December 2007 - Xymon - Xymon mailman web

Hobbit Client as a Service on Solaris 10
by Vernon.Everett＠woodside.com.au 06 Dec '07

06 Dec '07

Hi all I have set up my Hobbit clients to run (and auto-start) as a service on Solaris 10. Thought I would share this with the list. (If I have reinvented the wheel, then apologies for cluttering the list, but I couldn't find anything like this anywhere else) T.J. Are you still looking after the Hobbit Wiki? If enough people find this guide useful, it might be good to add it to the Wiki. For those of you familiar with Solaris 10, you should know about services, but for some, adding new ones is a little tricky. To get Hobbit working as a sevice we need to do the following. Create the file /var/svc/manifest/application/hobbit.xml In it, put the following <?xml version="1.0"?> <!DOCTYPE service_bundle SYSTEM "/usr/share/lib/xml/dtd/service_bundle.dtd.1">  <service_bundle type='manifest' name='Hobbit-monitor:hobbit'> <service name='application/hobbit' type='service' version='1'> <dependency name='filesystem' grouping='require_all' restart_on='none' type='service'> <service_fmri value='svc:/system/filesystem/local'/> </dependency> <dependency name='multi-user-server' grouping='optional_all' type='service' restart_on='none'> <service_fmri value='svc:/milestone/multi-user-server' /> </dependency> <exec_method type='method' name='start' exec='/usr/lib/hobbit/client/runclient.sh start' timeout_seconds='10'> <method_context> <method_credential user='hobbit'/> </method_context> </exec_method> <exec_method type='method' name='stop' exec='/usr/lib/hobbit/client/runclient.sh stop' timeout_seconds='10'> <method_context> <method_credential user='hobbit' /> </method_context> </exec_method> <exec_method type='method' name='restart' exec='/usr/lib/hobbit/client/runclient.sh restart' timeout_seconds='10' > <method_context> <method_credential user='hobbit' /> </method_context> </exec_method> <instance name='default' enabled='true' /> <stability value='Unstable' /> <template> <common_name> <loctext xml:lang='C'> Hobbit Monitor Client </loctext> </common_name> </template> </service> </service_bundle> Take note of lines 37, 47 and 57, the lines that start "exec=". You may need to edit the path to your Hobbit start script. To avoid confusion or possible issues, shut down your hobbit client at this point using the runclient script. Now, as root, run the command # svccfg import /var/svc/manifest/application/hobbit.xml We should now have a service called hobbit. # svcs | grep hobbit online 9:23:05 svc:/application/hobbit:default (It will probably have gone online at this point) You can now treat it as you would a regular service. If it hasn't gone online, kick it off as normal. # svcadm enable hobbit It may be necessary to do a disable and then an enable, but that should get it going. And because we have set the default as enabled, the service should start automagically when you do a reboot. Confirm it's all good by doing a # ps -efa | grep hobbit All the usual scripts should be running. If you don't want it as a serivce anymore, as root run # svccfg delete hobbit. This will remove the service, and allow you to continue running it from the runclient script. Have fun. Regards Vernon NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments.

3 2

[hobbit] Strange pinging issue
by oliver_grube＠campbellsoup.com 06 Dec '07

06 Dec '07

An HTML attachment was scrubbed... URL: <http://lists.xymon.com/pipermail/xymon/attachments/20071206/e86f3824/attach…>

