July 2011 - Xymon - Xymon mailman web

xymon-4.3.3 compile error need help
by DGarry＠coquitlam.ca 27 Jul '11

27 Jul '11

did export LIBRTDEF=-lrt gmake[1]: Leaving directory `/home/xymon/xymon-4.3.3/build' CC="gcc" CFLAGS="-g -O2 -Wall -Wno-unused -D_REENTRANT -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -DLINUX -I. -I`pwd`/include -DCLIENTONLY=1 -DLOCALCLIENT=1 -I/usr/local/include" LDFLAGS="" RPATHOPT="-Wl,--rpath,/usr/local/lib" RRDDEF="" RRDINCDIR="" PCREINCDIR="-I/usr/local/include" NETLIBS="" RRDLIBS="" PCRELIBS="-L/usr/local/lib -lpcre" XYMONTOPDIR="/home/xymon" XYMONHOME="/home/xymon/client" XYMONVAR="" XYMONLOGDIR="" XYMONHOSTNAME="" XYMONHOSTIP="142.219.34.210" XYMONHOSTOS="" XYMONUSER="xymon" CGIDIR="" SECURECGIDIR="" XYMONHOSTURL="" XYMONCGIURL="" SECUREXYMONCGIURL="" MAILPROGRAM=""mail"" RUNTIMEDEFS="" INSTALLWWWDIR="/home/xymon/client/www" INSTALLETCDIR="/home/xymon/client/etc" gmake -C xymond client gmake[1]: Entering directory `/home/xymon/xymon-4.3.3/xymond' gcc -o ../client/xymond_client -Wl,--rpath,/usr/local/lib xymond_client.o xymond_worker.o xymond_buffer.o client_config.o ../lib/xymonclient.a -L/usr/local/lib -lpcre -lrt xymond_worker.o: In function `net_worker_run': /home/xymon/xymon-4.3.3/xymond/xymond_worker.c:292: undefined reference to `locator_init' /home/xymon/xymon-4.3.3/xymond/xymond_worker.c:298: undefined reference to `locator_register_server' /home/xymon/xymon-4.3.3/xymond/xymond_worker.c:312: undefined reference to `locator_serverdown' xymond_worker.o: In function `net_worker_heartbeat': /home/xymon/xymon-4.3.3/xymond/xymond_worker.c:80: undefined reference to `locator_serverup' xymond_worker.o: In function `net_worker_run': /home/xymon/xymon-4.3.3/xymond/xymond_worker.c:317: undefined reference to `locator_serverdown' collect2: ld returned 1 exit status gmake[1]: *** [../client/xymond_client] Error 1 gmake[1]: Leaving directory `/home/xymon/xymon-4.3.3/xymond' gmake: *** [xymond-client] Error 2 Dennis Garry ICT, City of Coquitlam

3 2

Xymon 4.3.3 svcstatus
by brian.scott4＠det.nsw.edu.au 27 Jul '11

27 Jul '11

Hi, I was just setting up xymon 4.3.3 on a 64 bit FreeBSD system. Everything went as expected until I drilled down on a test (svcstatus.sh) and reliably got a server error. To cut a long story short, the programs svcstatus.c, history.c and reportlog.c (in web/ directory) have each had protective 'basename' calls put around file names received as web parameters. The basename function requires a function prototype because it returned 'char *' rather than int. Casting an int to a pointer works in the 32bit world but not when a pointer is 64 bits. On FreeBSD at least, the appropriate declaration is in libgen.h. From a quick look around it appears to be the same on a few other systems. My solution is to add a #include <libgen.h> To each of the files listed above. I don't know if this is something that would work everywhere or if it is a job for configure to work out. Hope this helps someone, Brian ********************************************************************** This message is intended for the addressee named and may contain privileged information or confidential information or both. If you are not the intended recipient please delete it and notify the sender. **********************************************************************

