[PVE-User] Losing quorum - cluster broken

[Shain Miley]

We had a similar issues when we upgraded a few months ago...around the same time we moved from cisco 3750s to Nexus 5ks....we had to enable a multicast querier on the switch and everything started working again.

Shain
Sent from my iPhone

> On Apr 24, 2015, at 7:14 AM, Nicolas Costes <nicolas.costes at univ-nantes.fr> wrote:
> 
> 
> When restarting cman (I often have to kill -9 processes to do so), multicast 
> works for a while :
> 
> # omping -m 239.192.205.35 yin hongcha
> 
> yin : waiting for response msg
> yin : joined (S,G) = (*, 239.192.205.35), pinging
> yin :   unicast, seq=1, size=69 bytes, dist=0, time=0.310ms
> yin : multicast, seq=1, size=69 bytes, dist=0, time=0.331ms
> yin :   unicast, seq=2, size=69 bytes, dist=0, time=0.427ms
> yin : multicast, seq=2, size=69 bytes, dist=0, time=0.391ms
> yin :   unicast, seq=3, size=69 bytes, dist=0, time=0.427ms
> yin : multicast, seq=3, size=69 bytes, dist=0, time=0.390ms
> [...]
> 
> Then after some minutes, indeed it stops and there is only unicast left.
> 
> As long as both nodes interfaces are in promiscuous mode (with tcpdump 
> running), the cluster stays in sync. It has been for 18 hours now.
> 
> -- 
> Nicolas Costes
> Responsable de parc informatique
> IUT de la Roche-sur-Yon
> Université de Nantes
> Tél.: 02 51 47 40 29
> _______________________________________________
> pve-user mailing list
> pve-user at pve.proxmox.com
> http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user