On Sun, May 22, 2005 at 08:15:51AM +0300, Dan Aloni wrote:
> On Sun, May 22, 2005 at 01:04:35PM +0800, Matt Johnston wrote:
> > On Sat, May 21, 2005 at 11:41:46PM +0200, Thomas Reitmayr wrote:
> > > FYI, this happened while pulling the database from colinux. As it's my
> > > first time trying to use monotone I don't know where the problem is so I
> > > send you the output as instructed.
> > > -Thomas Reitmayr.
> >
> > I can reproduce it here with a known-good monotone instance
> > (near-current head), it seems that the server is sending
> > (valid) revision data which doesn't match the associated
> > revision_id. Someone else reported the same problem to the
> > list about a week ago.
> >
> > (Also the netsync error handling isn't smart enough to know
> > it's not a 'bug in monotone' on the client side).
> >
> > Could the server admin run "db check" and see if that shows
> > anything? (I've bcced him) It seems most likely to be bad
> > hardware or perhaps a strange miscompilation bug, though
> > could be a server-side monotone bug.
> >
> > The revision in question seems to be the first revision of a
> > branch (no old_revision) and has a manifest_id of
> > 785e7d8bed37af1bd14b4b4a6339b8437204011a.
>
> Might this have something to do with the fact that the server is still
> running 0.18?
I don't think so, 0.18 and 0.19 should be compatible with
each other. Do you get output for either of
"monotone cat revision 9a85a4bb58702da9d1caf3836420179b410baa1b"
or
"monotone cat revision c41816932e8237b05bc28412a4a5e82877724cee" ?
Matt
|
