[DRBD-user] 'drbdadm connect' panic?

[Lars Ellenberg]

On Thu, Jul 19, 2007 at 05:15:18PM -0700, Alex Dean wrote:
> Lars Ellenberg wrote:
> 
> >ok...
> > interrupted during initial handshake, then
> > NULL pointer dereference in force_sig_info...
> >this appears to be because of a race-condition bug I remember vaguely.
> >
> >I'm not sure exactly which changelog item of which 0.7.x this corresponds 
> >to,
> >but it should be fixed in the newest 0.7.
> >
> 
> Lars : Thank you for the help.  We've scheduled an upgrade to the latest 
> 0.7 drbd.
> 
> After that, the correct proceedure to re-connect 2 nodes is to run 
> 'drbadm connect' on both, right?  (I think so, but the panic we saw make 
> me doubt myself.)

if you see both sides in "WFConnect", but they don't really establish a
connection, I'd do "drbdadm disconnect", then verify the network
connection by other means first. once that is sorted out, you can
"drbdadm connect" again. doing a "drbdadm connect" here is unlikely to
help: they try to connect already anyways.

if one of them is "StandAlone", you should first investigate (kernel
logs) why that happened, and then "drbdadm connect" it.

-- 
: Lars Ellenberg                            Tel +43-1-8178292-0  :
: LINBIT Information Technologies GmbH      Fax +43-1-8178292-82 :
: Vivenotgasse 48, A-1120 Vienna/Europe    http://www.linbit.com :
__
please use the "List-Reply" function of your email client.