Hi Jan,

On Thu, Feb 28, 2008 at 12:18:39PM -0500, Jan Harkes wrote:
> Well, really cfs forcereintegrate should just reintegrate everything
> unless there is a conflict/disconnection or other reason that prevents
> us from making further progress.
> 
> I've had no problem with cfs fr pushing everything back, although it
> used to be that partial reintegration would break us out of the loop
> too early. But the most recent release should definitely have that
> fixed.

Hmm, may be you are right about "cfs fr" as it is usually returning silently,
I did not check the result.
But "cfs fr <directory>" tells what happens.
I am running client code from git 2008-02-22 and "cfs fr <dir>" does not
reintegrate more than a "handful" of entries where "handful" varies
considerably from 1 (or may be even 0 sometimes?) to many.
Anyway, it tends to return "too early":
---------------------
$ mkdir testre
$ cd testre
$ for a in `seq -w 00 99`; do touch $a;done
$ cfs lv .
Status of volume 7f000139 (2130706745) named "/some/tmp"
  Volume type is ReadWrite
  Connection State is Reachable
  Reintegration age: 0 sec, time 15.000 sec
  Minimum quota is 0, maximum quota is unlimited
  Current blocks used are 138438
  The partition has 9656110 blocks available out of 33815996
  There are 201 CML entries pending for reintegration (49250 bytes)

$ cfs fr .
194 CML entries remaining for volume /some/tmp
$
---------------------
The volume is not replicated, if that matters.

If it is not meant to behave this way, then this is a bug?

Regards,
Rune