August 2008 - Xymon - Xymon mailman web

diference betwen hobbit and bb
by mosimodo＠gmail.com 07 Aug '08

07 Aug '08

Hi Im working with hobbit and bb at the same time!!! but now im thinking of change. I ´d like to use only hobbit clients . My question is .. what is the diference betwen bb and hobbit? i think that using hobbit I will be able to configure many monitors using only the server part , insted the ext scripts. for example scripts in the server part like oracle, cpu, memory, disk, tomcat,.... On the other hand, my server is a BB one, I´ll need to change it? regards

1 1

Script limits
by forums＠triadbiz.rr.com 05 Aug '08

05 Aug '08

What are the limits of data that is sent to hobbit via a script? I have thrown together a perl script that grabs data from a file and then sends to hobbit in a table format. At first I was trying to send the table with a kinda of pretty html format so that when you goto a view source it was easily readable. Well that wasn't working cause the file wasn't getting updated in hobbit. I piped the output of the script to a text file and and did a line count on it. It had about 4000 lines in the file that was being sent to hobbit. So I took all the lines returns away and made it all one line and it gets to hobbit now. But I am wanting to get a little bit more in there with graphing and didn't know if I was needing to look at a line limit that hobbit is able to accept. -Josh

2 2

Hobbit Database
by htaute＠media24.com 05 Aug '08

05 Aug '08

Hi We want to reinstall Hobbit on a new server but would like to keep the existing database as we have a long period of information that we would like to keep(backup) and extend on to. Were are the database/information stored on the current server and can we migrate it to a new server. Is there also a way to backup the data periodicly? Thanks Henri Taute Media24 Content Production Tel: +27 21 4062495 Fax: +27 21 4063143 Cell: +28 82 8276124 E-mail: htaute at media24.co.za ____________________________________________________________________________ This email and its contents are subject to an email legal notice that can be viewed at: http://www.naspers.com/email/disclaimer.html. Should you be unable to access the link provided, please email us for a copy at Helpdesk at Media24.com. Hierdie e-pos en sy inhoud is onderhewig aan 'n regskennisgewing oor elektroniese pos wat gelees kan word by http://www.naspers.com/epos/vrywaring.html. 'n Afskrif kan aangevra word by Helpdesk at Media24.com. ____________________________________________________________________________

4 5

help wth alarms
by mosimodo＠gmail.com 05 Aug '08

05 Aug '08

HI evereyone, Im using hobbit since 4 months , i did some little ext . But now im working with the alerts , they dont do what i want. this is one line . HOST=burgos.cmt.es SERVICE=conn,cpu,cpu2,disk,memory MAIL salvarez at cmt.es DURATION>30m COLOR=red MAIL salvarez at cmt.es REPEAT=60 RECOVERED the theme is that hobbit is sending us yellow alerts to and i only have configure red alerts. And some times is sending an alert that is recovered at the same minute. some help? regards, and have a good time!

3 5

Hobbit code available through Subversion
by henrik＠hswn.dk 05 Aug '08

05 Aug '08

As I wrote last week, the Hobbit sourcecode has now been moved into the SourceForge code archive. It is available through Subversion. The repository uses this URL: https://hobbitmon.svn.sourceforge.net/svnroot/hobbitmon So to checkout the current code, you would do svn co https://hobbitmon.svn.sourceforge.net/svnroot/hobbitmon hobbitmon where the latter "hobbitmon" is the target directory on your local machine. There's also a browser for the source code archive, at http://hobbitmon.svn.sourceforge.net/viewvc/hobbitmon/ Regards, Henrik

1 0

"bad" checks in default line?
by eric.jacobs＠thomastechsolutions.com 04 Aug '08

04 Aug '08

Every morning around 3:45 something happens on hobbit or our network that causes many very brief connection and http checks to fail. We haven't figured out why yet. So, to keep people from being paged needlessly, I tried to put time-limited badconn and badhttp checks in default line (ie. badhttp-*-0340-0400:1:2:3). But it doesn't seem to be working. Do I need to do that for every entry? Eric Jacobs

2 1

Enable file status and proc's graph
by Jerome.Meyer＠baldata.ch 04 Aug '08

04 Aug '08

Hi all Ok, i'm going further with Hobbit! I've read documentations and configured one server and three clients... However, I've questions about some configuration : files status ************ I've configurated hobbit-clients.cfg and the client-local.cfg but never appears in Hobbit, what's wrong? hobbit-clients.cfg HOST=%proxy1.* UP 30min LOAD 6.0 12.0 DISK /var/log 85 95 DIR red size>500K TRACK FILE /usr/local/sbin/test.txt red size<100K TRACK FILE /usr/local/sbin/linux_logfile.log TRACK client-local.cfg [%proxy1.*] dir:/var/log file:/usr/local/sbin/text.txt file:/usr/local/sbin/linux_logfile.log Graph ***** One client's has a graph for disk columns but a another one has no graph! Where could be the problem because both have same configuration! How could I enable the graph in the procs columns? Thanks a lot for your help... Best regards bâldata - die IT & Telecom Services der IWB Jérôme M. Meyer SIN - Network Engineer Phone______ +41 (0)61.275.5266 Fax________ +41 (0)61.275.5509 Mobile_____ +41 (0)79.349.2315 E-Mail_____ jerome.meyer at baldata.ch <mailto:vorname.name at iwb.ch> www.baldata.ch <http://www.iwb.ch> _____________________ Partner von Swisspower www.swisspower.ch *******************************************************************

