Initscript - add backup of corrupt non-persistent databases

Corrupt non-persistent databases never get analysed because ctdbd
zeroes them at startup.

Modify the initscript so that corrupt non-persistent databases are
moved aside to a backup.  If the number of backups for a particular
database exceeds $CTDB_MAX_CORRUPT_DB_BACKUPS (default 10) then the
oldest excess backups are garbage collected.

Abstracts from and cleans up the code for checking persistent
databases.

Logging of related messages is done to syslog or a log file as
specified.

Signed-off-by: Martin Schwenke <martin@meltin.net>

(This used to be ctdb commit 00cd75595685dae829758abf1a4cb644af7ed50e)

1 files changed, 93 insertions, 63 deletions

diff --git a/ctdb/config/ctdb.init b/ctdb/config/ctdb.init
index 0b863447b5..0a4d29b234 100755
--- a/ctdb/config/ctdb.init
+++ b/ctdb/config/ctdb.init
@@ -107,85 +107,112 @@ build_ctdb_options () {
     maybe_set "--max-persistent-check-errors" "$CTDB_MAX_PERSISTENT_CHECK_ERRORS"
 }
 
-check_tdb () {
-	local PDBASE=$1
-
-	test x"$TDBTOOL_HAS_CHECK" = x"1" && {
-		#
-		# Note tdbtool always exits with 0
-		#
-		local OK=`tdbtool $PDBASE check | grep "Database integrity is OK" | wc -l`
-		test x"$OK" = x"1" || {
-			return 1;
-		}
-
-		return 0;
-	}
-
-	tdbdump $PDBASE >/dev/null 2>/dev/null || {
-		return $?;
-	}
-
-	return 0;
-}
-
-check_persistent_databases () {
-    PERSISTENT_DB_DIR="${CTDB_DBDIR:-/var/ctdb}/persistent"
-    mkdir -p $PERSISTENT_DB_DIR 2>/dev/null
-    local ERRCOUNT=$CTDB_MAX_PERSISTENT_CHECK_ERRORS
+# Log given message or stdin to either syslog or a CTDB log file
+do_log ()
+{
+    if [ "$CTDB_SYSLOG" = "yes" -o \
+	"${CTDB_OPTIONS#*--syslog}" != "$CTDB_OPTIONS" ] ; then
 
-    test -z "$ERRCOUNT" && {
-	ERRCOUNT="0"
-    }
-    test x"$ERRCOUNT" != x"0" && {
-	return 0;
-    }
-
-    if test -x /usr/bin/tdbtool ; then
-        HAVE_TDBTOOL=1
+	logger -t "ctdb.init" "$@"
     else
-        HAVE_TDBTOOL=0
+	_l="${CTDB_LOGFILE:-/var/log/log.ctdb}"
+	{
+	    date
+	    if [ -n "$*" ] ; then
+		echo "$*"
+	    else
+		cat
+	    fi
+	} >>"$_l"
     fi
+}
 
-    if test x"$HAVE_TDBTOOL" = x"1" ; then
-        TDBTOOL_HAS_CHECK=`echo "help" | /usr/bin/tdbtool | grep check | wc -l`
+select_tdb_checker ()
+{
+    # Find the best TDB consistency check available.
+    use_tdb_tool_check=false
+    if [ -x /usr/bin/tdbtool ] && \
+	echo "help" | /usr/bin/tdbtool | grep -q check ; then
+
+	use_tdb_tool_check=true
+    elif [ -x /usr/bin/tdbtool -a -x /usr/bin/tdbdump ] ; then
+	    do_log <<EOF
+WARNING: The installed 'tdbtool' does not offer the 'check' subcommand.
+ Using 'tdbdump' for database checks.
+ Consider updating 'tdbtool' for better checks!
+EOF
+    elif [ -x /usr/bin/tdbdump ] ; then
+	do_log <<EOF
+WARNING: 'tdbtool' is not available.
+ Using 'tdbdump' to check the databases.
+ Consider installing a recent 'tdbtool' for better checks!
+EOF
     else
-        TDBTOOL_HAS_CHECK=0
+	do_log <<EOF
+WARNING: Cannot check databases since neither
+ 'tdbdump' nor 'tdbtool check' is available.
+ Consider installing tdbtool or at least tdbdump!
+EOF
+        return 1
     fi
+}
+
+check_tdb ()
+{
+    _db="$1"
 
