April 2017 - Xymon - Xymon mailman web

Anyone have a plugin for IBM Datapower?
by apocalysque＠yahoo.com 17 Apr '17

17 Apr '17

I wrote one that works with the REST API on ISAM appliances. I believe they’re pretty similar to the datapower appliances. There’s 2 parts to it and you’ll need to have node.js installed, as well as the jq - commandline JSON processor executable. The username and password for the appliance should go on the line immediately following the tag for this host. I found that putting the user/passwd on the same line as the host/tags made the information show up on the “info” page in xymon and I don’t think you want credentials published there. This particular test only checks to see if any of the junctions are “red” and reports a red status if so. The entire junction tree is reported along with green/red for good/bad junctions. I didn’t write in anything for junction specificity, though that’s probably coming soon. The first file I called isam-junctions.sh and the 2nd file I called isam-jct-parse.js I know it’s not exactly what you’re looking for but I hope it will help you to develop one that you can use for the datapower appliances. _____ #!/bin/bash # # get the status of junctions from ISAM appliances # HOSTTAG=junctions # What we put in hosts.cfg to trigger this test COLUMN=junctions # Name of the column, often same as tag in hosts.cfg CURLCOMMAND=/usr/bin/curl # where to find curl executable CURLOPTIONS="-skH 'Accept:application/json'" # set any preferred options here SEARCHTEXT='ServiceAccount' # set the content to look for here $XYMONHOME/bin/xymongrep $HOSTTAG | while read L do set $L # To get one line of output from xymongrep JCTUID=`grep -A 1 $2 /etc/xymon/hosts.cfg | egrep -io 'jctuid=[^[:space:]]+' | cut -f2 -d '='` JCTPWD=`grep -A 1 $2 /etc/xymon/hosts.cfg | egrep -io 'jctpwd=[^[:space:]]+' | cut -f2 -d '='` COLOR="green" # default to all is ok RESULT="0" HOSTIP="$1" MACHINEDOTS="$2" MACHINE=`echo $2 | sed -e's/\./,/g'` URL="https://$MACHINEDOTS/wga/widgets/health.json" # set the URL here RESPONSE=`$CURLCOMMAND $CURLOPTIONS --user "$JCTUID:$JCTPWD" -X GET $URL` # get the HTTP response from the URL HTML=`echo "$RESPONSE" | jq '.items[] | walk(if type == "array" then (sort_by(.label | ascii_downcase)) else . end)' | /etc/xymon/ext/isam-jct-parse.js` RESULT=`echo "$HTML" | grep -ic "color:red"` # check to see if search text is present in the response [[ "$RESULT" -ne "0" ]] && COLOR="red" $XYMON $XYMSRV "status $MACHINE.$COLUMN $COLOR `date` ${HTML}" done _____ #!/usr/bin/env node var parseJsonAsHTMLTree = function(jsn){ var result = ''; if(jsn.label){ switch(jsn.health){ case "0": result += '<ul style="list-style-type:none; color:green;"><li>' + jsn.label + "\n"; break; case "1": result += '<ul style="list-style-type:none; color:red;"><li>' + jsn.label + "\n"; break; } for(var i in jsn.children) result += parseJsonAsHTMLTree(jsn.children[i]); result += '</li></ul>'; } return result; } process.stdin.resume(); process.stdin.setEncoding('utf8'); process.stdin.on('data', function(data) { var result = parseJsonAsHTMLTree(JSON.parse(data)); process.stdout.write("<pre>" + result + "</pre>\n"); }); _____ From: Xymon [mailto:xymon-bounces at xymon.com] On Behalf Of McDonald, Daniel (Dan) Sent: Friday, March 3, 2017 7:29 AM To: xymon at xymon.com Subject: [Xymon] Anyone have a plugin for IBM Datapower? I’ve been asked to monitor an IBM Datapower XML gateway, and wanted to see if anyone else had a plugin to do it before I re-invented the wheel. They have a mib, so supposedly I can poll it with mrtg and feed statuses into xymon with bbmrtg.pl, but IBM claims that their web service has greater detail and information. So, if anyone has written a xymon plugin to poll a datapower webservice and come up with useful stats to display in xymon, I’d love to hear about it.

