Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Sat, May 12, 2007 at 04:45:41AM +0000, manjo wrote: > manjo <manjok at ...> writes: > > I realized I should not have formatted my new raided drives > > before adding them to drbd. Anyhow, as I mentioned in my post, > > I did do a full resync afterwords (it was automatic). So now my > > drbd mirror is functioning well with 450G size on 750G > > capacity physical disks. The device is named drbd0. > > > > Should I now call drbdadm resize on both machines and then do > > resize_reiserfs drbd0 on both machines? When doing filesystem > > resize do I still have to specify a size 128M less than the disk > > capacity? > > > Problem solved! Thought I will post it. > > I was missing the step of resizing the filesystem on the raided > volume. My raided volume was called md1. Calling > "resize_reiserfs /dev/md1" on both the nodes solved > the problem. Earlier I was only calling "drbdadm resize all" > followed by "reisize_reiserfs /dev/drbd0" on both nodes. > > Also after the resize, I was not able to mount drbd0 on > either nodes. Unknown superblock error. Running fsck fixed > that issue. this is definetly not how you do this with drbd. I described earlier how you grow a file system with drbd. you got it wrong again. it may appear now that it works, but you bypassed drbd again. drbd has no means anymore to know which blocks you modified, and you need a full sync to get to some "defined" state. if anything, then you made things worse. I get the feeling that something in your setup is "just wrong". -- : 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 :