Import 3ware-status

packaging/debian/3ware-status/3ware-status.3ware-statusd.init

@@ -0,0 +1,191 @@
+#! /bin/sh
+
+# Author: Petter Reinholdtsen <pere@hungry.com>
+# License: GNU General Public License v2 or later
+#
+### BEGIN INIT INFO
+# Provides:          3ware-statusd
+# Required-Start:    $remote_fs $syslog
+# Required-Stop:     $remote_fs $syslog
+# Default-Start:     2 3 4 5
+# Default-Stop:      0 1 6
+# Short-Description: Check 3ware-status values in the background.
+### END INIT INFO
+
+PATH=/sbin:/bin:/usr/sbin:/usr/bin
+DESC="3ware-status monitor"
+NAME=3ware-statusd
+PIDFILE=/var/run/$NAME.pid
+STATUSFILE=/var/run/$NAME.status
+SCRIPTNAME=/etc/init.d/$NAME
+
+
+# Do not touch you can configure this in /etc/default/3ware-statusd
+MAILTO=root   # Where to report problems
+PERIOD=600    # Seconds between each check    (default 10 minutes)
+REMIND=7200   # Seconds between each reminder (default 2 hours)
+RUN_DAEMON=yes
+
+[ -e /etc/default/3ware-statusd ] && . /etc/default/3ware-statusd
+
+# Gracefully exit if the package has been removed.
+test -x /usr/sbin/3ware-status || exit 0
+
+. /lib/lsb/init-functions
+[ -e /etc/default/rcS ] && . /etc/default/rcS
+
+if [ $RUN_DAEMON = "no" ] ; then
+	log_begin_msg "3ware-statusd is disabled in /etc/default/3ware-statusd, not starting."
+    	log_end_msg 0
+	exit 0
+fi
+
+check_3ware() {
+    echo $$ > $PIDFILE.new && mv $PIDFILE.new $PIDFILE
+    while true ; do
+        # Check ever $PERIOD seconds, send email on every status
+        # change and repeat ever $REMIND seconds if the raid is still
+        # bad.
+        if (3ware-status) |grep -q 'NOT OPTIMAL' ; then
+            BADRAID=true
+            logger -t 3ware-statusd "detected non-optimal RAID status"
+        else
+            BADRAID=false
+        fi
+        STATUSCHANGE=false
+        if [ true = "$BADRAID" ] ; then
+            # RAID not OK
+            (3ware-status) > $STATUSFILE.new
+            if [ ! -f $STATUSFILE ] ; then # RAID just became broken
+                STATUSCHANGE=true
+                mv $STATUSFILE.new $STATUSFILE
+            elif cmp -s $STATUSFILE $STATUSFILE.new ; then
+                # No change.  Should we send reminder?
+                LASTTIME="`stat -c '%Z' $STATUSFILE`"
+                NOW="`date +%s`"
+                SINCELAST="`expr $NOW - $LASTTIME`"
+                if [ $REMIND -le "$SINCELAST" ]; then
+                    # Time to send reminder
+                    STATUSCHANGE=true
+                    mv $STATUSFILE.new $STATUSFILE
+                else
+                    rm $STATUSFILE.new
+                fi
+            else
+                STATUSCHANGE=true
+                mv $STATUSFILE.new $STATUSFILE
+            fi
+        else
+            # RAID OK
+            if [ -f $STATUSFILE ] ; then
+                rm $STATUSFILE
+                STATUSCHANGE=true
+            fi
+        fi
+                
+        if [ true = "$STATUSCHANGE" ]; then
+            hostname="`uname -n`"
+            (
+                cat <<EOF 
+This is a RAID status update from 3ware-statusd.
+The 3ware-status program reports that one of the RAIDs changed state:
+
+EOF
+                if [ -f $STATUSFILE ] ; then
+                    cat $STATUSFILE
+                else
+                    (3ware-status)
+                fi
+                echo
+                echo "Report from $0 on $hostname"
+            ) | mail -s "info: 3ware raid status change on $hostname" $MAILTO
+        fi
+        sleep $PERIOD
+    done
+}
+
+check_daemon() {
+        # Let's check if there is a daemon which is really running and not timing out
+        DAEMON_RUN=`ps aux | grep "/etc/init.d/3ware-statusd check_3ware" | grep -v grep | grep -v daemon`
+        if [ -n "$DAEMON_RUN" ] ; then
+                return 1;
+        else
+                return 0;
+        fi
+}
+
+#
+#       Function that starts the daemon/service.
+#
+d_start() {
+    [ -f $PIDFILE ] && PID="`cat $PIDFILE`"
+    if [ "$PID" ] ; then
+        log_progress_msg "Daemon already running. Refusing to start another"
+        return 0
+    elif check_daemon ; then
+        # Use the daemon package to turn this script into a daemon
+        start-stop-daemon --start --quiet --pidfile $PIDFILE \
+                --oknodo --exec /usr/bin/daemon /usr/bin/daemon $SCRIPTNAME check_3ware
+	return 0
+    else
+        log_progress_msg "Daemon is already running. Refusing to start another"
+        return 0
+    fi
+}
+
+#
+#       Function that stops the daemon/service.
+#
+d_stop() {
+        if [ -f $PIDFILE ] ; then
+             	start-stop-daemon --stop --oknodo --quiet --pidfile $PIDFILE > /dev/null 2>&1 
+		rm -f $PIDFILE
+        else
+                log_progress_msg "Daemon is already stopped."
+                return 0
+        fi
+}
+
+# This is a workaround function which does not directly exit and
+# therefore can be used by a restart
+d_stop_by_restart() {
+        if [ -f $PIDFILE ] ; then
+                start-stop-daemon --oknodo --stop --quiet --pidfile $PIDFILE
+                rm -f $PIDFILE
+                log_end_msg 0
+        else
+                log_progress_msg "Daemon is already stopped."
+                log_end_msg 0
+        fi
+}
+
+case "$1" in
+  start)
+        echo -n ""
+        log_begin_msg "Starting $DESC: $NAME"
+        d_start ; CODE=$?
+        log_end_msg $CODE
+        ;;
+  stop)
+        log_begin_msg "Stopping $DESC: $NAME"
+        d_stop ; CODE=$?
+        log_end_msg $CODE
+        ;;
+  check_3ware)
+        check_3ware
+        ;;
+  restart|force-reload)
+        log_begin_msg "Restarting $DESC: $NAME"
+        d_stop_by_restart
+        sleep 1
+        d_start || CODE=$?
+        log_end_msg $CODE
+        ;;
+  *)
+        # echo "Usage: $SCRIPTNAME {start|stop|restart|reload|force-reload}" >&2
+        echo "Usage: $SCRIPTNAME {start|stop|restart|force-reload}" >&2
+        exit 1
+        ;;
+esac
+
+exit 0

