November 2012 - Xymon - Xymon mailman web

Duplicate email Recipients in Info page
by Don.Kuhlman＠schawk.com 10 May '13

10 May '13

Hi folks. Still working on cleaning up my xymon infrastructure. I believe I'm getting a better handle on things with my analysis.cfg and alerts.cfg settings now. I made some more changes to the analysis and alerts files to add other groups for emails. So now, I think I've duplicated a notify group in one or the other places. When I click on a certain host's info link on an application page, I'm seeing the same target recipient for things like conn, cpu, disk, memory, etc. What's the best way to troubleshoot my rules/settings to figure out where I'm setting the duplicated target recipients at? PS – I can send along the alerts and analysis files if that helps. Thanks much! Don K

5 13

xymon e-mail alerts show up as attachments instead of e-mail
by icepickjazz＠gmail.com 29 Mar '13

29 Mar '13

Not a wild thought. Yes you are correct. I was getting ready to post my findings. We recently moved our xymon server from fedora core 3 to SLES 11. Fedora used mail and SLES 11 mail is linked to mailx. The windows clients are sending carriage returns in the status ^M (carriage return). The problem occured on SuSe server but not on redhat server, because mailx expects input text to be in Unix format, with lines separated by *newline*(^J, \n) characters only. Non-Unix text files that use *carriage return* (^M, \r) characters in addition will be treated as binary data. Hence the reason we get notifications that show up as attachments. Also explains why AIX servers don't have the problem and the windows servers do. Thanks On Mon, Sep 19, 2011 at 9:36 PM, Scott Pfister <icepickjazz at gmail.com>wrote: > Not a wild thought. Yes you are correct. I was getting ready to post my > findings. We recently moved our xymon server from fedora core 3 to SLES > 11. Fedora used mail and SLES 11 mail is linked to mailx. The windows > clients are sending carriage returns in the status ^M (carriage return). > The problem occured on SuSe server but not on redhat server, because mailx > expects input text to be in Unix format, with lines separated by *newline*(^J, \n) characters only. Non-Unix text files that use > *carriage return* (^M, \r) characters in addition will be treated as > binary data. Hence the reason we get notifications that show up as > attachments. Also explains why AIX servers don't have the problem and the > windows servers do. > > Thanks > > > On Mon, Sep 19, 2011 at 8:08 PM, Ralph Mitchell <ralphmitchell at gmail.com>wrote: > >> Wild thought - could the Windows reports be turned into attachments in >> your email server and/or client because the line-ends are <cr><lf> instead >> of just <lf> ?? >> >> Ralph Mitchell >> >> >> On Mon, Sep 19, 2011 at 6:06 PM, Scott Pfister <icepickjazz at gmail.com>wrote: >> >>> Good day, >>> >>> We are running xymon 4.2.3 and are experiencing some particularities with >>> e-mail alerts showing up as attachments like att00001.bin instead of the >>> normal message. The problem is sporadic. For example windows server throws >>> CPU alter. E-mail sent shows up as an attachment in the recipient e-mail. >>> AIX servers throw cpu alerts don't experience this problem. Client >>> shouldn't matter. Any tips for troubleshooting this issue? If we open the >>> attachment in the editor it shows what should be in the body of the email. >>> Hobbit server is running postfix. >>> >>> red Mon Sep 19 15:07:50 EDT 2011 [indyctx8.companyname.com] up: 10:12, 1 >>> users, 224 procs, load=100%, PhysicalMem: 3072MB(74%) >>> >>> >>> >>> Memory Statistics >>> Total Physical memory: 3220598784 bytes (3.00GB) >>> Available Physical memory: 850558976 bytes (812.00MB) >>> Total PageFile size: 4294967295 bytes (4.00GB) >>> Available PageFile size: 4294967295 bytes (4.00GB) >>> Total Virtual memory size: 2147352576 bytes (2.00GB) >>> Available Virtual memory size: 2106589184 bytes (1.96GB) >>> >>> Most active processes >>> 96.23% oexplore (0x26a4 [9892]) >>> 00.60% svchost (0xbf0 [3056]) >>> 00.32% lsass (0x23c [572]) >>> 00.26% oexplore (0x29c8 [10696]) >>> 00.25% oexplore (0x255c [9564]) >>> 00.21% Smc (0x3c4 [964]) >>> 00.20% SmcGui (0x3230 [12848]) >>> 00.19% oexplore (0x20dc [8412]) >>> 00.16% oexplore (0x14d0 [5328]) >>> 00.13% XTE (0xfc4 [4036]) >>> >>> See >>> http://hobbit.companyname.docm/hobbit-cgi/bb-hostsvc.sh?HOST=indyctx8&SERVI… >>> >>> >>> thanks, >>> >>> Scott >>> >>> >>> >>> _______________________________________________ >>> Xymon mailing list >>> Xymon at xymon.com >>> http://lists.xymon.com/mailman/listinfo/xymon >>> >>> >> >