2 1

Bug with Xymon Client Data Parsing?
by tlewick＠tradebotsystems.com 27 Jul '11

27 Jul '11

I posted a related message a couple weeks back. We continue to see periodic issues with hobbit tests having a problem parsing certain strings. I don't know if its happening on only things we've setup a regex to look for, or if those are just the ones we notice. I suspect both, as sometimes the client data received via the xymon server is garbled, and sometimes its clean. I've seen this in the following tests: disk cpu procs See below for an example of garbled data 9375 368 tr]debot 21:27:22 S 19 0.0 00:00:00 0. 5424 24] 6 -bash] However, when I click on the client data link on the webpage, to see what xymon is parsing, I see the following: 9375 tradebot 20 0 24556 5424 1720 S 0 0.0 0:00.17 bash So I've seen scenarios where the client data looks ok, as in above, and when the client data looks garbled. When last I posted, the suggested fixes involved increasing the values of MAXLINE, MAXMSG_STATUS, MAXMSG_CLIENT, and MAXMSG_DATA in hobbitserver.cfg, but I had already increased those much beyond the defaults due to previous truncation issues. We've got a few clients sending some very long PROCS and CPU messages. The xymon logs don't show anything. We are running Xymon 4.3.0-0.beta2 Is this a known bug with regex parsing?

2 1

two tests showing purple and clear
by Paul.A.Gallegos＠wellsfargo.com 26 Jul '11

26 Jul '11

Ok, I'm stumped. I'm running Xymon 4.3.2 on a group of 10 RedHat 5.4 x64 servers, 4 of which are DB servers. 3 out of the four DB servers are showing green, the fourth shows clear for msgs (it says "no data reported") and purple for processes (over 20 days since something was reported). However, a ps command shows that the processes it should be checking for exist, and I can tail the log to see things scrolling through. I reinstalled the client on that server and it still won't go green. The configuration for all four DB servers is the same, outside of one script in ext/ directory. What could possibly be the problem?

1 0

Spikes in activity.
by Neil.Simmonds＠express-gifts.co.uk 25 Jul '11

25 Jul '11

Hi all, I've looked through the documentation for Xymon and I can't seem to find any way to only have an icon colour change if a threshold has been passed for x number of consecutive tests. I'm aware that alerts allow this kind of thing but would really like the dashboard not to change colour if we just get a spike in disk, cpu or memory usage (amongst other things). Is there any way to do this? Regards, Neil Simmonds Operations Analyst Operations Support Group Express Gifts Ltd. neil.simmonds at express-gifts.co.uk Tel :- 01254 303092 Fax :- 01254 303100 Name & Registered Office: EXPRESS GIFTS LIMITED, 2 GREGORY ST, HYDE, CHESHIRE, ENGLAND, SK14 4TH, Company No. 00718151. Express Gifts Limited is authorised and regulated by the Financial Services Authority ------------ NOTE: This email and any information contained within or attached in a separate file is confidential and intended solely for the Individual to whom it is addressed. The information or data included is solely for the purpose indicated or previously agreed. Any information or data included with this e-mail remains the property of Findel PLC and the recipient will refrain from utilising the information for any purpose other than that indicated and upon request will destroy the information and remove it from their records. Any views or opinions presented are solely those of the author and do not necessarily represent those of Findel PLC. If you are not the intended recipient, be advised that you have received this email in error and that any use, dissemination, forwarding, printing, or copying of this email is strictly prohibited. No warranties or assurances are made in relation to the safety and content of this e-mail and any attachments. No liability is accepted for any consequences arising from it. Findel Plc reserves the right to monitor all e-mail communications through its internal and external networks. If you have received this email in error please notify our IT helpdesk on +44(0) 1254 303030

4 5

Status board script
by marc＠moremagic.com 22 Jul '11

22 Jul '11

Here is a script I wrote for displaying an actionable overview of alerts on a large format display - perhaps it could be of use. We currently use a wall-mounted Samsung 460UX-2 with a few iframes to display various sites. Thanks to pnixon for his starting version that did something similar (this one has about the same degree of polish :P). Installation involves the normal sticking it in your server/ext directory and adding an entry in etc/hobbitlaunch.cfg. I think it does rely on an led graphics addon. ~marc

1 0

HTTP Monitoring issue
by pnixon＠gmail.com 22 Jul '11

22 Jul '11

Hello, I have http monitoring setup using something similar to: 0.0.0.0 website # noconn http://www.patricknixon.com=178.102.132.142/ http://www.patricknixon.com=108.143.174.204/ The monitoring itself is working, but the rrd graphs are getting large chunks of NaNs (like hours at a time). Is there a better way to monitor this configuration? Thanks much! --Patrick

3 2

no client data received by xymon server every day during same hours.
by malcolm.hunter＠gmx.com 22 Jul '11

