init-ceph: continue after failure doing osd data mount

If we are starting many daemons and hit an error, we normally note it and
move on.  Do the same when doing the pre-mount step.

Fixes: #8554
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit 6a7e20147cc39ed4689809ca7d674d3d408f2a17)

diff --git a/src/ceph_common.sh b/src/ceph_common.sh
index 01781b754157fa94369f08b158f4f58fbe0a224b..d78f831f5e866b9e1b9bf92763a485179174f259 100644 (file)
--- a/src/ceph_common.sh
+++ b/src/ceph_common.sh
@@ -137,6 +137,24 @@ do_root_cmd() {
     fi
 }
 
+do_root_cmd_okfail() {
+    ERR=0
+    if [ -z "$ssh" ]; then
+       [ $verbose -eq 1 ] && echo "--- $host# $1"
+       ulimit -c unlimited
+       whoami=`whoami`
+       if [ "$whoami" = "root" ]; then
+           bash -c "$1" || { [ -z "$3" ] && echo "failed: '$1'" && ERR=1 && return 1; }
+       else
+           sudo bash -c "$1" || { [ -z "$3" ] && echo "failed: '$1'" && ERR=1 && return 1; }
+       fi
+    else
+       [ $verbose -eq 1 ] && echo "--- $rootssh $2 \"if [ ! -d $sshdir ]; then mkdir -p $sshdir; fi; cd $sshdir ; ulimit -c unlimited ; $1\""
+       $rootssh $2 "if [ ! -d $sshdir ]; then mkdir -p $sshdir; fi; cd $sshdir ; ulimit -c unlimited ; $1" || { [ -z "$3" ] && echo "failed: '$rootssh $1'" && ERR=1 && return 1; }
+    fi
+    return 0
+}
+
 get_local_daemon_list() {
     type=$1
     if [ -d "/var/lib/ceph/$type" ]; then

diff --git a/src/init-ceph.in b/src/init-ceph.in
index 846bd573cef760444ea502e493780c32ed9d266c..1effb564543572a088cb10d56494c3f5d5c2d260 100644 (file)
--- a/src/init-ceph.in
+++ b/src/init-ceph.in
@@ -311,10 +311,14 @@ for name in $what; do
 
                if [ "$fs_type" = "btrfs" ]; then
                    echo Mounting Btrfs on $host:$fs_path
-                   do_root_cmd "modprobe btrfs ; btrfs device scan || btrfsctl -a ; egrep -q '^[^ ]+ $fs_path' /proc/mounts || mount -t btrfs $fs_opt $first_dev $fs_path"
+                   do_root_cmd_okfail "modprobe btrfs ; btrfs device scan || btrfsctl -a ; egrep -q '^[^ ]+ $fs_path' /proc/mounts || mount -t btrfs $fs_opt $first_dev $fs_path"
                else
                    echo Mounting $fs_type on $host:$fs_path
-                   do_root_cmd "modprobe $fs_type ; egrep -q '^[^ ]+ $fs_path' /proc/mounts || mount -t $fs_type $fs_opt $first_dev $fs_path"
+                   do_root_cmd_okfail "modprobe $fs_type ; egrep -q '^[^ ]+ $fs_path' /proc/mounts || mount -t $fs_type $fs_opt $first_dev $fs_path"
+               fi
+               if [ "$ERR" != "0" ]; then
+                   EXIT_STATUS=$ERR
+                   continue
                fi
            fi