Distributed Replicated Block Device (DRBD) development
 help / color / mirror / Atom feed
* [Drbd-dev] Summary (2nd try)
@ 2004-08-25 13:01 Philipp Reisner
  2004-08-25 13:09 ` Lars Marowsky-Bree
  0 siblings, 1 reply; 2+ messages in thread
From: Philipp Reisner @ 2004-08-25 13:01 UTC (permalink / raw)
  To: drbd-dev

Hi,

After I realized that we no longer make any node primary automatically :)
Here the suggestion of the new configure option:

after-primary-primary = 
 disconnect	No automatic resynchronisation gets performed. One
		node should drop its net-conf (preferable the
		node that would become sync-target)
		I think this should be default.
 ass-older	Auto sync source is the oder primary (current default i.t.s.)
 ass-younger	Auto sync source is the younger primary
 ass-furthest   Auto sync source is the node that did more modifications
 ass-NODENAME   Auto sync source is the named node 

Falls es schon einen pirmary gibt (also dieses:

      N1    N2
  P --- S   Everything ok.
  P - - S   N1 is failing, but for the moment being just can no
            longer answer the network; but it is still able to update
            drbds generation counts
  ?   - S   Now N1 may be dead, or maybe not
  X   - S   A sane Cluster-mgr makes N2 primary, but stonith N1 first ...
  X   - P   N1 now is really dead.
  S --- P   N1 comes back
  S - : P   oops, N1 has "better" generation counts than N2
            N2 shall become sync target, but since it is
            currently Primary, it will refuse this.
            It goes standalone.

) Dann:

 disconnect   (current behaviour in the situation.)
 ass-*        Auto sync, but the current primary gets sync-source.

What do you think ?

-Philipp
-- 
: Dipl-Ing Philipp Reisner                      Tel +43-1-8178292-50 :
: LINBIT Information Technologies GmbH          Fax +43-1-8178292-82 :
: Schönbrunnerstr 244, 1120 Vienna, Austria    http://www.linbit.com :

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [Drbd-dev] Summary (2nd try)
  2004-08-25 13:01 [Drbd-dev] Summary (2nd try) Philipp Reisner
@ 2004-08-25 13:09 ` Lars Marowsky-Bree
  0 siblings, 0 replies; 2+ messages in thread
From: Lars Marowsky-Bree @ 2004-08-25 13:09 UTC (permalink / raw)
  To: drbd-dev

On 2004-08-25T15:01:59,
   Philipp Reisner <philipp.reisner@linbit.com> said:

> Hi,
> 
> After I realized that we no longer make any node primary automatically :)
> Here the suggestion of the new configure option:
> 
> after-primary-primary = 

This is in case where after the reconnect there is no primary (s/s),
right?

>  disconnect	No automatic resynchronisation gets performed. One
> 		node should drop its net-conf (preferable the
> 		node that would become sync-target)
> 		I think this should be default.
>  ass-older	Auto sync source is the oder primary (current default i.t.s.)
>  ass-younger	Auto sync source is the younger primary
>  ass-furthest   Auto sync source is the node that did more modifications
>  ass-NODENAME   Auto sync source is the named node 
> 
> Falls es schon einen pirmary gibt (also dieses:
> 
[snip]

> ) Dann:
> 
>  disconnect   (current behaviour in the situation.)
>  ass-*        Auto sync, but the current primary gets sync-source.
> 
> What do you think ?

Sounds good, one scenario is missing.

P-P scenario:

disconnect	(same same)
ass-*		the node which would be SyncTarget needs to crash
		itself to 'clear' the higher levels, and then would
		(after the reboot) go through the first scenario.

This may be useful for unattended embedded scenarios, but as it is
pretty severe recovery for a grave error, may even value an additional
flag.


Sincerely,
    Lars Marowsky-Brée <lmb@suse.de>

-- 
High Availability & Clustering	   \\\  /// 
SUSE Labs, Research and Development \honk/ 
SUSE LINUX AG - A Novell company     \\// 


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2004-08-25 13:09 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-08-25 13:01 [Drbd-dev] Summary (2nd try) Philipp Reisner
2004-08-25 13:09 ` Lars Marowsky-Bree

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox