On Mon, May 20, 2013 at 01:20:34PM -0400, Josef Bacik wrote: > I'm not sure how the numbering is supposed to work now that we've split > everything out so I'm just going with the next number in the directory. This is > a regression test for btrfs send, we had a problem where we'd try to send a file > that had been deleted in the source snapshot. This is just to make sure we > don't have the same problem in the future. Thanks, You are holding open an unlinked file? You want to use src/multi_open_unlink, then. /* * multi_open_unlink path_prefix num_files sleep_time * e.g. * $ multi_open_unlink file 100 60 * Creates 100 files: file.1, file.2, ..., file.100 * unlinks them all but doesn't close them all until after 60 seconds. */ If you need to hold open existing files, add support for that into multi_open_unlink.c (i.e. allow it to ignore EEXIST when trying to create files). > +# need this so that tail doesn't error out racing with the rm > +sleep 1 > +rm -f $SCRATCH_MNT/foo > +$BTRFS_UTIL_PROG subvol snap -r $SCRATCH_MNT $SCRATCH_MNT/snap1 > /dev/null 2>&1 > +$BTRFS_UTIL_PROG send -f /dev/null -p $SCRATCH_MNT/snap $SCRATCH_MNT/snap1 >/dev/null 2>&1 I'd send this output to $seqres.full, so if the test fails there's debug output to look at... Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