packaging/debian/3ware-status/README.Debian

@@ -0,0 +1,23 @@
+README.Debian for 3ware-status package
+--------------------------------------
+
+Possible configuration:
+-----------------------
+
+If you want to change the default configuration of the init script you
+can create the file /etc/default/3ware-statusd and specify the following
+values.
+
+MAILTO=
+PERIOD=
+REMIND=
+
+Use MAILTO to specify which user shall get the status mails 
+(default is root).
+
+With PERIOD you can fix the seconds between each check.
+
+And REMIND specifies the seconds between each reminder.
+
+
+ -- Adam Cécile (Le_Vert) <gandalf@le-vert.net>  Wed, 10 Oct 2007 10:19:28 +0200 

packaging/debian/3ware-status/changelog

@@ -0,0 +1,26 @@
+3ware-status (0.4) unstable; urgency=low
+
+  * Now handle different tw-cli output for 9000 series cards.
+  * Will now fail if tw-cli binary cannot be found.
+  * Add --nagios parameter (run through nrpe).
+  * Import initscript changes from mpt-status 1.2.0-7 package.
+
+ -- Adam Cécile (Le_Vert) <gandalf@le-vert.net>  Thu, 02 Feb 2012 11:31:45 +0100
+
+3ware-status (0.3) unstable; urgency=low
+
+  * Do not fail with empty ports.
+
+ -- Adam Cécile (Le_Vert) <gandalf@le-vert.net>  Mon, 16 Aug 2010 09:27:55 +0200
+
+3ware-status (0.2) unstable; urgency=low
+
+  * Switch logical disk id and disk id.
+
+ -- Adam Cécile (Le_Vert) <gandalf@le-vert.net>  Wed, 10 Oct 2007 10:55:57 +0200
+
+3ware-status (0.1) unstable; urgency=low
+
+  * Intial release.
+
+ -- Adam Cécile (Le_Vert) <gandalf@le-vert.net>  Wed, 10 Oct 2007 10:19:28 +0200

