The rmdir should succeed, but if it doesn't, it's hard to tell what was
there that prevented cleanup. The find will usually print nothing but
when we do have a sloppy task that didn't clean up we'll be able to tell
what it was...
Signed-off-by: Sage Weil <sage@redhat.com>
# just cram an rm -rf here
run.wait(
ctx.cluster.run(
- args=['rmdir', '--', testdir],
+ args=['find', testdir, '-ls',
+ run.Raw(';'),
+ 'rmdir', '--', testdir],
wait=False,
),
)