Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Tue, Aug 21, 2007 at 11:37:10AM +0100, Irvine, Doug - Corporate Core - ICT Services wrote: > Hi, > > We have a number of secondary schools which use a RHEL3 pair of servers > as their internet cache. > > > > Unfortunately the initial disk allocation has proved to be too small in > the / partition. > > > Filesystem 1K-blocks Used Available Use% Mounted on > /dev/sda3 10394720 3245996 6620852 33% / > /dev/sda1 198321 25640 162442 14% /boot > none 1282636 0 1282636 0% /dev/shm > /dev/drbd0 62669424 55076 59430904 1% /data > > > I recently used gparted to resize the disks. > > This all seemed to work fine initially, but now we get errors in > /var/log/messages saying the OS is trying to write off the end of the > device. > > I have : > > On the standby server > #drbdadm disconnect all > #mke2fs /dev/sda2 - reports correct disk allocation if you do a new mkfs, this has absolutely nothing to do with resizing. > #drbdadm connect all - do same on primary node > #drbdadm invalidate all if /dev/sda2 is the "lower level disk" of drbd0, DO NOT: mke2fs /dev/sda2. DO: mke2fs -j /dev/drbd0. http://thread.gmane.org/gmane.linux.network.drbd/12690/focus=12692 may help to understand what is going on. > Secondary server then 'syncs' > > Then failover the servers and re-run on the other standby server. > > After a day of running, fsck will report > filesystem size (according to superblock) is xxxxxxxxxxxxxxxxxxx > Physical size of device is yyyyyyyyyyyyyyyyyy > > quit <y> > > I am assuming drbd holds information about partition size differently to > the OS. > > I tried: > # drbdadm resize all > On both servers but I still get the same error. -- : Lars Ellenberg Tel +43-1-8178292-0 : : LINBIT Information Technologies GmbH Fax +43-1-8178292-82 : : Vivenotgasse 48, A-1120 Vienna/Europe http://www.linbit.com : __ please use the "List-Reply" function of your email client.