Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
/ 2006-10-02 16:25:04 -0700
\ Robinson, Eric:
> Lars said...
>
> <snip>
> ---------------------------------------
> I'll quote the code, to clarify the intention for when
> "degr-wfc-timeout" is used:
> /* If I am currently not Primary,
> * but meta data primary indicator is set,
> * I just now recover from a hard crash,
> * and have been Primary before that crash.
> *
> * Now, if I had no connection before that crash
> * (have been degraded Primary), chances are that
> * I won't find my peer now either.
> *
> * In that case, and _only_ in that case,
> * we use the degr-wfc-timeout instead of the default,
> * so we can automatically recover from a crash of a
> * degraded but active "cluster" after a certain timeout.
> */
> ---------------------------------------
> </snip>
>
> Hmmm... why not always start both nodes as Secondary with no timeout,
> then let heartbeat force the right one to be Primary? Wouldn't that
> avoid blocking and split-brain?
thats why it is configurable...
just make sure that your heartbeat won't decide to make a node primary
that happens to have long-since outdated data
cluster fine
secondary crash [first spike of a brown out]
time passes
primary crash [well, now its a real black out]
... [power back]
previously secondary comes up, heartbeat decides to make it primary
*** you are primary with outdated data ***
previously primary needs a lot longer (recounts its scsi devices,
thinks it needs to fsck its root, whatever)...
same effect as split brain: diverging data sets.
> Is it possible to set up something like this...
>
> Cluster fine, connected.
> Pull plug from primary machine A.
> Heartbeat on B takes over, forces drbd on B to be primary.
> Plug machine A back in.
> Heartbeat on A detects it is secondary, forces drbd on A to go
> secondary.
>
> If this is a stupid question, forgive me. I am shameless about showing
> my ignorance. :-)
well, it depends what plug you pull,
and what you can teach heartbeat to detect.
--
: Lars Ellenberg Tel +43-1-8178292-0 :
: LINBIT Information Technologies GmbH Fax +43-1-8178292-82 :
: Schoenbrunner Str. 244, A-1120 Vienna/Europe http://www.linbit.com :
__
please use the "List-Reply" function of your email client.