On 2008-06-03T16:35:22, "Hildebrand, Nils, 232" <[EMAIL PROTECTED]> wrote:
> > - [Quality not important for hardware?]
> I think the truth is somewhere in between. If you have a cluster it is
> not that important if a node goes down (depending on how available your
> services have to be).
Right, but telling customers to save on hardware because they are
getting clustering is, uhm, not something I'd advocate ;-)
> > [no network bonding needed?]
> Same thing. Network cards very rarely do crash. And if so - failover,
> repair it and failback.
Wrong. A system without redundant network media is not supported. It
doesn't need to have bonding, but you really, really want redundant comm
channels.
> > - [STONITH needed?]
> I think more important than stonith is having at least two independent
> heartbeat-lines.
Wrong. (Sorry, I've been told I should stop making such absolute
statements. Let me rephrase:)
Not quite. STONITH does more than just recover from split-brain, it is
also needed for stop recovery in some cases. STONITH is the ultimate
split-brain protection though and possibly more important than dual
comm channels.
In practice, I'll disregard any data corruption caused by not having
both stonith and redundant comm channels as "doh, your own fault" ;-)
> > - [time is important - but only for logs]
> Why is there a heartbeat-message "time has jumped back - compensating"?
That doesn't track whether time is set correctly, but whether the system
time as reported is monotonously increasing. And if it is not, it is
compensated for.
That can happen when the clock is adjusted abruptly, ie it is a possible
side effect of running ntpdate via cron too. Not all applications can
compensate this, some will exhibit rather strange behaviour because they
are more trusting than our rather paranoid libraries ;-)
xntpd avoids this by smoothing all jumps and only making minimal
adjustments. The right combination is to use ntpdate once at system boot
and then run xntpd (like the xntp init script does.)
Regards,
Lars
--
Teamlead Kernel, SuSE Labs, Research and Development
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
"Experience is the name everyone gives to their mistakes." -- Oscar Wilde
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