-    if test -x /usr/bin/tdbdump ; then
-        HAVE_TDBDUMP=1
+    if $use_tdb_tool_check ; then
+	# tdbtool always exits with 0  :-(
+	if tdbtool "$_db" check 2>/dev/null |
+	    grep -q "Database integrity is OK" ; then
+	    return 0
+	else
+	    return 1
+	fi
     else
-        HAVE_TDBDUMP=0
+	tdbdump "$_db" >/dev/null 2>/dev/null
+	return $?
     fi
+}
 
-    if test x"$HAVE_TDBDUMP" = x"0" -a x"$TDBTOOL_HAS_CHECK" = x"0" ; then
-        echo "WARNING: Cannot check persistent databases since"
-        echo "neither 'tdbdump' nor 'tdbtool check' is available."
-        echo "Consider installing tdbtool or at least tdbdump!"
-        return 0
-    fi
+check_persistent_databases ()
+{
+    _dir="${CTDB_DBDIR_PERSISTENT:-${CTDB_DBDIR:-/var/ctdb}/persistent}"
+    mkdir -p "$_dir" 2>/dev/null
 
-    if test x"$HAVE_TDBDUMP" = x"1" -a x"$TDBTOOL_HAS_CHECK" = x"0" ; then
-        if test x"$HAVE_TDBTOOL" = x"0"; then
-            echo "WARNING: 'tdbtool' is not available. Using 'tdbdump' to"
-            echo "check the persistent databases."
-            echo "Consider installing a recent 'tdbtool' for better checks!"
-        else
-            echo "WARNING: The installed 'tdbtool' does not offer the 'check'"
-            echo "subcommand. Using 'tdbdump' for persistent database checks."
-            echo "Consider updating 'tdbtool' for better checks!"
-        fi
-    fi
+    [ "${CTDB_MAX_PERSISTENT_CHECK_ERRORS:-0}" = "0" ] || return 0
 
-    for PDBASE in `ls $PERSISTENT_DB_DIR/*.tdb.[0-9] 2>/dev/null`; do
-	check_tdb $PDBASE || {
-	    echo "Persistent database $PDBASE is corrupted! CTDB will not start."
+    for _db in $(ls "$_dir/"*.tdb.*[0-9] 2>/dev/null) ; do
+	check_tdb $_db || {
+	    do_log "Persistent database $_db is corrupted! CTDB will not start."
 	    return 1
 	}
     done
 }
 
+check_non_persistent_databases ()
+{
+    _dir="${CTDB_DBDIR:-/var/ctdb}"
+    mkdir -p "$_dir" 2>/dev/null
+
+    for _db in $(ls "${_dir}/"*.tdb.*[0-9] 2>/dev/null) ; do
+	check_tdb $_db || {
+	    _backup="${_db}.$(date +'%Y%m%d.%H%M%S.%N').corrupt"
+	    do_log <<EOF
+WARNING: database ${_db} is corrupted.
+ Moving to backup ${_backup} for later analysis.
+EOF
+	    mv "$_db" "$_backup"
+
+	    # Now remove excess backups
+	    ls -td "${_db}."*".corrupt" |
+	    tail -n +$((${CTDB_MAX_CORRUPT_DB_BACKUPS:-10} + 1)) |
+	    xargs rm -f
+	    
+	}
+    done
+}
+
 set_ctdb_variables () {
     # set any tunables from the config file
     set | grep ^CTDB_SET_ | cut -d_ -f3- | 
@@ -239,7 +266,10 @@ start() {
     # instance of ctdb got killed with -9 or similar
     drop_all_public_ips
 
-    check_persistent_databases || return $?
+    if select_tdb_checker ; then
+	check_persistent_databases || return $?
+	check_non_persistent_databases
+    fi
 
     if [ "$CTDB_SUPPRESS_COREFILE" = "yes" ]; then
 	ulimit -c 0