1 0

Strange pinging issue
by shawn.maschino＠sabic-ip.com 05 Dec '07

05 Dec '07

Hi all, we are seeing a strange fping issue with Hobbit and I was wondering if anyone may have seen the same. We have changed the IP addresses on a large number of systems we are monitoring due to moving them to a new location. We updated the bb-hosts file to have the new IPs, and DNS was also updated. However Hobbit occasionally falls back to testing the old IP for no apparent reason. This happens quite regularly as well, as seen here: Fri Nov 30 10:56:08 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:56:08_2007> 0:04:19 Fri Nov 30 10:48:31 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:48:31_2007> 0:07:37 Fri Nov 30 10:48:08 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:48:08_2007> 0:00:23 Fri Nov 30 10:42:27 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:42:27_2007> 0:05:41 Fri Nov 30 10:39:53 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:39:53_2007> 0:02:34 Fri Nov 30 10:36:49 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:36:49_2007> 0:03:04 Fri Nov 30 10:31:57 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:31:57_2007> 0:04:52 Fri Nov 30 10:25:27 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:25:27_2007> 0:06:30 Fri Nov 30 10:23:58 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:23:58_2007> 0:01:29 Fri Nov 30 10:19:19 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:19:19_2007> 0:04:39 Fri Nov 30 10:16:26 2007 <http://lacin002v.gep.ge.com/cgi-bin/bb-histlog.sh?HOST=SACIN053V&SERVIC E=conn&TIMEBUF=Fri_Nov_30_10:16:26_2007> 0:02:53 The conn status page when green shows the correct IP address, and when red shows the wrong (old) IP address. Nothing in the bb-hosts or DNS change during this time. I ran a "fping -AeC 1000 -i 5000 <host>" from the command line to see if fping was switching IPs, and it does not seem to be, even when Hobbit shows red, the fping from the command line still his the new/current IP. I've tried forcing a "testip" in bb-hosts to see if DNS was involved, and with that forced it still changes in Hobbit. Any suggestions would be much appreciated. Thanks, Shawn

3 7

Attn Henrik: [hobbit] problem with CLASS definition (PATCH)]
by jonescr＠cisco.com 05 Dec '07

