[DRBD-user] Oops - drbd will not start on either node

David david at davidbranford.net
Fri May 20 17:08:50 CEST 2005

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


Hi all,

I have thus far been unsuccessfull in starting drbd on either node in my 
test setup.

A quick summary of my setup:

Both machines running redhat 9, with (i686) redhat-built 2.4.20-31.9 kernel, 
kernel sources installed for that same kernel, drbd 0.7.10 compiled against 
said kernel source, using instructions from latest INSTALL.TXT.

I have simply a 400MB partition on each system (/dev/hdc) which I am trying 
to mirror (on both systems). I have a dedicated 260MB partition (/dev/hda5) 
for drbd meta data (on both systems).

I found that there were no /dev/drbd* entries created at redhat install 
time, nor were there any after make install; therefore I used the text file 
suggested method to create them:

# for i in `seq 0 15` ; do mknod /dev/drbd$i b 147 $i; done

on both systems.

drbd.conf:
------------------------------------------------------
global {
    dialog-refresh 5; # 5 seconds
    disable-io-hints;
}

resource r0 {

  protocol C;

  incon-degr-cmd "echo '!DRBD! pri on incon-degr' | wall ; sleep 60 ; 
halt -f";

  startup {
    degr-wfc-timeout 120;    # 2 minutes.
  }

  disk {
    on-io-error   detach;
  }

  net {
  }

  syncer {
    rate 1M;
    group 1;

    al-extents 257;
  }

  on celery {
    device     /dev/drbd0;
    disk       /dev/hdc1;
    address    192.168.0.8:7788;
    meta-disk  /dev/hda5[0];
  }

  on pii {
    device    /dev/drbd0;
    disk      /dev/hdc1;
    address   192.168.0.13:7788;
    meta-disk /dev/hda5[0];
  }

}
------------------------------------------------------

The two test systems have hostnames "celery" (192.168.0.8) and "pii" 
(192.168.0.13).

I can provide more detail if needed. The trouble I have starting drbd occurs 
thusly (after reboot):


# insmod drbd
Using /lib/modules/2.4.20-31.9/kernel/drivers/block/drbd.o

# service drbd start
Starting DRBD resources:    [ d0 /etc/init.d/drbd: line 131:  2625 
Segmentation fault      $CMD

cmd /sbin/drbdsetup /dev/drbd0 disk /dev/hdc1 /dev/hda5 
0 --on-io-error=detach  failed!


...ok, so:

# drbdadm up all

...just hangs, and in system log I get:

