Subject: check file timestamp bug? - msg#00075

Previous Message by Date: click to view message preview

Re: Monit refuse to start/stop a java process

Hi,Sorry for distrubing you. I have find my problem. It was about the configuration of my JAVA program, wich need some parameters from my personnal environment, and not available in root mode. 2007/9/19, pierrick grasland <pierrick.grasland@xxxxxxxxx>: Hi,I'm testing monit for a project, and i'm trying to monit a java process, under a debian etch (monit 4.8.1)I'm currently using this configuration for my service : check process easip with pidfile /var/run/easip.pid        group easip_gr         start program = "/home/pierrick/easip/easip.sh start"        stop program = "/home/pierrick/easip/easip.sh stop" #       if 3 restarts within 5 cycles then timeout  # desactivated for test         depends on easip_filecheck file easip_file with path /home/pierrick/easip/easip.sh         group easip_gr        if failed permission 755 then unmonitorwith this script, in order to have a pidfile : #/bin/sheasipStart() {         echo $$ > /var/run/easip.pid        exec 2>&1 java -jar /home/pierrick/easip/start.jar }easipStop() {         kill -15 `cat /var/run/easip.pid`        echo "" >/var/run/easip.pid }case $1 in  start)        easipStart         ;; stop)         easipStop        ;;  restart)        easipStop        sleep 3         easipStart        ;;  *)        echo "usage: easip.sh {start|stop|restart}" ;; esacWith this configuration, I can check the status of my process, ie running or not, with the web interface. Process 'easip'  status                            running   monitoring status                 monitored  pid                               23418   parent pid                        19170  uptime                            1m   childrens                         0  memory kilobytes                  76484   memory kilobytes total            76484  memory percent                    3.6%   memory percent total              3.6%  cpu percent                       0.0%   cpu percent total                 0.0%  data collected                    Wed Sep 19 15:41:38 2007In order to verify my script (needed for pidfile), i monitor the file File 'easip_file'  status                            accessible   monitoring status                 monitored  permission                        755   uid                               0  gid                               1000   timestamp                         Wed Sep 19 15:10:48 2007   size                              389 B  data collected                    Wed Sep 19 15:41:38 2007So, as you can see, i have all access to my script and monit can check the pid. But, when i try to stop / start my process with web interface, monit don't pass these operations : Sep 19 15:47:08 localhost monit[23566]: 'easip' stop: /home/pierrick/easip/easip.sh Sep 19 15:47:29 localhost monit[23566]: 'easip' failed to stopSep 19 15:50:29 localhost monit[23566]: 'easip' process is not running Sep 19 15:50:29 localhost monit[23566]: 'easip' trying to restart Sep 19 15:50:29 localhost monit[23566]: monit: pidfile `/var/run/easip.pid' does not contain a valid pidnumberSep 19 15:50:29 localhost monit[23566]: monit: pidfile `/var/run/easip.pid' does not contain a valid pidnumber Sep 19 15:50:29 localhost monit[23566]: 'easip' start: /home/pierrick/easip/easip.sh Sep 19 15:50:29 localhost monit[23566]: monit: pidfile `/var/run/easip.pid' does not contain a valid pidnumberSep 19 15:50:29 localhost last message repeated 2 times Sep 19 15:50:29 localhost monit[23566]: 'easip' failed to start Sep 19 15:50:30 localhost monit[23566]: monit: pidfile `/var/run/easip.pid' does not contain a valid pidnumberMy script works fine and I don't have any clue about what to do next in order to make monit do his job. Thank you,-- Grasland Pierrick -- Grasland PierrickENSSAT - LSI 3 -- To unsubscribe: http://lists.nongnu.org/mailman/listinfo/monit-general