05 Dec '07

Thus far I have gotten no acknowledgment that you are aware of this bug. I have a vested interest in seeing this patch be part of 4.3, as I have production servers utilizing this feature to handle automatic fail-overs. Happy Holidays! -Charles -------- Original Message -------- Subject: Re: [hobbit] problem with CLASS definition (PATCH) Date: Tue, 16 Oct 2007 16:14:01 -0700 From: Charles Jones <jonescr at cisco.com> Reply-To: hobbit at hswn.dk To: hobbit at hswn.dk References: <4706C711.4030201 at cisco.com> <470D1348.3090405 at cisco.com> <470D1781.5070606 at cisco.com> <470D1E8C.7050403 at cisco.com> <470D2138.2000705 at cisco.com> Henrik, Just making sure that you saw this bug that I found, and hopefully implement a fix (I'm sure you have a more elegant way to fix it than I did) for Hobbit 4.3 -Charles Charles Jones wrote: > Okay I came up with a workaround. The problem is that CONFIGCLASS gets > reset to a default value (in hobbitclient.cfg). So, I created another > variable, "CLASS". I then made a small mod to hobbitclient.sh that > checks to see if CLASS got set to something, and if it did, it sets > CONFIGCLASS to that value. Here are the changes: > > --- runclient.sh 2006-08-09 20:09:58.000000000 +0000 > +++ /tmp/runclient.sh 2007-10-10 18:55:00.000000000 +0000 > @@ -32,6 +32,7 @@ > ;; > --class=*) > CONFIGCLASS="`echo $1 | sed -e 's/--class=//' | tr > '[A-Z]' '[a-z]'`" > + CLASS=$CONFIGCLASS # Use a variable that doesn't get > reset > ;; > --help) > echo "Usage: $0 [--hostname=CLIENTNAME] > [--os=rhel3|linux22] [--class=CLASSNAME] start|stop > " > @@ -61,7 +62,7 @@ > > MACHINE="`echo $MACHINEDOTS | sed -e 's/\./,/g'`" > > -export MACHINE MACHINEDOTS BBOSTYPE BBOSSCRIPT HOBBITCLIENTHOME > CONFIGCLASS > +export MACHINE MACHINEDOTS BBOSTYPE BBOSSCRIPT HOBBITCLIENTHOME > CONFIGCLASS CLASS > > case "$CMD" in > "start") > @@ -128,6 +129,5 @@ > break; > > esac > - > exit 0 > > ----- > > --- hobbitclient.sh 2006-08-09 20:09:58.000000000 +0000 > +++ /tmp/hobbitclient.sh 2007-10-10 18:55:28.000000000 +0000 > @@ -48,6 +48,9 @@ > echo "@@client#1|0|127.0.0.1|$MACHINEDOTS|$BBOSTYPE" >> > $MSGTMPFILE > fi > > +if [ -n "$CLASS" ]; then #If CLASS was set to something, override the > default > + CONFIGCLASS=$CLASS > +fi > echo "client $MACHINE.$BBOSTYPE $CONFIGCLASS" >> $MSGTMPFILE > $BBHOME/bin/$BBOSSCRIPT >> $MSGTMPFILE > # logfiles > > > > Charles Jones wrote: >> I also just discovered that CONFIGCLASS gets converted to lowercase >> (which I should have seen plain as day in the sed command). However >> this is not the problem as the CONFIGCLASS in the client message is >> still using the default value of BBOSTYPE. I'm currently attempting >> to hack in a work around. >> >> -Charles >> >> Charles Jones wrote: >>> I did some poking around, and as far as I can tell, the problem is >>> due to hobbitclient.cfg overriding the CONFIGCLASS variable. >>> Here are some snippets from runclient.sh: >>> >>> --class=*) >>> CONFIGCLASS="`echo $1 | sed -e 's/--class=//' | tr >>> '[A-Z]' '[a-z]'`" >>> >>> export MACHINE MACHINEDOTS BBOSTYPE BBOSSCRIPT HOBBITCLIENTHOME >>> CONFIGCLASS >>> >>> hobbitclient.cfg contains this: >>> ./hobbitclient.cfg:CONFIGCLASS="$BBOSTYPE" # Default >>> configuration class for logfiles >>> >>> hobbitclient.sh uses the CONFIGCLASS variable when sending the >>> client message: >>> ./bin/hobbitclient.sh:echo "client $MACHINE.$BBOSTYPE $CONFIGCLASS" >>> >> $MSGTMPFILE >>> >>> Heres the actual client message header Im sending (not that >>> CONFIGCLASS is being set to "linux" instead of the specified class: >>> client host-app-23.linux linux >>> >>> -Charles >>> >>> >>> Charles Jones wrote: >>>> Anyone have any thoughts on this? I'm currently dead in the water >>>> without this working as it should. I tried emailing Henrik as well >>>> but I'm sure he is too busy to reply. >>>> >>>> -Charles >>>> >>>> Charles Jones wrote: >>>>> I have a host whose client was started via: "./runclient.sh >>>>> --class=APP_PRIMARY start" >>>>> >>>>> In the hobbit servers hobbit-clients.cfg, I have something similar >>>>> to: >>>>> >>>>> CLASS=APP_PRIMARY >>>>> PROC "someproc" >>>>> PROC "/some/other/proc" >>>>> .... >>>>> DEFAULT >>>>> PROC sshd >>>>> >>>>> The problem is that only the default proc "sshd" is being checked >>>>> on the host. The expected behavior is that the CLASS would be >>>>> used and other other procs checked instead. I know this has >>>>> nothing to do with the DEFAULT setting, as using HOST=hostname >>>>> works fine. I just cant get it to check the procs by CLASS. Is >>>>> there any way to debug this? Shouldn't the CLASS show up in the >>>>> "info" section? Do I need to put the class name in quotes on the >>>>> command line? >>>> >>>> To unsubscribe from the hobbit list, send an e-mail to >>>> hobbit-unsubscribe at hswn.dk >>> >>> To unsubscribe from the hobbit list, send an e-mail to >>> hobbit-unsubscribe at hswn.dk >> >> To unsubscribe from the hobbit list, send an e-mail to >> hobbit-unsubscribe at hswn.dk > > To unsubscribe from the hobbit list, send an e-mail to > hobbit-unsubscribe at hswn.dk To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