2 1

Alert transition yellow -> red with repeat problem
by john.r.rothlisberger＠accenture.com 11 Apr '17

11 Apr '17

This is a problem I have seen for a long long time and have actually brought it up on the list before. Xymon 4.3.21 Ubuntu 14.04LTS The problem I have (in this instance) a warning which is to be repeated daily suddenly goes red and triggers a single alert but doesn't repeat again until the repeat time of the warning has passed. >From the notification.log: Tue Mar 28 05:31:28 2017 ServerA.disk (IP) disk_warn 1490697082 100 <- warning sets a repeat time of 1 day Tue Mar 28 05:53:32 2017 ServerA.disk (IP) disk_alert 1490698405 100 <- minutes later it goes red (red repeat time is 15 minutes but no further alerts are generated) Next alert comes out 1 day after the above warning: Wed Mar 29 05:31:31 2017 ServerA.disk (IP) disk_alert 1490783486 100 <- 1 day after previous warning. This should have been repeated every 15 minutes. Wed Mar 29 05:46:40 2017 ServerA.disk (IP) disk_alert 1490784394 100 <- now, the repeat time is 15 minutes ... <- and continues every 15 minutes. This alert went a full 24 hours with only a single notification. :( I have seen this before (not always) where the repeat time in a warning overrides a follow up alert until the warning repeat time has expired. Alert rules: SCRIPT /home/xymon/server/ext/pg/exwarn_SCRIPT disk_warn DURATION>30 REPEAT=1d COLOR=yellow SERVICE=disk FORMAT=TEXT UNMATCHED SCRIPT /home/xymon/server/ext/pg/exalert_SCRIPT disk_alert DURATION>20 REPEAT=15 COLOR=red SERVICE=disk FORMAT=TEXT UNMATCHED Ideas/thoughts? Thanks, John Upcoming PTO: 4/3 _____________________________________________________________________ John Rothlisberger IT Strategy, Infrastructure & Security - Technology Growth Platform TGP for Business Process Outsourcing Accenture 312.693.3136 office _____________________________________________________________________ ________________________________ This message is for the designated recipient only and may contain privileged, proprietary, or otherwise confidential information. If you have received it in error, please notify the sender immediately and delete the original. Any other use of the e-mail by you is prohibited. Where allowed by local law, electronic communications with Accenture and its affiliates, including e-mail and instant messaging (including content), may be scanned by our systems for the purposes of information security and assessment of internal compliance with Accenture policy. ______________________________________________________________________________________ www.accenture.com

1 0

Internal Server Error when clicking Acknowledge Alert only for one group of hosts
by vadud3＠gmail.com 05 Apr '17

05 Apr '17

(xymon)@xymon:~$ tail -1 /var/log/apache2/error.log [Wed Apr 05 14:15:17.305430 2017] [cgi:error] [pid 3744] [client 10.13.136.222:40560] End of script output before headers: acknowledge.sh, referer: https://xymon.example.net/xymon/groupA/groupA_syslog/groupA_syslog.html page groupA GroupA SERVERS subpage groupA_syslog GroupA Syslog Servers group-compress <H3><I>GorupA Syslog SERVERS</I></H3> 192.168.1.250 foo.example.net 192.168.2.250 bar.example.net If I am a layer up Acknowledge Alert works without giving any internal server error. Any suggestion on where to look? Thanks -- Asif Iqbal PGP Key: 0xE62693C5 KeyServer: pgp.mit.edu A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing?

2 1

False positives on RAM and DISK F19
by john＠executech.com 04 Apr '17

04 Apr '17

I realized that I didn’t give all the details. This only happens once or twice a week per server. It will go this way and then recover a few minutes later. Right now, everything is normal (green) but then it will happen again in a couple days. I’m getting the exact behavior on my 15 other Fedora 19 machines. John Tullis From: Xymon [mailto:xymon-bounces at xymon.com] On Behalf Of Bruce Ferrell Sent: Wednesday, March 22, 2017 10:05 AM To: xymon at xymon.com Subject: Re: [Xymon] False positives on RAM and DISK F19 On 3/22/17 8:21 AM, JT Tullis wrote: I can't seem to find a solution for a problem I'm having: I'm running xymon-client on several Fedora 19 servers. The client version is 4.3.28 release 1.fc19. There isn't an option to upgrade the servers right now from F19. Memory report showing over 100%: server:memory red [758975] red Wed Mar 22 08:56:44 MDT 2017 - Memory CRITICAL Memory Used Total Percentage &red Physical 1236M 123M 1004% &red Actual 9293M 123M 7555% &green Swap 2266M 2867M 79% Disk showing 0 units free even though that isn't the case: server:disk red [208517] red Wed Mar 22 07:42:39 MDT 2017 - Filesystems NOT ok &red 11933080 4623272 73% / (0 units free) has reached the PANIC level (512 units) &red 636 6149980 1% /dev/shm (0 units free) has reached the PANIC level (512 units) &red 1016 6149600 1% /run (0 units free) has reached the PANIC level (512 units) &red 0 6150616 0% /sys/fs/cgroup (0 units free) has reached the PANIC level (512 units) &red 264948 5885668 5% /tmp (0 units free) has reached the PANIC level (512 units) &red 117488 62417 66% /boot (0 units free) has reached the PANIC level (512 units) Filesystem 1024-blocks Used A ail]ble Capacity Mounted on /dev/mapper/fedora_pvapp05-root 17411420 11933080 4623272 73% / tmpfs 6150616 636 6149980 1% /dev/shm tmpfs 6150616 1016 6149600 1% /run tmpfs 6150616 0 6150616 0% /sys/fs/cgroup tmpfs 6150616 264948 5885668 5% /tmp /dev/xvda1 194241 117488 62417 66% /boot Does anyone know if there is a patch for this or has anyone successfully run a newer version of the client on an older OS? Thanks, John _______________________________________________ Xymon mailing list Xymon at xymon.com<mailto:Xymon at xymon.com> http://lists.xymon.com/mailman/listinfo/xymon John, This *may* not by a xymon issue. I'm seeing similar things with my systems. They are reporting bogus info to my xymon and xymon reports ( and alarms ) on that. what do free and df report at the command line of the effected systems?

