[DRBD-user] Error "1 corrupt AL transactions found" after DRBD upgrade

Lars Ellenberg lars.ellenberg at linbit.com
Tue May 26 11:12:06 CEST 2015

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, May 18, 2015 at 11:18:05AM +0000, Christoph Mitasch wrote:
> > This should address all your concerns:
> >
> > http://git.drbd.org/gitweb.cgi?p=drbd-
> > utils.git;a=commitdiff;h=7e63e763a86ea691c5c5eda5b3e6bd747d74b186
> >
> > commit 7e63e763a86ea691c5c5eda5b3e6bd747d74b186
> > Author: Lars Ellenberg <lars.ellenberg at linbit.com>
> > Date:   Mon Sep 9 14:53:01 2013 +0200
> >
> >     drbdmeta apply-al: don't scare users with words like "corrupt"
> >
> >     When converting drbd 8.3 to 8.4 on-disk meta data,
> >     some people have been scared by the message
> >     "XY corrupt or uninitialized AL transactions found."
> >
> >     Calm them down: usually, this is diagnostic only,
> >     and nothing to worry about.
> Great, thanks for clarification.
> Is there any way to fix it except creating new metadata?

To fix what, exactly?

iirc (which is not necessarily the case, because it's been some time)
If this was 8.3 meta data, we cannot really distinguish between
"corrupt" or "never initialized" transaction slots.  So even if you
create new 8.3 meta data, it will probably behave the same.

The activity log is only relevant after a primary crash.
Upgrade procedure should look like: make both nodes
Connected Secondary/Secondary, then cleanly shutdown drbd.

Which is why the above mentioned commit changed the message to read:
                fprintf(stderr, "%u corrupt or uninitialized AL transactions found\n", mx - found_valid);
+               fprintf(stderr, "You can safely ignore this if this node was cleanly stopped (no crash).\n");

: Lars Ellenberg
: http://www.LINBIT.com | Your Way to High Availability
: DRBD, Linux-HA  and  Pacemaker support and consulting

DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.
please don't Cc me, but send to list   --   I'm subscribed

More information about the drbd-user mailing list