22 Jul '11

Hi Oliver, > ----- Original Message ----- > From: Oliver.Zhang at weirminerals.com > Sent: 07/22/11 03:09 AM > To: xymon at xymon.com > Subject: [Xymon] no client data received by xymon server every day during same hours. > We have a server running win 2008 R2 64-bit with bbwin0.12 installed. Every > work day during 10am-10pm, the client data from bbwin will not be received > by the server. How can I troubleshoot it? On other windows servers, some of > them have issue in getting the uptime data sent to the sever. From time to > time, the uptime will become purple. Could you please advise what to do? What's the uptime of your server? I'm not 100% sure, but there may be a problem with BBWin after the Windows server has been up for 500 days. Malcolm

1 0

Alerts do not seem to be triggering or otherwise executing my script
by chris.allen＠affinesystems.com 21 Jul '11

21 Jul '11

Seemingly the same problem as this guy: http://lists.xymon.com/archive/2008-June/019952.html I've crawled the hell out of the official Xymon documentation as well as the wikibook, I can't find anything that tells me what's going on. There is an alert that should have been triggering (It's covered by the time displayed in the log output below). The shutdown message is me running /etc/init.d/hobbit restart in an attempt to make it aware of the problem. (Xymon is currently red) Is the hobbitd_alert test/debug command supposed to actually execute the script/MAIL rule? Knowing this would help me a little, but not a whole lot as the alert doesn't seem to be otherwise triggering anyway. I am quite puzzled and unable to otherwise intelligently test or debug this, doubly disappointing is that I'm rather fond of this monitoring system and would rather not be forced to discard it because of a lack of documentation and abundance of inexplicable behavior. I have tested running my script standalone but with identical context to what would happen if hobbit executed it, and it worked. I also have debugging commands in the script to let me know if it gets executed at all. It never does during a --test nor naturally as it should be given the currently red status of one of my tests. Disregard that I am not utilizing $RCPT, I eliminated that to reduce the change it could be something unexpected with variable. It wasn't, the script just isn't getting executed. What am I missing? Log and command output follows this line: /etc/hobbit/scripts/alert.sh #!/bin/sh echo `set` > /tmp/envvars /data/software/aws/dist/current/sns/bin/sns-publish "arn:aws:sns:us-east-1:871321084716:ops" --subject "Server alert from Hobbit/Xymon" --message "$BBALPHAMSG" > /tmp/debug-alerts 2>&1 hobbit-alerts.cfg HOST=* COLOR=red SCRIPT /etc/hobbit/scripts/alert.sh "ops" REPEAT=1h tail -10 page.log 2011-07-21 14:15:01 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:17:01 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:17:10 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:19:10 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:20:02 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:20:02 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:20:13 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:22:13 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:22:45 Tried to down BOARDBUSY: Invalid argument 2011-07-21 14:22:45 Got a shutdown message 2011-07-21 14:22:45 Terminated by signal 15 2011-07-21 14:23:09 Peer not up, flushing message queue 2011-07-21 14:23:09 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:25:03 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:25:03 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:26:12 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:27:46 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:28:46 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:29:15 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:29:15 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg Running bbcmd hobbitd_alert --debug --test live proc (live is a real server/hostname and proc is a test criteria we have) returns: dev:/var/log/hobbit# bbcmd hobbitd_alert --debug --test live proc 2011-07-21 14:31:21 Using default environment file /usr/lib/hobbit/client/etc/hobbitserver.cfg 2011-07-21 14:31:21 Opening file /usr/lib/hobbit/server/etc/bb-hosts 2011-07-21 