heitor.lessa at hotmail
Jul 2, 2012, 7:14 AM
Post #4 of 4
(754 views)
Permalink
|
|
Re: Pacemaker/Corosync issue on Amazon VPC (ec2) [solved]
[In reply to]
|
|
Solved.
I installed a new distribution, configured and installed exactly the same as I did on old ones, then we got a different issue - No nodes were appearing in crm status.
Increased corosync logs to Debug and sorted some issues that raised, then changed /etc/corosync/service.d/pcmk file in order to use version 0, restart corosync and worked !
Changed now to Version 1 as I have read in some documentations, rebooted both machines and still working perfectly.
Thanks.
> From: heitor.lessa [at] hotmail
> To: linux-ha [at] lists
> Date: Sat, 30 Jun 2012 17:10:43 +0000
> Subject: Re: [Linux-HA] Pacemaker/Corosync issue on Amazon VPC (ec2)
>
> Hi Andreas,
>
> Thanks for your answer firstly.
>
> As I said before I created sockets using random ports TCP And UDP, both worked perfectly in tests with netcat and iperf.
>
> For sake I also chose all traffic allowed and issue continues.
>
> I am testing using a different distributions to make sure.
>
> Date: Sat, 30 Jun 2012 15:08:34 +0200
> From: andreas [at] hastexo
> To: linux-ha [at] lists
> Subject: Re: [Linux-HA] Pacemaker/Corosync issue on Amazon VPC (ec2)
>
> On 06/29/2012 05:22 PM, Heitor Lessa wrote:
> >
> > Hi,
> > I have installed DRBD+OCFS2 and working Amazon EC2, however as a previous thread suggested we should use Pacemaker in order to get OCFS modified in runtime (modify/del nodes).
> > Pacemaker/corosync and other components were very straight forward installing via Lucid-Cluster and Ubuntu-HA, but at the first steps I experienced some problems with CoroSync regarding network connectivity.
> > Unfortunately, Amazon does not allow Multicast, so I used udpu once it would be the only way to get it working, but when I started I got same error on logs Even with all traffic allowed, no apparmor (ubuntu), no iptables locally at all:
> >
> > Jun 29 15:11:11 corosync [TOTEM ] Totem is unable to form a cluster because of an operating system or network fault. The most common cause of this message is that the local firewall is configured improperly.
> > Just for sake, I used iperf and netcat to send UDP packets and it is working fine in several ports, so we can rule out firewall issue.
> > Any thoughts?
>
> yes .. security groups, adjustable in your EC2 management console.
>
> Regards,
> Andreas
>
> --
> Need help with Pacemaker?
> http://www.hastexo.com/now
>
> > Thank you very much.
> >
> >
> > _______________________________________________
> > Linux-HA mailing list
> > Linux-HA [at] lists
> > http://lists.linux-ha.org/mailman/listinfo/linux-ha
> > See also: http://linux-ha.org/ReportingProblems
> >
>
>
>
>
>
>
>
> _______________________________________________
> Linux-HA mailing list
> Linux-HA [at] lists
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
> _______________________________________________
> Linux-HA mailing list
> Linux-HA [at] lists
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
|
signature.asc
