Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hi Martin
Checking the drbddisk , the help say :
drbddisk [resource] {start|stop|status}
about it , the resource refer to the name of the
resource on the drbd.conf ?
So , the drbddisk::drbd0 mean that heartbeat execute :
/etc/init.d/drbddisk drbd0 start/stop ??
It generate to me that no switch to Node2 to Primary .
When i perform a heartbeat stop on Node1, happen that
Node 1 umount the partition and Node2 mount the
partition , but the state Primary continue on Node1 :
Node1 :
ersion: 0.7.2 (api:75/proto:74)
SVN Revision: 1483:1485 build by root at cpms1.isc,
2004-08-13 22:23:45
0: cs:Connected st:Primary/Secondary ld:Consistent
ns:132316 nr:272 dw:132588 dr:357 al:32 bm:71 lo:0
pe:0 ua:0 ap:0
Node2
version: 0.7.2 (api:75/proto:74)
SVN Revision: 1483:1485 build by root at cpms2.isc,
2004-08-13 23:28:19
0: cs:Connected st:Secondary/Primary ld:Consistent
ns:272 nr:132316 dw:132588 dr:4285 al:1 bm:3 lo:0
pe:0 ua:0 ap:0
But if i change on haresource : drbddisk::correo (
name of the resource) if work well .
Let me know if i'm correct
Thanks
--- Martin Bene <martin.bene at icomedias.com> wrote:
> Hi German,
>
> > 1- What is the best way for to integrate it with
> > heartbeat since the datadisk changed by drbdisk ?
> > I remember that datadisk do the work for switch
> > primary / secondary and mount and umount
> partition. On
> > 7.2 ( if i not mistake ) only swith , but not
> mount /
> > umount partition . Let me know a good way for to
> do
> > it.
>
> Example from my haressources file:
>
> test-neu1 10.192.64.63 \
> drbddisk::drbd1
> Filesystem::/dev/drbd1::/var/dbase::jfs::noatime \
> postgresql pg_autovacuum
> MailTo::support at icomedias.com::test-neu-db
>
> So mount/unmount is handled by using the
> "Filesystem" ressource script
> provided with heartbeat.
>
> > 2- If one of my two nodes fail (disk corruption ,
> hw
> > problems , etc) and i need to change a new one for
> > recovery the redundance, How can i do for to
> > recontruct DrBd ? ( forcing syncs , etc ) .
>
> If you're using internal metadata, drbd should
> realize that the disk is
> new and resync by itself. If you want to explicitely
> resync a device you
> can use "drbdadm invalidate <ressource>" on the side
> with bad data.
>
> > 4- How can i monitoring that DRBD is working well
> > without testing switchover ?
>
> You'll definitely want to try failovers before
> putting a system in
> production. you can monitor current status of drbd
> by checking the
> information from cat /proc/drbd.
>
> Bye, Martin
>
> _______________________________________________
> drbd-user mailing list
> drbd-user at lists.linbit.com
> http://lists.linbit.com/mailman/listinfo/drbd-user
>
___________________________________________________________ALL-NEW Yahoo! Messenger - all new features - even more fun! http://uk.messenger.yahoo.com