[Pacemaker] Unicast communication is working, but mcastaddr is no working

[Andrew Beekhof]

On 3 Jul 2014, at 7:46 am, Ziqing Zhuang <ziqing.zhuang at avid.com> wrote:

> I am trying to configure Corosync here. At first, I tried to use unicast, here is part of my corosync.conf file (I did not change others):
> 
>   interface {
> 
>                 # The following values need to be set based on your environment
> 
>                 member {
> 
>                         memberaddr: 192.168.254.131
> 
>                 }
> 
>                 member {
> 
>                         memberaddr: 192.168.254.134
> 
>                 }
> 
>                 ringnumber: 0
> 
>                 bindnetaddr: 192.168.254.0
> 
>                 mcastport: 5405
> 
>         }
> 
>         transport: udpu
> 
> Then I run crm_mon -1,
> 
> Everything works fine. It says 2 nodes configured.
> 
> But if I use multicast, by changing this interface part to the following:
> 
>  
> 
> interface {
> 
>                 # The following values need to be set based on your environment
> 
>                 ringnumber: 0
> 
>                 bindnetaddr: 192.168.254.0
> 
>                 mcastaddr: 227.94.1.2
> 
>                 mcastport: 5405
> 
>    }
> 
> Then the system doesn’t recognize any nodes. It just said 0 node configured.
> 
> So I use
> 
> root at devenv1: corosync-cfgtool –s
> 
> Printing ring status.
> 
> Local node ID 117352640
> 
> Could not get the ring status, the error is: 6
> 
> I am assuming mcastaddr is not correct, so I tried several different address, but it still doesn’t work.
> 
> Any advice?

Bad switch?  Not all of them like multicast (out of the box at least)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 841 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20140703/52f020c2/attachment-0003.sig>