14:31:21 Opening file /usr/lib/hobbit/server/etc/hobbit-alerts.cfg 2011-07-21 14:31:21 Opening file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:31:21 Cannot open configuration file /usr/lib/hobbit/server/etc/hobbit-holidays.cfg 2011-07-21 14:31:21 send_alert live:proc state 0 00007255 2011-07-21 14:31:21 send_alert live:proc state Paging 00007255 2011-07-21 14:31:21 Matching host:service:page 'live:proc:' against rule line 122 00007255 2011-07-21 14:31:21 *** Match with 'HOST=* COLOR=red' *** 2011-07-21 14:31:21 Found a first matching rule 00007255 2011-07-21 14:31:21 Matching host:service:page 'live:proc:' against rule line 122 00007255 2011-07-21 14:31:21 *** Match with 'HOST=* COLOR=red' *** 2011-07-21 14:31:21 repeat live|proc|script|"ops" at 0 2011-07-21 14:31:21 Alert for live:proc to "ops" 00007255 2011-07-21 14:31:21 Script alert with command '/etc/hobbit/scripts/alert.sh' and recipient "ops" 2011-07-21 14:31:21 No more secondary matching rule dev:/var/log/hobbit# tail ./notifications.log Thu Jul 21 12:22:01 2011 nfs-1.procs (75.101.157.144) chris.allen at affinesystems.com[123] 1311276121 300 Thu Jul 21 12:52:31 2011 live.conn (75.101.158.206) engineeering at affinesystems.com[122] 1311277951 500 Thu Jul 21 12:52:31 2011 live.conn (75.101.158.206) chris.allen at affinesystems.com[123] 1311277951 500 Thu Jul 21 12:52:31 2011 nfs-1.procs (75.101.157.144) engineeering at affinesystems.com[122] 1311277951 300 Thu Jul 21 12:52:31 2011 nfs-1.procs (75.101.157.144) chris.allen at affinesystems.com[123] 1311277951 300 Thu Jul 21 13:23:01 2011 live.conn (75.101.158.206) ops at affinesystems.com 1311279781 500 Thu Jul 21 13:23:01 2011 nfs-1.procs (75.101.157.144) ops at affinesystems.com 1311279781 300 Thu Jul 21 13:50:01 2011 ads2.procs (68.168.100.16) arn:aws:sns:us-east-1:871321084716:ops 1311281397 300 Thu Jul 21 13:52:09 2011 ads2.procs (68.168.100.16) "arn:aws:sns:us-east-1:871321084716:ops" 1311281521 300 Thu Jul 21 13:59:58 2011 ads4.procs (68.168.100.18) "ops" 1311281995 300 dev:/var/log/hobbit# tail ./rrd-data.log 2011-07-21 13:42:27 Cache flush completed 2011-07-21 13:42:37 Peer not up, flushing message queue 2011-07-21 14:04:33 Tried to down BOARDBUSY: Invalid argument 2011-07-21 14:04:33 Shutting down, flushing cached updates to disk 2011-07-21 14:04:33 Cache flush completed 2011-07-21 14:04:42 Peer not up, flushing message queue 2011-07-21 14:22:45 Tried to down BOARDBUSY: Invalid argument 2011-07-21 14:22:45 Shutting down, flushing cached updates to disk 2011-07-21 14:22:45 Cache flush completed 2011-07-21 14:22:53 Peer not up, flushing message queue dev:/var/log/hobbit# tail ./history.log 2011-07-21 14:04:42 Peer not up, flushing message queue 2011-07-21 14:22:51 Peer not up, flushing message queue 2011-07-21 14:32:47 Will not update /var/lib/hobbit/hist/ads2.msgs - color unchanged (purple) 2011-07-21 14:32:47 Will not update /var/lib/hobbit/hist/ads2.ssh - color unchanged (purple) 2011-07-21 14:32:47 Will not update /var/lib/hobbit/hist/ads3.msgs - color unchanged (purple) 2011-07-21 14:32:47 Will not update /var/lib/hobbit/hist/ads4.msgs - color unchanged (purple) 2011-07-21 14:32:47 Will not update /var/lib/hobbit/hist/dev.files - color unchanged (purple) 2011-07-21 14:32:47 Will not update /var/lib/hobbit/hist/dev.ports - color unchanged (purple) 2011-07-21 14:32:47 Will not update /var/lib/hobbit/hist/live.msgs - color unchanged (clear) 2011-07-21 14:32:47 Will not update /var/lib/hobbit/hist/vcr1.msgs - color unchanged (purple)

1 0

Problem with long ipv6 address
by anders.liljegren＠uadm.uu.se 20 Jul '11

20 Jul '11

Hello! The command netstat under Linux truncates long ipv6 adresses. This has the effect that rules like the following does not work as expected: PORT LOCAL=2001:6b0:b:231:130:238:31:23:22 STATE=LISTEN MAX=1 "TEXT=SSHD v6" An ugly workaround is to truncate the string in the rule also: PORT LOCAL=2001:6b0:b:231:130:238:3:22 STATE=LISTEN MAX=1 "TEXT=SSHD v6" A better solution would be for xymon to use the flag "-T" on the netstat command under Linux. Other operative systems may need other flags or need no flag. Best regards -- Anders Liljegren Mail: IT och inköp, Uppsala universitet Phone: +46 18 4717751 Box 887 mailto:anders.liljegren at uadm.uu.se SE-751 08 UPPSALA http://www.anst.uu.se/andelilj Sweden

1 0