tests: fix spurious failures in bug-1190734.t (remove-brick)

This is the second leading cause of spurious failures, including those
in tests for other spurious-regression-failure fixes (creating a bit of
a "catch 22" situation).  While these failures have been hard to
reproduce except during full regression-test runs, two changes have been
made that might make this test more resilient to certain types of
failures.

 * Use a specific "ls" instead of a general "find" to list/count only
   the files we're interested in, without (possibly) including transient
   artifacts from the "remove-brick" command.

 * Retry the file count up to five times, just in case there are other
   transient conditions causing it to yield the wrong result.

Also, "inlining" some of the functions for removing the brick might help
to highlight exactly which command within those functions was failing.

Change-Id: I5a462b91fb4e04d9e9a53cc60f9db11b89101107
Signed-off-by: Jeff Darcy <jdarcy@redhat.com>
Reviewed-on: http://review.gluster.org/10013
Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>

0 files changed, 0 insertions, 0 deletions