3 2

ISO documentation on flapping status?
by betsy.schwartz＠gmail.com 05 Feb '13

05 Feb '13

We've got a bunch of servers showing as "flapping". I understand that this is a new feature but haven't found the docs on how to tweak it. Where and how can this be changed? thanks Betsy

3 3

Custom step and heartbeats. The rrddefinitions.cfg
by cornmaster＠gmail.com 24 Dec '12

24 Dec '12

I created a test that reports back some file sizes from a Windows server. The files don't change much, so I set the test to run every 15 minutes. My rrd file has all NaN, even though I confirmed the data is reported using the lastupdate command in rrdtool. I understand that this is because the default step/heartbeat are configured for a 5 minute window of checks, and my 15 minute checkins fall outside of this range, and therefore aren't added to the file. These are my assumptions thus far, and if they are wrong, please correct me. So what I did was use rrdtool tune command to change the heartbeat of my 5 data types to 1000 seconds. Which should allow my 15 minute updates to get logged correctly. Not sure if that has worked yet, hasn't been 15 minutes. But I wanted to ask about the rrddefinitions.cfg file. I understand that I can change the step/heartbeat settings on RRD file creation using this file. But I can't understand part of the config. Here is the contents of that file: # 576 datapoints w/ 5 minute interval = 48 hours @ 5 min avg. RRA:AVERAGE:0.5:1:576 # 576 datapoints w/ 6*5 minute averaged = 12 days @ 30 min avg. RRA:AVERAGE:0.5:6:576 # 576 datapoints w/ 24*5 minute averaged = 48 days @ 2 hour avg. RRA:AVERAGE:0.5:24:576 # 576 datapoints w/ 288*5 minute averaged = 576 days @ 1 day avg. RRA:AVERAGE:0.5:288:576 I won't say I understand it all, but this is what I have deduced so far: - RRA:AVERAGE: - needs to be there. - 0.5: - Don't know what this is for. I assume it have something to do with the span of the step/heartbeat. - 1: or 6: or 24: etc... - The number of datapoints collected per graph interval. - 576 - The total datapoints saved. Can anyone provide some insight into this file? Thanks, Thomas

1 1

holiday as planned downtime
by Jurgen.Claessens＠cegeka.be 11 Dec '12

11 Dec '12

Hey all, I'm running xymon 4.3.7 and I have been trying to find a way so that the holidays are being showed as what would be the same for the alerting. For me this would mean that if it's a holiday they would show as planned downtime and not show up in the allnongreen window. The alerting itself works without problems, but I cannot find a way for the planned downtime act the same way. Can anyone point me in the right direction? This e-mail and all files transmitted as attachment(s) thereto are confidential and solely intended for the individual to whom or the organization to which they are addressed. If you received this e-mail by mistake, please notify Cegeka's Service Desk at cegeka.support at cegeka.be or call +32 (0)11 240 363. We thank you in advance. Cegeka hereby confirms that this message has been swept by Sophos for the presence of viruses.

2 3

MAPÉ - Monitoring App Page Extension
by cami＠hack.co.za 10 Dec '12

10 Dec '12

Howdy All, We're working on a new (animated) front-end for Xymon which allows you to have different views into your environment. It's called "MAPÉ" (pronounced: MAP-EE). The project was started because the default "Main View" / "Non-green" view is quite outdated and such a great monitoring system needs a new face. We would love to get some feedback on the current screenshots and the roadmap (preferably on the mailing list). Instead of subjecting everyone to screenshots, you can take a look at it: http://mape.hack.co.za The first milestone has been reached with a dynamic / animated tree and will be up for download once we got a bit of discussion going around where we should be headed. Regards, Cami

9 27

migrate server
by ejacobs＠thomaspublishing.com 08 Dec '12

08 Dec '12

Our current Xymon server is running on a vmware vm. The powers-that-be want it to run on a physical machine. Unfortunately there is no V2P converter. So I've set up a new Linux box. Looking for the best way to migrate from old server to new one -- Eric Jacobs Thomas Publishing Company Infrastructure and operations Information Technology Group Phone: 215-494-7312 Email: ejacobs at thomaspublishing.com

6 9

[Newbie] Smaller, lighter Xymon based program?
by kfc.android＠gmail.com 05 Dec '12

05 Dec '12

Hi guys, I'm very new to this Xymon-land, spent some time to install and configure Xymon on my virtual machine runs CentOS 6.3 Here's my thought, since I have very limited skill in this monitoring field, my target is only to have a program which can test the 'ping' only. Is that doable if I can use Xymon and disable some of the functions? Has anyone ever think about this before? Maybe some hint or guides, thanks! Btw, I'm working on this as my school project, hopefully I didn't break/offend anyone's work and copyright, if so, please let me know. Many thanks :D -- Best, James

