On Wed, Nov 24, 2010 at 08:43:00PM +0100, pebcak wrote:
> Heya all,
>
> I have a problem with corosync via iba & ipv6.
>
> I want to build a storage cluster (sheepdog) with corosync while using
> infiniband. However when I use "transport: iba" I get a segmentation
> fault when starting corosync
(sounds like a bug)
Can you get a backtrace?
gdb corosync
run -f
# after segfault
thr a a bt
Thanks
Angus
>
> I took the source from the debian-packages and rebuild them with
> --enable-rdma.
>
> --
>
> corosync [MAIN ] main.c:1300 Corosync Cluster Engine ('1.2.1'): started
> and ready to provide service.
> corosync [MAIN ] main.c:1301 Corosync built-in features: nss rdma
>
> --
>
> The relevant part of my config looks like this:
>
> --
> transport: iba
> interface {
> ringnumber: 0
> bindnetaddr: 2a02:2478:0:1:218:f104:396:e131
> mcastport: 5405
> mcastaddr: ff04::1
> }
> --
>
> And that's what I get when start it:
>
> --
>
> r...@caspar:~# corosync -f
> corosync [MAIN ] main.c:1300 Corosync Cluster Engine ('1.2.1'): started
> and ready to provide service.
> corosync [MAIN ] main.c:1301 Corosync built-in features: nss rdma
> corosync [MAIN ] main.c:1376 Successfully read main configuration file
> '/etc/corosync/corosync.conf'.
> corosync [TOTEM ] totemnet.c:204 Initializing transport (Infiniband/IP).
> Segmentation fault
>
> --
>
> Everything works fine when I change transport to udp, using the same
> device and config.
>
> Did I forget or overlook something?
>
> Cheers,
> Christine
>
> _______________________________________________
> Openais mailing list
> [email protected]
> https://lists.linux-foundation.org/mailman/listinfo/openais
_______________________________________________
Openais mailing list
[email protected]
https://lists.linux-foundation.org/mailman/listinfo/openais