packaging/debian/3ware-status/compat

@@ -0,0 +1 @@
+5

packaging/debian/3ware-status/control

@@ -0,0 +1,14 @@
+Source: 3ware-status
+Section: admin
+Priority: extra
+Maintainer: Adam Cécile (Le_Vert) <gandalf@le-vert.net>
+Build-Depends: debhelper (>= 5)
+Standards-Version: 3.9.2
+
+Package: 3ware-status
+Architecture: all
+Depends: ${shlibs:Depends}, ${misc:Depends}, python, lsb-base, daemon, bsd-mailx | mailx, tw-cli (>= 9.5.0)
+Description: get RAID status out of 3ware Eskalad HW RAID controllers
+ The 3ware-status software is a query tool to access the running
+ configuration and status of 3Ware HBAs. 3ware-status allows you to
+ monitor the health and status of your RAID setup.

packaging/debian/3ware-status/copyright

@@ -0,0 +1,26 @@
+This package was debianized by Adam Cécile (Le_Vert) <gandalf@le-vert.net> on
+Wed, 10 Oct 2007 10:19:28 +0200.
+
+It was downloaded from http://hwraid.le-vert.net
+
+Copyright Holder: 
+  Copyright (C) 2007-2012 Adam Cécile (Le_Vert) <gandalf@le-vert.net>
+
+License:
+
+This program is free software; you can redistribute it and/or
+modify it under the terms of the GNU General Public License, version 2
+as published by the Free Software Foundation.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANDABILITY of FITNESS FOR A PARTICULAR PURPOSE.
+See the GNU General Public License for details.
+
+
+On Debian GNU/Linux systems, the complete text of the GNU General Public
+License version 2 (or later) see `/usr/share/common-licenses/GPL-2'.
+
+This package is highly based on Steffen Joeris <white@debian.org> works done
+for the mpt-status package.
+Thanks a lot Steffen!

packaging/debian/3ware-status/dirs

@@ -0,0 +1 @@
+usr/sbin

packaging/debian/3ware-status/install

@@ -0,0 +1 @@
+3ware-status	usr/sbin

packaging/debian/3ware-status/rules

@@ -0,0 +1,42 @@
+#!/usr/bin/make -f
+
+# Uncomment this to turn on verbose mode.
+#export DH_VERBOSE=1
+
+build: build-arch build-indep
+build-arch:
+build-indep:
+
+clean:
+	dh_testdir
+	dh_testroot
+	dh_clean 
+
+install: build
+	dh_testdir
+	dh_testroot
+	dh_clean -k 
+	dh_installdirs
+
+
+binary-arch: build install
+binary-indep: build install
+	dh_testdir
+	dh_testroot
+	dh_installchangelogs 
+	dh_installdocs
+	dh_install
+	dh_installinit --name 3ware-statusd \
+		       --update-rcd-params="start 21 2 3 4 5 . stop 19 0 1 6 ."
+	dh_link
+	dh_strip
+	dh_compress
+	dh_fixperms
+	dh_installdeb
+	dh_shlibdeps
+	dh_gencontrol
+	dh_md5sums
+	dh_builddeb
+
+binary: binary-indep binary-arch
+.PHONY: build clean binary-indep binary-arch binary install

