOT to keep using v1 style 
> configs for the moment?
> 
> The problems that made me revert the configuration to v1:
> 
>  - couldn't get colocation of two drbd resources on the same 
> node to work
>  - gui doesn't have full support for master/slave resources
>  - got weird results after (non monitored) resource failure: 
> stonith of both nodes (sequentially  - first node1 shot 
> node2, then after node2 rebooted and rejoined the cluster, 
> node1 got shot - the sequence was triggered by running 
> crm_resource -P -r <failed resource>
> 
> Bye, Martin
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
> 
> 
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to