9 13

mysql password file on centos 5, xymon client
by leon＠one.co.il 03 Dec '12

03 Dec '12

Hi, I have a question. Recently I've added a script to monitor the mysql replication. I got it off the internet: ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- #!/bin/bash ############################################################################################ test -e /usr/local/mysql/bin/mysql && PROG=/usr/local/mysql/bin/mysql test -e /usr/bin/mysql && PROG=/usr/bin/mysql MAXDRIFT=5 ############################################################################################ MYSQLCOLOR="" MYSQLLINE="" HOSTIP=127.0.0.1 IPADDR=127.0.0.1 MACHINE=`uname -n` MAXSEC=60 MYSQLUSER=root MYSQLPORT=3306 LINE=`echo "show slave status \G" | ${PROG} -h ${HOSTIP} -u ${MYSQLUSER} -P ${MYSQLPORT} 2>&1` MASTERPOS=`echo $LINE | grep Read_Master_Log_Pos | awk -F 'Read_Master_Log_Pos: ' '{print $2}' | awk '{print $1}'` SLAVEPOS=`echo $LINE | grep -i Exec_Master_Log_Pos | awk -F 'Exec_Master_Log_Pos: ' '{print $2}' | awk '{print $1}'` SLAVEUP=`echo $LINE | grep Running | wc -l` SLAVE_SECOND_BEHIND_MASTER=`echo $LINE | grep 'Seconds_Behind_Master:' | awk -F 'Seconds_Behind_Master: ' '{print $2}' | awk '{print $1}'` test $SLAVEUP -eq "1" && DRIFT=$((${MASTERPOS} - ${SLAVEPOS})) echo "${LINE}" | ${GREP} "Running: No" >/dev/null if test "$?" -eq "0" then MYSQLCOLOR="red" MYSQLLINE="`${DATE}` MYSQL Slave in error ${LINE}" elif test ${SLAVEUP} -ne "1" then MYSQLCOLOR="green" MYSQLLINE="`${DATE}` MYSQL Slave is Stopped ${LINE}" elif test ${SLAVE_SECOND_BEHIND_MASTER} -ge ${MAXSEC} then MYSQLCOLOR="yellow" MYSQLLINE="`${DATE}` MYSQL Slave is ${DRIFT} statements behind! ${LINE}" else MYSQLCOLOR="green" MYSQLLINE="`${DATE}` MYSQL Slave is ${DRIFT} statements behind! ${LINE}" fi #echo "${BB} ${BBDISP} status ${MACHINE}.mysqlslave ${MYSQLCOLOR} ${MYSQLLINE}" ${BB} ${BBDISP} "status ${MACHINE}.mysqlslave ${MYSQLCOLOR} ${MYSQLLINE}" ----------------- To make it working, you need to create the file : ~hobbit/.my.cnf [mysql] password=XXXXXXXX [mysqlclient] password= XXXXXXXX [mysqladmin] password= XXXXXXXX [mysqlshow] password= XXXXXXXX [mysqldump] password= XXXXXXXX ------------------------------------------- chown hobbit: ~hobbit/.my.cnf chmod 600 ~hobbit/.my.cnf Add the section in clientlaunch.cfg [mysqlslave] ENVFILE $HOBBITCLIENTHOME/etc/hobbitclient.cfg CMD $HOBBITCLIENTHOME/ext/mysqlslave.sh LOGFILE $HOBBITCLIENTHOME/logs/mysqlslave.log INTERVAL 5m Restart hobbit-client. ------------------------------------------------------------------------------------------------------------------------ as you can see, the author instructs to put the .my.cnf file with all the passwords in the ~hobbit/.my.cnf directory. While that went fine on the Ubuntu 12.04 server, it doesn't work on the CentOS 5 server. I got the xymon client.rpm from some website and it just doesn't recognize this file no matter where I put it. The version is 4.3.10. the client is installed in these directories: /etc/xymon-client /usr/libexec/xymon-client /usr/share/xymon-client I also noticed that there's no /var/lib/xymon-client directory. The Ubuntu version was installed via apt-get and still has "hobbit" naming. On CentOS client I renamed all the "hobbit" into "xymon". If I put the pass into the script itself - it works. But I really don't want to do that. Any ideas? Thanks, Lenny

4 5

Solaris 10 xymon SMF
by Rolands＠logicaltech.com.au 30 Nov '12

30 Nov '12

Hi, There is a minor issue with the Solaris 10 SMF for xymon by Galen Johnson that I improved a bit. Thanks Galen for the nice work :D At least it works fine for us. When you disable the xymon service it could wait for up to 5 min before disabling it because the iostat and vmstat are running. (300 min) I been trying some different approaches in the manifest but it doesn't really work properly. I had to change the default timeout (30) to 300 to cater for the 300min possible run time of iostat, vmstat. Otherwise it will go into maintenance mode. Today I had a different approach and changed the xymon.sh script instead, that worked really nice. In server/bin/xymon.sh I added these simple lines to the stop function. (works in Solaris 10) "stop") if test -s /usr/local/xymon/logs/xymonlaunch.pid then kill -TERM `cat /usr/local/xymon/logs/xymonlaunch.pid` echo "Xymon stopped" else echo "Xymon is not running" fi rm -f /usr/local/xymon/logs/xymonlaunch.pid # kill off the remaining xymon processes, ie vmstat iostat etc. sleep 2 pkill 'iostat|vmstat' -U xymon ;; Maybe it could help someone else out there. I'm running 4.3.10 - Roland

3 5