Presently timeout is picked up via sysctl command
under sanity-quota/18. Which is picking up timeout
from incorrect location. This patch fixes the issue
by reading via $LCTL directly.
sysctl output:
timeout=$(sysctl -n lustre.timeout)
sysctl: cannot stat /proc/sys/lustre/timeout: No such file or directory
Test-Parameters: trivial fstype=zfs testlist=sanity-quota env=ONLY=18
Signed-off-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Change-Id: I4604d31c615fa65bf79598cc09f05d9c7c7abf1b
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51337
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Hongchao Zhang <hongchao@whamcloud.com>
$LFS setstripe $TESTFILE -i 0 -c 1 || error "setstripe $TESTFILE failed"
chown $TSTUSR.$TSTUSR $TESTFILE || error "chown $TESTFILE failed"
- local timeout=$(sysctl -n lustre.timeout)
+ local timeout=$($LCTL get_param -n timeout)
if [ $io_type = "directio" ]; then
log "Write 100M (directio) ..."
if at_is_enabled; then
timeout=$(at_max_get mds)
else
- timeout=$(lctl get_param -n timeout)
+ timeout=$($LCTL get_param -n timeout)
fi
while [ true ]; do