[DRBD-user] is a resize really needed on both nodes?

Zing zing+drbduser at fastmail.fm
Mon Aug 27 21:47:51 CEST 2007

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


Hello,

I'm testing drbd 8.0.5+(recent actlog patch) resizing (/dev/drbd0 on top
of an lvm volume), and I don't seem to need to do a "drbdadm resize r0" on
the secondary node.  Is this ok?  The man page says I need to do a resize
on _both_ nodes.

Basically, I just lvextend the logical volume on both nodes, and then do
"drbdadm resize r0" on the primary:

drbd0: Moving meta-data.
drbd0: writing of bitmap took 1 jiffies
drbd0: 1536 MB marked out-of-sync by on disk bit-map.
drbd0: 1572892 KB now marked out-of-sync by on disk bit-map.
drbd0: Writing meta data super block now.
drbd0: drbd_bm_resize called with capacity == 5242648
drbd0: resync bitmap: bits=655331 words=20480
drbd0: size = 2559 MB (2621324 KB)
drbd0: writing of bitmap took 19 jiffies
drbd0: 2048 MB marked out-of-sync by on disk bit-map.
drbd0: 2097164 KB now marked out-of-sync by on disk bit-map.
drbd0: No resync, but 524291 bits in bitmap!
drbd0: BUG! md_sync_timer expired! Worker calls drbd_md_sync().
drbd0: Writing meta data super block now.

The secondary looks similar without any drbd resizing.  It seems to
realize what's going on and just do the resize also.  I can grow the
filesystem on the primary and then switch secondary to primary and
everything looks good.

thanks.



More information about the drbd-user mailing list