wrapper-scripts/3ware-status

@@ -0,0 +1,153 @@
+#!/usr/bin/python
+
+import os
+import re
+import sys
+
+binarypath = "/usr/sbin/tw-cli"
+
+if len(sys.argv) > 2:
+    print 'Usage: 3ware-status [--nagios]'
+    sys.exit(1)
+
+nagiosmode=False
+nagiosoutput=''
+nagiosgoodarray=0
+nagiosbadarray=0
+nagiosgooddisk=0
+nagiosbaddisk=0
+
+if len(sys.argv) > 1:
+    if sys.argv[1] == '--nagios':
+        nagiosmode=True
+    else:
+        print 'Usage: 3ware-status [--nagios]'
+        sys.exit(1)
+
+# Check binary exists (and +x), if not print an error message
+# or return UNKNOWN nagios error code
+if os.path.exists(binarypath) and os.access(binarypath, os.X_OK):
+        pass
+else:
+        if nagiosmode:
+                print 'UNKNOWN - Cannot find '+binarypath
+        else:
+                print 'Cannot find '+binarypath+'. Please install it.'
+        sys.exit(3)
+
+
+# Get command output
+def getOutput(cmd):
+    output = os.popen(cmd)
+    lines = []
+    for line in output:
+        if not re.match(r'^$',line.strip()):
+            lines.append(line.strip())
+    return lines
+ 
+def returnControllerList(output):
+    lines = []
+    for line in output:
+        if re.match(r'^c[0-9]+\s.*$',line.strip()):
+	    lines.append(line.split()[0])
+    return lines
+
+def returnDiskList(output):
+    lines = []
+    for line in output:
+        if re.match(r'^[p][0-9]+\s.*$',line.strip()):
+            # Shoudl contain something like 'u0'
+            # '-' means the drive doesn't belong to any array
+            # If is NOT PRESENT too, it just means this is an empty port
+            if not line.split()[2].strip() == '-' and not line.split()[1].strip() == 'NOT-PRESENT':
+                lines.append(line.split())
+    if fake_failure:
+        lines[0][1] = 'NOT PRESENT'
+    return lines
+
+def returnArrayList(output):
+    lines = []
+    for line in output:
+        if re.match(r'^[u][0-9]+\s.*$',line.strip()):
+            lines.append(line.split())
+    if fake_failure:
+        lines[0][2] = 'DEGRADED'
+    return lines
+
+# A way to force a fake failure
+fake_failure = False
+if os.path.exists('/root/fake_3ware_failure'):
+    fake_failure = True
+
+cmd = binarypath+' info'
+output = getOutput(cmd)
+controllerlist = returnControllerList(output)
+
+bad = False
+
+
+# List available controller
+if not nagiosmode:
+    print '-- Controller informations --'
+    print '-- ID | Model'
+    for controller in controllerlist:
+        cmd = binarypath+' info '+controller+' model'
+        model = getOutput(cmd)[0].split(' = ')[1].strip()
+        print controller+' | '+model
+    print ''
+
+# List arrays
+if not nagiosmode:
+    print '-- Arrays informations --'
+    print '-- ID\tType\tSize\tStatus'
+for controller in controllerlist:
+    cmd = binarypath+' info '+controller
+    output = getOutput(cmd)
+    arraylist = returnArrayList(output)
+    for array in arraylist:
+        type = array[1].replace('-','')
+        id = controller+array[0]
+        size = array[6].split('.')[0]+'G'
+        status = array[2]
+        if not status == 'OK':
+            bad = True
+            nagiosbadarray=nagiosbadarray+1
+        else:
+            nagiosgoodarray=nagiosgoodarray+1
+        if not nagiosmode:
+            print id+'\t'+type+'\t'+size+'\t'+status
+if not nagiosmode:
+    print ''
+
+# List disks
+if not nagiosmode:
+    print '-- Disks informations'
+    print '-- ID\tModel\t\t\tStatus'
+for controller in controllerlist:
+    cmd = binarypath+' info '+controller
+    output = getOutput(cmd)
+    disklist = returnDiskList(output)
+    for disk in disklist:
+        id = controller+disk[2]+disk[0]
+        cmd = binarypath+' info '+controller+' '+disk[0]+' model'
+        model = getOutput(cmd)[0].split(' = ')[1].strip()
+        cmd = binarypath+' info '+controller+' '+disk[0]+' status'
+        status = getOutput(cmd)[0].split(' = ')[1].strip()
+	if not status == 'OK':
+            bad = Truei
+            nagiosbaddisk=nagiosbaddisk+1
+        else:
+            nagiosgooddisk=nagiosgooddisk+1
+        if not nagiosmode:
+            print id+'\t'+model+'\t'+status
+
+if nagiosmode:
+    if bad:
+        print 'RAID ERROR - Arrays: OK:'+str(nagiosgoodarray)+' Bad:'+str(nagiosbadarray)+' - Disks: OK:'+str(nagiosgooddisk)+' Bad:'+str(nagiosbaddisk)
+        sys.exit(2)
+    else:
+        print 'RAID OK - Arrays: OK:'+str(nagiosgoodarray)+' Bad:'+str(nagiosbadarray)+' - Disks: OK:'+str(nagiosgooddisk)+' Bad:'+str(nagiosbaddisk)
+else:
+    if bad:
+        print '\nThere is at least one disk/array in a NOT OPTIMAL state.'
+        sys.exit(1)