summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorEmmanuel Dreyfus <manu@netbsd.org>2015-02-27 03:50:40 +0100
committerNiels de Vos <ndevos@redhat.com>2015-02-27 00:50:29 -0800
commit6a77db6d19dba5367c02cbf2a5883ac49cef94e2 (patch)
treec381c657bb2a3845efc5a1a6c78bb0decd83cb6a
parent5dc152aac25ac7680ce0f83edc580d2a953d27a7 (diff)
Tests: give sane default for NetBSD NFS mount retry count
Default NetBSD NFS retry count is 10000, which means tests will loop for a long time if the server is not available. We fix this by setting a default retry count to 2 (1 seems to low and breaks regression). If mount_nfs is called with a retry option, it will overrride this default. BUG: 1129939 Change-Id: I1ae16f8caa74d6e9af1aa7a55fd111178af0ad78 Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org> Reviewed-on: http://review.gluster.org/9763 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Niels de Vos <ndevos@redhat.com>
-rw-r--r--tests/nfs.rc2
1 files changed, 1 insertions, 1 deletions
diff --git a/tests/nfs.rc b/tests/nfs.rc
index f79be701a2d..2e0241709d3 100644
--- a/tests/nfs.rc
+++ b/tests/nfs.rc
@@ -29,7 +29,7 @@ function mount_nfs ()
for o in ${opt//,/ }; do
case $OSTYPE in
NetBSD)
- test "x${nopt}" = "x" && nopt="tcp,"
+ test "x${nopt}" = "x" && nopt="tcp,-R=2,"
case $o in
nolock|noac|actimeo=*|mountproto=udp)