RE: [SSI-devel] Another rfork bug(s?)
Brought to you by:
brucewalker,
rogertsang
From: Walker, B. J <bru...@hp...> - 2004-03-19 15:01:13
|
I believe your root node had panic'd and there was no failover node so the other nodes paniced as well. Bruce > -----Original Message----- > From: ssi...@li...=20 > [mailto:ssi...@li...] On=20 > Behalf Of Maxime Ritter > Sent: Friday, March 19, 2004 12:47 AM > To: ssi...@li... > Subject: Re: [SSI-devel] Another rfork bug(s?) >=20 >=20 > On Thu, Mar 18, 2004 at 12:37:07PM -0800, Brian J. Watson wrote: > > Maxime Ritter wrote: > > >After some new crashes, I also noticed that most of the=20 > time, master node=20 > > >begins > > >to say "node 1 died", and "node 2 died".=20 > > >Then the kernel panic (same time or after the message on=20 > master node ?=20 > > >don't know. Maybe even before) and node 1 & 3 :=20 > > ><4>ics_llprobe_clms: receive of master info failed, error=20 > =3D -11; possibly=20 > > >node w > > >ent down<0>Kernel panic: No secondaries for CLMS failover ! > >=20 > > This obtuse panic message means that a node is no longer receiving=20 > > heartbeat packets from either the active root node=20 > (initnode) or any=20 > > failover root nodes. It believes that it can no longer=20 > access the root=20 > > filesystem, so it panics. I'm going to change this message=20 > to read "Lost=20 > > network connection to all potential root nodes", so that=20 > it's more clear=20 > > what's happening. >=20 > That's what I understood, so it wasn't so much obscure :-) >=20 > > The problem you're seeing could be a result of your network switch=20 > > getting saturated, thus preventing heartbeat packets from=20 > being delivered. >=20 > Doesn't OpenSSI try to resend its network messages when it receive no=20 > answer ? >=20 > --=20 > Maxime Ritter >=20 >=20 > ------------------------------------------------------- > This SF.Net email is sponsored by: IBM Linux Tutorials > Free Linux tutorial presented by Daniel Robbins, President and CEO of > GenToo technologies. Learn everything from fundamentals to system > = administration.http://ads.osdn.com/?ad_id=3D1470&alloc_id=3D3638&op=3Dcli= ck > _______________________________________________ > ssic-linux-devel mailing list > ssi...@li... > https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel >=20 |