Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Mon, Mar 09, 2009 at 11:31:39AM +0100, Sebastian Reese wrote: > Hi, > > i am running drbd version Version: 8.0.11 (api:86) on two ubuntu hardy > servers. > drbd ist running on top of lvm with internal metadata. > Kernel is: 2.6.24-19-xen > > Till now i got no problems with drbd. I got over 40 resources up and > running. > > Today i wanted to add resources 44 and 45. > > so what i made was the following (each on both nodes, via cluster ssh, > so i made the exact same things): > > - created the lvm disks via lvcreate > - created the internal metadata > - added the new resources to the drbd.conf > - performed an /etc/init.d/drbd reload > > that worked on one host, but not on the other. > It told me: > root at xen3:~# /etc/init.d/drbd reload > * Reloading DRBD configuration > Failure: (127) Device minor not allocated > Command '/sbin/drbdsetup /dev/drbd44 disk /dev/data1/vsrv-www2-disk > /dev/data1/vsrv-www2-disk internal --set-defaults --create-device' > terminated with exit code 10 > > so wheres the problem? you apparently hit an "out of (some particular kind of) memory" problem, and a bug in DRBD with unclean handling of that. > in dmesg i found: > [12574226.859838] sysfs: duplicate filename 'drbd44' can not be created > [12574226.859851] WARNING: at > /build/buildd/linux-2.6.24/debian/build/custom-source-xen/fs/sysfs/dir.c:424 > sysfs_add_one() > [12574226.859854] Pid: 6569, comm: cqueue/2 Not tainted 2.6.24-19-xen #1 > [12574226.859855] > [12574226.859856] Call Trace: > [12574226.859877] [<ffffffff802efe68>] sysfs_add_one+0xb8/0xf0 > [12574226.859880] [<ffffffff802f0480>] create_dir+0x60/0xb0 > [12574226.859884] [<ffffffff802f0501>] sysfs_create_dir+0x31/0x50 > [12574226.859887] [<ffffffff80336d12>] kobject_get+0x12/0x20 > [12574226.859890] [<ffffffff80337333>] kobject_add+0xb3/0x200 > [12574226.859893] [<ffffffff802e9a76>] register_disk+0x56/0x240 > [12574226.859897] [<ffffffff8032da14>] add_disk+0x34/0x40 > [12574226.859908] [<ffffffff88322d7e>] :drbd:drbd_new_device+0xde/0x1e0 > [12574226.859912] [<ffffffff883020f0>] :cn:cn_queue_wrapper+0x0/0x30 > [12574226.859920] [<ffffffff88324c7d>] :drbd:ensure_mdev+0x4d/0x110 right. on the first attempt, DRBD added the device to the kernel, but then did abort for some other reason, without removing it from the kernel again. this is a DRBD bug. what you see here is a second (or else) attempt, where drbd tries to add it again, and the kernel does complain (correctly): > [12574226.859968] kobject_add failed for drbd44 with -EEXIST, don't try > to register things with the same name in the same directory. exactly. > can anyone help me out? you need to wait until we fix the bug, and then upgrade. and then see why DRBD was not able to completely construct the device structures in the first attempt, and fix that. -- : 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