Next Message by Date: click to view message preview

monit+heartbeat and drbd

After read 'MONIT WITH HEARTBEAT' section of Monit documentation i make some little changes for '/etc/init.d/monit-node1' example script: #!/bin/bash # # script for starting/stopping all services for given clusters node # prog="/usr/local/bin/monit -g $1" start() { echo -n "Starting Monit services for group '$1':" $prog start all RETVAL=$? echo } stop() { echo -n "Stopping Monit services for group '$1':" $prog stop all RETVAL=$? echo } case "$#" in start) start ;; stop) stop ;; *) echo "Usage: $0 groupname {start|stop}" RETVAL=1 esac exit $RETVAL Now we can use this script on both nodes without any tunes and i called it 'monit-hb'. For use this form of script we must change its call from 'heartbeat: /etc/ha.d/haresources' as follow: node1 IPaddr::172.16.100.1 monit-hb::node1 node2 IPaddr::172.16.100.2 monit-hb::node2 ---------- I not understand why in section 'INIT SUPPORT' you tell that we must use '-I' parameters: mo:2345:respawn:/usr/local/bin/monit -Ic /etc/monitrc but in section 'MONIT WITH HEARTBEAT' you are start 'monit' in daemon mode: mo:2345:respawn:/usr/local/bin/monit -d 10 -c /etc/monitrc -g local Yet another question. '-d 10' option means that 'monit' daemon restart after 10 seconds or it is daemon poll interval (as 'set daemon' option)? ---------- I want to use also 'drbd' with 'monit+heartbeat' and found some info about this configuration in your maillist archive but without content of 'ha-fs' script that mention there. I send to you my variant and think that not will be bad if you add info about 'monit+heartbeat+drbd' into your documentation. My variant 'monitrc': check device fs with path /dev/drbd0 start program = "/etc/ha.d/resource.d/ha-fs start" stop program = "/etc/ha.d/resource.d/ha-fs stop" if failed permission 660 then unmonitor if failed uid root then unmonitor if failed gid root then unmonitor if space usage > 80% then alert if space usage > 99% then stop mode manual group node1 check process postgresql with pidfile /var/run/postgresql.pid start program = "/etc/init.d/postgresql start" stop program = "/etc/init.d/postgresql stop" depends fs alert foo@bar mode manual group node1 check process app-server with pidfile /var/run/app-server.pid start program = "/etc/init.d/app-server start" stop program = "/etc/init.d/app-server stop" depends postgresql alert foo@bar mode manual group node1 # # node2 services # check process asterisk with pidfile /var/apache/logs/httpd.pid start program = "/etc/init.d/apache start" stop program = "/etc/init.d/apache stop" alert foo@bar mode manual group node2 and content of '/etc/ha.d/resource.d/ha-fs' with my values: #!/bin/bash # # script for starting/stopping services underlies of 'fs' device # scrd="/etc/ha.d/resource.d" start() { echo -n "Starting 'fs' device services:" $scrd/drbddisk pgdisk start && \ $scrd/Filesystem /dev/drbd0 /mnt/cluster reiserfs start RETVAL=$? echo } stop() { echo -n "Stopping 'fs' device services:" $scrd/Filesystem /dev/drbd0 /mnt/cluster reiserfs stop $scrd/drbddisk pgdisk stop RETVAL=$? echo } case "$#" in start) start ;; stop) stop ;; *) echo "Usage: $0 {start|stop}" RETVAL=1 esac exit $RETVAL I don't test this configuration but will do it at near time. I will be much grateful to you for your opinion about my proposal and any fixes. Thank you very much. Aleksandr Shubnik --------- Процесс обучения индивидуален. Курсы иностранных языков 'Streamline' предлагают своим клиентам комфортное и эффективное обучение в группах V.I.P. численностью до 8-ми человек. http://www.str.by/ -- To unsubscribe: http://lists.nongnu.org/mailman/listinfo/monit-general

Previous Message by Thread: click to view message preview

custom check program

Hello list! I have one little question about 'monit' using. As i understand from documentation now i can't use my own program/script for use in 'monit' check procedure. Am i right or not? Thank you very much. Aleksandr Shubnik --------- Канцелярские и офисные товары на http://www.ofiston.by -- To unsubscribe: http://lists.nongnu.org/mailman/listinfo/monit-general

Next Message by Thread: click to view message preview

Re: check file timestamp bug?

Hi, I confirmed this behavior on monit 4.9 on Linux 2.6.x and monit 4.10.beta1 Using the original block from your email: # monit -v monit: Debug: Adding credentials for user 'ogg'. /etc/monitrc:247: Error: the executable does not exist '/usr/local/bfrt-monit/scripts/nodeControl' /etc/monitrc:247: Error: the path '/var/run/bfrt-monit/pid-files/testMat.pid' used in the TIMESTAMP statement does not exist 'warn' Using this: check file testMat_pidfile with path /var/run/monitbug.pid if changed timestamp 3 times within 12 cycles then exec "/usr/bin/yes" mode manual group testGroup # monit -V This is monit version 4.10-beta1 Copyright (C) 2000-2007 by the monit project group. All Rights Reserved. # which yes /usr/bin/yes # monit -v monit: Debug: Adding credentials for user 'ogg'. /etc/monitrc:247: Error: the path '/var/run/monitbug.pid' used in the TIMESTAMP statement does not exist '/usr/bin/yes' So the bug appears to still be there. One thing that monit -v doesn't seem to print is the version info... Alex Stewart wrote: Dear All, Am I missing something here or do people agree this is not desired behaviour (if so what is the mechanism for filing a bug report)? -Alex check file statement from monitrc file: ------------------------------------------------------------ check file testMat_pidfile with path /var/run/bfrt-monit/pid-files/testMat.pid if changed timestamp 3 times within 12 cycles then exec "/usr/local/bfrt-monit/scripts/nodeControl warn" mode manual group testGroup ------------------------------------------------------------ -- To unsubscribe: http://lists.nongnu.org/mailman/listinfo/monit-general