Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hello,
(This must be a FAQ but a few hours of digging the net failed to come
up with an answer).
My environment are a couple of CentOS 5 x86_64 Xen DomU machines built
just to learn how to use DRBD (and heartbeat). They have DRBD
8.0.6-1.el5.centos installed on them.
I got DRBD up and running without a problem initially with some test data on it.
After a couple of weeks of messing around with heartbeat, where:
1. The test DRBD partition haven't been touched.
2. The machines were rebooted repeatedly.
3. Heartbeat was NOT configured to manage this resource yet.
This is the state of the DRBD resource as I found it today:
On the primary:
# cat /proc/drbd
version: 8.0.6 (api:86/proto:86)
SVN Revision: 3048 build by buildsvn at c5-x8664-build, 2007-10-25 06:25:16
1: cs:WFConnection st:Secondary/Unknown ds:Inconsistent/DUnknown C r---
ns:0 nr:0 dw:0 dr:0 al:0 bm:0 lo:0 pe:0 ua:0 ap:0
resync: used:0/31 hits:0 misses:0 starving:0 dirty:0 changed:0
act_log: used:0/127 hits:0 misses:0 starving:0 dirty:0 changed:0
On the secondary:
# cat /proc/drbd
version: 8.0.6 (api:86/proto:86)
SVN Revision: 3048 build by buildsvn at c5-x8664-build, 2007-10-25 06:25:16
1: cs:WFConnection st:Secondary/Unknown ds:Inconsistent/DUnknown C r---
ns:0 nr:0 dw:0 dr:0 al:0 bm:0 lo:0 pe:0 ua:0 ap:0
resync: used:0/31 hits:0 misses:0 starving:0 dirty:0 changed:0
act_log: used:0/127 hits:0 misses:0 starving:0 dirty:0 changed:0
The primary boots as secondary, naturaly, and agrees to change to both
primary and secondary when I do a manual "drbdadm primary/secondary
all".
But the secondary refuses to do so, giving:
# drbdadm primary all
State change failed: (-2) Refusing to be Primary without at least one
UpToDate disk
Command 'drbdsetup /dev/drbd1 primary' terminated with exit code 11
drbdsetup exited with code 11
I think I understand correctly that the secondary thinks that it's out
of date (in fact, I forced it to think that way using various ways in
order to trigger a re-sync. But it doesn't help.
So what can I do?
I can loose this data, it's just for a test, but I need to learn what
to do on a production system where this data is crucial.
I understand that in production, heartbeat is supposed to manage this
automatically, but I need to understand what needs to be done manually
too.
Thanks,
--Amos