2 1

Help with jmxstat
by xymon＠asitshouldbe.com 01 Apr '17

01 Apr '17

Thanks, Martin. I'll check out Jolokia. I've started experimenting with jboss-cli.sh to gather some of the more critical stats. I've got that feeding into Xymon now. The next step is to get graphing working. Ron Martin Lenko <lenko99 at gmail.com> wrote .. > Hi, > jboss doesn't use standard jmx protocol but it's own remoting protocol. > I remember working around it by taking jboss-client.jar from jboss and > rebuilding jmxsh.jar. It also required a bit of a code change to get jmxsh > use it. At the end I abandoned the idea of using this approach because the > remoting client seemed to have a memory leak, using up all the heap space - > basically the monitoring was killing jboss apps. > > Have a look at Jolokia https://jolokia.org/ > It should allow you to query JMX over HTTP. > > Regards, > Martin > > On 29 March 2017 at 23:14, <xymon at asitshouldbe.com> wrote: > > > I'm trying to set up jmxstat from xymonton.org to monitor information > > about JBoss 7.1 on 64-bit Oracle Linux 6.7 with Java 1.7.0_75. I have > > followed the instructions for both the client and server side. The jmxstat > > test appears on the Xymon server and is updating every 5 minutes as > > expected. > > > > I'm continuously getting the error: > > > > myclient.mydomain.com JMX - connection failed for JVM > > > > I don't know enough about Java or JBoss to know where to start looking. > > > > The instructions for jmxstat say this: > > > > "The JVM will need to be restarted with additional options that enable the > > JMX module and define the port & credentials." > > > > What additional options are needed and where are these configured? > > > > The admin port is 9999 and netstat shows that it's listening. > > > > I did use the add-user.sh script to create the username and password in > > the mgtm-users.properties file for JBoss. > > > > Any suggestions? > > > > Thanks. > > > > > > _______________________________________________ > > Xymon mailing list > > Xymon at xymon.com > > http://lists.xymon.com/mailman/listinfo/xymon > > > >

1 0

Help with jmxstat
by lenko99＠gmail.com 01 Apr '17

01 Apr '17

Hi, jboss doesn't use standard jmx protocol but it's own remoting protocol. I remember working around it by taking jboss-client.jar from jboss and rebuilding jmxsh.jar. It also required a bit of a code change to get jmxsh use it. At the end I abandoned the idea of using this approach because the remoting client seemed to have a memory leak, using up all the heap space - basically the monitoring was killing jboss apps. Have a look at Jolokia https://jolokia.org/ It should allow you to query JMX over HTTP. Regards, Martin On 29 March 2017 at 23:14, <xymon at asitshouldbe.com> wrote: > I'm trying to set up jmxstat from xymonton.org to monitor information > about JBoss 7.1 on 64-bit Oracle Linux 6.7 with Java 1.7.0_75. I have > followed the instructions for both the client and server side. The jmxstat > test appears on the Xymon server and is updating every 5 minutes as > expected. > > I'm continuously getting the error: > > myclient.mydomain.com JMX - connection failed for JVM > > I don't know enough about Java or JBoss to know where to start looking. > > The instructions for jmxstat say this: > > "The JVM will need to be restarted with additional options that enable the > JMX module and define the port & credentials." > > What additional options are needed and where are these configured? > > The admin port is 9999 and netstat shows that it's listening. > > I did use the add-user.sh script to create the username and password in > the mgtm-users.properties file for JBoss. > > Any suggestions? > > Thanks. > > > _______________________________________________ > Xymon mailing list > Xymon at xymon.com > http://lists.xymon.com/mailman/listinfo/xymon > >

1 0