[DRBD-user] see replicated data

agag rysic at vp.pl
Thu May 31 15:10:02 CEST 2012

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


OK! So main task for DRBD is to replicate data? If main server crashes then
Pacemaker prometes secondary? If I'll have GFS then for what do I nead DRBD?

Can DRBD also replicate ethernet interfaces? (it is not necessary becouse of
vrrp but may be that is something what DRBD offers).



I'm sorry, may be that are stupid questions but I'm not native english
speaker - I can mix something ;)
What does it mean stonith? I can't find it in dictionary :)



Lars Ellenberg wrote:
> 
> On Thu, May 31, 2012 at 05:23:44AM -0700, agag wrote:
>> 
>> I read Guide but not everything is clear for me yet :)
>> 
>> About "understanding how primary-primary" works is not much written:
>> http://www.drbd.org/users-guide-8.3/s-dual-primary-mode.html.
>> 
>> My first idea is to have in example WWW server and when one crashes,
>> second
>> is working without any additional admin tasks. But if I don't see the
>> same
>> data in bouth primary servers (or data is not mounted in secondary) then
>> it
>> will not work like that... - Yes I can use GFS and centralized database
>> server and replicate php code files but then DRBD is starting to be not
>> neaded - am I right?
> 
> If you need data accessible at the same time on both nodes you need a
> cluster file system, or a network file system.
> 
> You can put OCFS2 or GFS2 on top of dual-primary DRBD.
> As I said, doing cluster file systems right is complex,
> and requires stonith among other things.
> 
> If it is sufficient to have the data available on both nodes,
> but exclusively only on one at any given time,
> we have a much more simple failover configuration.
> 
>> Second idea is to have almost identical operating systems. For example to
>> use drbd replicated data for /etc and /bin (wich will be modified only in
>> primary/first serverd when changes or updates are done). Is it possible
>> somehow?
> 
> Usually you put your *data* on DRBD, and use a failover configuration.
> e.g. you have pacemaker mount /dev/drbd0 to /var/www (or so), move an IP
> over, then start or restart apache.
> 
> On failover, the other node will promote drbd to primary,
> mount it, configure the IP, and (re)start apache.
> 
> Or any other service...
> 
> -- 
> : Lars Ellenberg
> : LINBIT | Your Way to High Availability
> : DRBD/HA support and consulting http://www.linbit.com
> 
> DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.
> __
> please don't Cc me, but send to list   --   I'm subscribed
> _______________________________________________
> drbd-user mailing list
> drbd-user at lists.linbit.com
> http://lists.linbit.com/mailman/listinfo/drbd-user
> 
> 

-- 
View this message in context: http://old.nabble.com/see-replicated-data-tp33895972p33938096.html
Sent from the DRBD - User mailing list archive at Nabble.com.




More information about the drbd-user mailing list