On Tue, Jan 20, 2004 at 03:32:59AM -0500, Ivan Popov wrote:
> it looks like I am hitting rvm fragmentation or some other phenomenon that
> makes further volume modifications fail (create or delete files).

No, this is not RVM fragmentation.

> (the logs complain about no space in the volume log, trying to wrap around
> and failing)

You have one (or more) clients that are not talking to both servers.
Possibly weak reintegration, or a server was down for a while. As a
result one server has a stale copy of the volume. Luckily the other
server kept meticulous track of what operations the missing server
hasn't seen yet.

The problem you face is that we ran out of the allocated space to log
the operations.

> At that point the copy was not yet finished, and both of my replicated
> servers died (10 minutes one after the other), the first one logging:

That would probably be around the time that the client started to
trigger resolution. We need a log entry to commit the success of the
resolution before we can truncate, or at least something like that.

For each volume replica of the replicated volume do,

    volutil -h <server> setlogparms <volid> reson 4 logsize 16384

This should double (or quadruple?) the size of the resolution log on
each volume. After that you probably would want to run a recursive ls
through the volume to make sure both replicas are in sync again.

Jan