1 0

[hobbit] "Hobbit" name legal stuff; slight name change will h appen
by bewhite＠fellowes.com 03 Aug '08

03 Aug '08

Shhh.... Don't let the lawyers know! -----Original Message----- From: Martin Flemming [mailto:martin.flemming at desy.de] Sent: Thursday, July 31, 2008 12:56 PM To: hobbit at hswn.dk Subject: RE: [hobbit] "Hobbit" name legal stuff; slight name change will happen Hmm, it's something already there http://www.fastpath.it/products/palantir/ martin On Thu, 31 Jul 2008, Galen Johnson wrote: > I like it...it's exactly what the Palantir were for... > > =G= > > -----Original Message----- > From: Mike Eggleston [mailto:mikeegg1 at mac.com] > Sent: Thursday, July 31, 2008 1:34 PM > To: hobbit at hswn.dk > Subject: Re: [hobbit] "Hobbit" name legal stuff; slight name change > will happen > > On Thu, 31 Jul 2008, Kern, Thomas might have said: > >> How about Palantir ? > > No, 'precious'! > > Mike > > 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 ---------------------------------------------------- Note: The information contained in this message may be privileged and confidential and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. Fellowes, Inc. ----------------------------------------------------

1 0

[hobbit] SCRIPT using expect
by bewhite＠fellowes.com 03 Aug '08

03 Aug '08

Eric, I have done some scripts which act upon a change of status to attempt to correct the issue (none using expect). I find it is best to somehow track that you have attempted the change in the past and after X number of attempts (I usually only allow one automated fix attempt), forget the automated fix and just set the status to red, so I get paged. It is easy to do this by writing a small file with any needed information from the attempt to have the issue resolved via my monitoring script. If the problem goes away and the status will be reported as green, I delete the info file. If the problem is still present and the info file exists, I leave the file there and start the paging process in motion. .....Bruce -----Original Message----- From: Hubbard, Greg L [mailto:greg.hubbard at eds.com] Sent: Thursday, July 31, 2008 10:57 AM To: hobbit at hswn.dk Subject: RE: [hobbit] SCRIPT using expect Eric, You need to make sure that your script sets up its own environment prior to running because the Hobbit server may not have the same settings as what you have when you develop the script. This is the same issue that you would have to conquer if you were to put the script under cron. Secondly, you probably need to write a "director" script that is invoked through the alert rule, and let it examine the parameters in order to determine which "worker" script to call. I think this is the only way to run a different script when something has recovered. You also must take into account that you will get recovered messages even if no alert was sent. You should probably diagram out the whole thing from one end to the other -- e.g. Something breaks Hobbit test X goes to color C1 my script is run for test X, color C1 My script does something to resolve the problem Hobbit text X goes to color C2 My script is run for test X, color C1, resolved. Etc. GLH -----Original Message----- From: news [mailto:news at ger.gmane.org] On Behalf Of Eric Jacobs Sent: Thursday, July 31, 2008 9:53 AM To: hobbit at hswn.dk Subject: [hobbit] SCRIPT using expect I have been asked by those who must be obeyed to see if I can get Hobbit to make a configuration change on a switch if a certain connection test goes red. Been trying to do so via an "expect" script. Script runs fine when run manually but fails from within Hobbit. First attempt failed because script apparently didn't wait long enough after spawning a telnet session. So I slowed it down and it got session started but apparently didn't "send" password. Has anyone done anything like this? Even if I get it to work, a few other questions arise. 1. Will it continue to run script if connection failure continues? 2. If I turn "RECOVER" on, can I run an alternate script upon recovery? Eric Jacobs 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 ---------------------------------------------------- Note: The information contained in this message may be privileged and confidential and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. Fellowes, Inc. ----------------------------------------------------

1 0

client over-ride of tests processed on server, vmware esx
by seanj＠xyke.com 02 Aug '08

02 Aug '08

This post is about how to modify the [cpu] test on a per client basis. I am new to Hobbit. I have developed some extra client side tests that I have successfully used to log up to the server. My question is about the cpu column on the hobbit web display. I see that this test result is processed from the top output as sent by the client/hobbitclient-$OS.sh script On the client side, [top] sends the data to the server, which is then parsed and turned into what would look like a [cpu] test from the client. I tried sending my own [cpu] test to hobbit. But it would not show up. I changed the test name to 'xcpu' and I get an extra column. So I was sending it properly. If I have a couple hosts where I would like to send my own test results for existing columns that are processed on the server, where would I configure that? Is it even possible? I traced through hobbit server source for where some of this data is processed, hobbitd_client but couldn't immediately grok the configuration. I am developing custom tests for measuring cpu/memory and disk for vmware esx server. The problem with the cpu test is that top only esx only returns the load and processes from the host os, not the clients. We would like to get the stats for the whole machine. This is available through the `esxtop` program, but the outputs are very different. It would take a couple days to transform esxtop -b -n 1 > output.txt Into the same output as regular top. Currently I have some python code that parses the output of the esxtop command and supplies back the color and message to the server. I would like my tests _in place of_ the tests displayed on the server for my esx clients. Sean

1 0