23:14:29 celery kernel: drbd: initialised. Version: 0.7.10 (api:77/proto:74)
23:14:29 celery kernel: drbd: SVN Revision: 1743 build by 
root at localhost.localdomain, 2005-05-15 03:00:03
23:14:29 celery kernel: drbd: registered as block device major 147
23:14:42 celery kernel: drbd0: Creating state block
23:14:42 celery kernel: drbd0: Both nodes diskless!
23:14:42 celery kernel: drbd0: Assuming that all blocks are out of sync (aka 
FullSync)
23:14:42 celery kernel: drbd0: drbd_bm_set_all: (!(b && b->bm)) in 
drbd_bitmap.c:553
23:14:42 celery kernel: d595de98 d894f1b7 d8963d20 00000000 d89635b1 
d896972d 00000229 00000000
23:14:42 celery kernel:        00000000 d8951294 d5dd0000 00000000 000001c9 
00000246 d72aa005 00000002
23:14:42 celery kernel:        00000000 00000000 00000000 00001601 d7534380 
d69dcc80 0000000f 00000001
23:14:42 celery kernel: Call Trace:   [<d894f1b7>] drbd_bm_set_all [drbd] 
0x137 (0xd595de9c))
23:14:42 celery kernel: [<d8963d20>] .rodata.str1.32 [drbd] 0x280 
(0xd595dea0))
23:14:42 celery kernel: [<d89635b1>] __func__.17 [drbd] 0x0 (0xd595dea8))
23:14:42 celery kernel: [<d896972d>] .rodata.str1.1 [drbd] 0x0 (0xd595deac))
23:14:42 celery kernel: [<d8951294>] drbd_ioctl_set_disk [drbd] 0x624 
(0xd595debc))
23:14:42 celery kernel: [<d89528b6>] drbd_ioctl [drbd] 0x846 (0xd595df28))
23:14:42 celery kernel: [<c014ec6e>] blkdev_ioctl [kernel] 0x3e 
(0xd595df80))
23:14:42 celery kernel: [<c01571a9>] sys_ioctl [kernel] 0xc9 (0xd595df94))
23:14:42 celery kernel: [<c010953f>] system_call [kernel] 0x33 (0xd595dfc0))
23:14:42 celery kernel:
23:14:42 celery kernel: Unable to handle kernel NULL pointer dereference at 
virtual address 00000000
23:14:42 celery kernel:  printing eip:
23:14:42 celery kernel: d894f0ae
23:14:42 celery kernel: *pde = 00000000
23:14:42 celery kernel: Oops: 0000
23:14:42 celery kernel: drbd parport_pc lp parport autofs sis900 keybdev 
mousedev hid input usb-ohci usbcore ext3 jbd
23:14:42 celery kernel: CPU:    0
23:14:42 celery kernel: EIP:    0060:[<d894f0ae>]    Not tainted
23:14:42 celery kernel: EFLAGS: 00010082
23:14:42 celery kernel:
23:14:42 celery kernel: EIP is at drbd_bm_set_all [drbd] 0x2e (2.4.20-31.9)
23:14:42 celery kernel: eax: 00000000   ebx: d5b52280   ecx: 00000001   edx: 
00000000
23:14:42 celery kernel: esi: d5dd0000   edi: d5dd0000   ebp: 00000000   esp: 
d595dea0
23:14:42 celery kernel: ds: 0068   es: 0068   ss: 0068
23:14:42 celery kernel: Process drbdsetup (pid: 2593, stackpage=d595d000)
23:14:42 celery kernel: Stack: d8963d20 00000000 d89635b1 d896972d 00000229 
00000000 00000000 d8951294
23:14:42 celery kernel:        d5dd0000 00000000 000001c9 00000246 d72aa005 
00000002 00000000 00000000
23:14:42 celery kernel:        00000000 00001601 d7534380 d69dcc80 0000000f 
00000001 00000000 00000000
23:14:42 celery kernel: Call Trace:   [<d8963d20>] .rodata.str1.32 [drbd] 
0x280 (0xd595dea0))
23:14:42 celery kernel: [<d89635b1>] __func__.17 [drbd] 0x0 (0xd595dea8))
23:14:42 celery kernel: [<d896972d>] .rodata.str1.1 [drbd] 0x0 (0xd595deac))
23:14:42 celery kernel: [<d8951294>] drbd_ioctl_set_disk [drbd] 0x624 
(0xd595debc))
23:14:42 celery kernel: [<d89528b6>] drbd_ioctl [drbd] 0x846 (0xd595df28))
23:14:42 celery kernel: [<c014ec6e>] blkdev_ioctl [kernel] 0x3e 
(0xd595df80))
23:14:42 celery kernel: [<c01571a9>] sys_ioctl [kernel] 0xc9 (0xd595df94))
23:14:42 celery kernel: [<c010953f>] system_call [kernel] 0x33 (0xd595dfc0))
23:14:42 celery kernel:
23:14:42 celery kernel:
23:14:42 celery kernel: Code: 81 3c 90 67 02 74 83 74 33 c7 44 24 0c 2e 02 
00 00 8b 15 50


I have repeated the installation from scratch several times (o/s and drbd), 
and received the same results. And these results are identical on both 
machines.

I have checked and checked the running kernel version, the version of kernel 
source I have (from redhat also, btw), uname -r reports "2.4.20-31.9" which 
is correct.

Sorry about the length of this post; I thought it was necessary to provide 
all these details as simply/clearly as possible... if anyone can offer any 
guidance as to where to go from here it would be much appreciated ! :)

Regards,

David. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20050521/a23c0f7f/attachment.htm>


More information about the drbd-user mailing list