[DRBD-user] drbd 0.7.7 disconnects at mount

[Johan Karlsson]

Hello,

I have a setup with linux 2.4.28, LVM and drbd 0.7.7.
The shared resource is mapped onto two LVM volumes, each of
equal size, like this


  on mail1-p {
    device     /dev/drbd0;
    disk       /dev/vg00/mail;
    address    192.168.0.74:7788;
    meta-disk  internal;

  }

  on standby1-p {
    device    /dev/drbd0;
    disk      /dev/vg00/mail-standby;
    address   192.168.0.75:7788;
    meta-disk internal;
  }


They connect to each other and synchronize just fine.
I then drbdadm primary all  on host mail1-p, and /proc/drbd on both sides
indicate that all is well;

mail1-p:~# cat /proc/drbd 
version: 0.7.7 (api:77/proto:74)
SVN Revision: 1680 build by root at host, 2005-01-02 23:21:04
 0: cs:Connected st:Primary/Secondary ld:Consistent
    ns:1052672 nr:0 dw:0 dr:1052672 al:0 bm:514 lo:0 pe:0 ua:0 ap:0

standby1-p:~# cat /proc/drbd 
version: 0.7.7 (api:77/proto:74)
SVN Revision: 1680 build by root at host, 2005-01-02 23:21:04
 0: cs:Connected st:Secondary/Primary ld:Consistent
    ns:0 nr:2092836 dw:2092836 dr:0 al:0 bm:511 lo:0 pe:0 ua:0 ap:0


So - the next step is of course to mount the disk - but here comes the
problem: both mounts and fscks of drbd0 make the connection disappear;

mail1-p:~# fsck /dev/drbd0
[hangs, open new terminal]
mail1-p:~# cat /proc/drbd 
version: 0.7.7 (api:77/proto:74)
SVN Revision: 1680 build by root at host, 2005-01-02 23:21:04
 0: cs:BrokenPipe st:Primary/Unknown ld:Consistent
    ns:1054984 nr:0 dw:2312 dr:1057256 al:372 bm:514 lo:0 pe:578 ua:0 ap:577

standby1-p:~# cat /proc/drbd 
version: 0.7.7 (api:77/proto:74)
SVN Revision: 1680 build by root at host, 2005-01-02 23:21:04
 0: cs:WFConnection st:Secondary/Unknown ld:Consistent
    ns:0 nr:2095148 dw:2095148 dr:0 al:0 bm:511 lo:0 pe:0 ua:0 ap:0

On mail1-p, dmesg indicates

drbd0: Secondary/Secondary --> Primary/Secondary
drbd0: sock was shut down by peer
drbd0: drbd0_receiver [391]: cstate Connected --> BrokenPipe
drbd0: short read expecting header on sock: r=0
drbd0: worker terminated

While being a bit more informative on standby1-p
drbd0: Secondary/Secondary --> Secondary/Primary
drbd0: PingAck did not arrive in time.
drbd0: drbd0_asender [576]: cstate Connected --> NetworkFailure
drbd0: asender terminated
drbd0: drbd0_receiver [523]: cstate NetworkFailure --> BrokenPipe
drbd0: short read expecting header on sock: r=-512
drbd0: worker terminated
drbd0: drbd0_receiver [523]: cstate BrokenPipe --> Unconnected
drbd0: Connection lost.
drbd0: drbd0_receiver [523]: cstate Unconnected --> WFConnection


The only way to recover from this condition seem to be a reboot of
mail1-p, as drbd cannot be restarted due to fsck/mount having opened
the device. This has occurred every time I've tried to access the
shared resource.

Any help / suggestions would be greatly appreciated.