1 0

Can I subpage a subpage ?
by Keller_Liora＠emc.com 05 Dec '07

05 Dec '07

Hi, I would like to have a tree of more than just a page/subpage/subparent with titles. It looks like subparent can do titles. Thanks, Liora

2 1

How to create a graph from a personnal extension
by linuxmasterjedi＠free.fr 05 Dec '07

05 Dec '07

Hello once again ;) I'm making a custom script on my hobbit server. It will measure the time-response of an application. The script will return a RED or YELLOW status in the request time in under a define trigger. But is there a way to graph that on Hobbit like it's done for "CONN" or so ? Thanks by advance for any information or links for a tut! CU

3 4

[hobbit] Critical Systems page displaying "Internal Server Error"
by gumby3203＠gmail.com 04 Dec '07

04 Dec '07

It seems that the Critical Systems page has decided to stop working on me. Any time I load the page, it just displays an Internal Server Error. I enabled the --debug option, and in the apache error logs, I see the following message: [Sun Dec 02 11:11:46 2007] [error] [client x.x.x.x] malformed header from script. Bad header=HOST=: hobbit-nkview.sh "HOST" happens to be the first entry in the hobbit-nkview.cfg file. I have tried a couple of changes to the config file to see what would happen. I first tried deleting the entire contents of the configuration file, and still get the Internal Server Error. When this happens, I don't see any entry added to the apache error log. If I go back to the original hobbit-nkview.cfg file and just delete the first entry, the error above changes to "HOSTB", where HOSTB was the host below HOST, but is now the first. Essentially, I have determined that whatever is the first entry in the hobbit-nkview.cfg file is the host that appears as the "Bad header=" in the error above. The hobbit-nkview.cfg file has not changed in at least a week, and the Critical Systems page was working as of only 3 days ago. Any idea what's going on? I have seen the "Internal Server Error" problem come up before, but it would usually be fine on the next reload of the page. Also, the HOST that the error message is complaining about doesn't have any non-green alerts, so it appears to be some problem with the parsing of the configuration file. Is there additional logging I can turn on? I tried enabling the --debug for the bb-display module, but that just produced a huge amount of logging, but nothing that helped to narrow down the problem.

2 5

[hobbit] Sending alerts to remote system running qpage
by Keller_Liora＠emc.com 04 Dec '07

04 Dec '07

I just wanted to thank everyone for the input. Eric's script works really well, qpage.pl, and I configured the qpage/client on the hobbit server, changing the #define SNPP_SERVER in config.h file to point to our original qpage server. Thanks so much!!! Liora > _____________________________________________ > From: Keller, Liora > Sent: Monday, December 03, 2007 9:13 AM > To: 'hobbit at hswn.dk' > Subject: RE: [hobbit] Sending alerts to remote system running > qpage > > > It will save us so much time, to be able to forward the pages to the > already configured remote qpage. > > Can you please email me the qpage.pl script you are running on the > Hobbit server to forward the paging to the qpage server? > > Thanks, > Liora

1 0

Test subcribe
by michael.nemeth＠lmco.com 04 Dec '07

04 Dec '07

Sorry they bocked the news archive and all hobbit sites expect for sorceforge. Hope to get email working again! -- +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ | _p_ Mike Nemeth | ___| |_____ email(w) michael.nemeth at lmco.com Work: 856 359-1425 |><___________) | Home Page:http://www.geocities.com/mjnemeth/ | Work Page:http://scorpion/~nemethm/ | Work Page:http://phlccase/SWUNIX/mnemeth/ |++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

2 1

[hobbit] Sending alerts to remote system running qpage
by Keller_Liora＠emc.com 04 Dec '07

04 Dec '07

It will save us so much time, to be able to forward the pages to the already configured remote qpage. Can you please email me the qpage.pl script you are running on the Hobbit server to forward the paging to the qpage server? Thanks, Liora

3 2