- Xymon - Xymon mailman web

start up script for automatic loading
by don.munyak＠gmail.com 20 Jan '07

20 Jan '07

Hello, I'd like to automate startup of hobbit at boot time. If I make a link from /usr/local/etc/rc.d/hobbit.sh to the startup script in /usr/local/...hobbit/server/hobbit.sh, how do I get it to start using the user account 'hobbit' ? Is there another preferred technic ? Thanks

2 3

Custom SMS Alert
by rleon＠racenter.com 19 Jan '07

19 Jan '07

How can I change my SMS alert to display just the top line of the regular email alert. For example I just want this to be sent via sms: yellow Fri Jan 19 12:48:14 CST 2007 - Filesystems NOT ok &yellow /f01 (83% used) has reached the WARNING level (83%) The end goal is to customize the alert to only send. - Filesystems NOT ok &yellow /f01 (83% used)

1 0

Errors with btest
by mprice＠hst.nasa.gov 19 Jan '07

19 Jan '07

hello, Some times I have hosts that will go off line and the "conn" connection goes red because they can not be pinged. Which is ok, and perfectly make sense. Well, the side affect is that the "bbtest" goes yellow from timeouts of the pings timed out. Here are the errors... Error output: Timeout waiting for data from child, killing it Timeout waiting for data from child, killing it Child process terminated with signal 15 How can I prevent the "bbtest" from reacting this way.. thanks, michael -- Michael A. Price Performance Network Engineering NASA/GSFC Code 440.8/LMB Greenbelt, Maryland 20770 Phone: 240-684-1356 Cell: 410-507-7476 e-mail: mprice at hst.nasa.gov ----------------------------------------------

1 0

[hobbit] Dependency sytem in Hobbit
by emarganne＠be.tiauto.com 19 Jan '07

19 Jan '07

Hello, Well yes indeed it could work well even though all the work necessary to build the right configuration file will be huge for a network such as ours. The idea, to explain everybody in the mailing list, would be to link the different devices together by the means of the combo test. They work as follow : BigTest1 = hostA.test && hostB.test BigTest2 = hostC.test || hostD.test BigTest3 = BigTest1 || BigTest2 To propagate correctly the informations through a network path, do as follow: A - 1 - 2 - 3 - F 1' = A (to remember the test done on A) 2' = 1' and 1 (if you want to join 2 you must at least having 1', A let's say and 1) 3' = 2' and 2 (same idea for 3) F' = 3' and 3 Another kind of topology: /- 1 -\ A /-- 2 -- F \-- 3 -/ 1' = A (useful to propagate surely the information) 2' = A 3' = A F = (1' and 1) || (2' and 2) || (3' and 3) Last kind of topology (full mesh): /- 1 -\ / | \ A /--- 2 -- F \ | / \-- 3 -/ 1' = A (useful to propagate surely the information) 2' = A 3' = A F = A and (1 || 2 || 3) Here it is. If you think about network topology I forgot, feel free to tell them (with the solution if available ;-)). Of course this does not solve directly the alerts flood problem. I will try to find out what could be possible with alerts based on the combo tests and no more on simple tests. If this solution can look pretty, at least to me, at the first read you have to remember that it is still a lot work, even with scripts doing (you have to write the scripts do not you ?). For the people with two ore more Hobbit servers, this is even worse because the views are most likely not the same. Thanks for reading and taking part, Etienne Marganne, TI Automotive -----Original Message----- From: Charles Goyard [mailto:cgoyard at cvf.fr] Sent: vendredi 19 janvier 2007 14:28 To: hobbit at hswn.dk Subject: Re: [hobbit] Dependency sytem in Hobbit Hi, try the bbcombotest to add HA logic and make the depend tag point to the aggregated dot. See bbcombotest.cfg -- Charles Goyard - cgoyard at cvf.fr - (+33) 1 45 38 01 31 To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk The information contained in this transmission may contain privileged and confidential information. It is intended only for the use of the person(s) named above. If you are not the intended recipient, you are hereby notified that any review, dissemination, distribution or duplication of this communication is strictly prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.

1 0

No commercial Hobbit support from me
by henrik＠hswn.dk 19 Jan '07

19 Jan '07

Some time ago I floated the idea of providing commercial support for Hobbit, and asked if anyone would be interested. A handful of you came forward and said you would be interested. However, the past few months have made it clear to me that committing myself to support Hobbit with guaranteed responsetimes etc. will not be possible. When there's a quiet period at my full-time job I'd have plenty of time to support Hobbit, but when the workload is the way it's been these past few months I would not be able to provide a reasonable level of support. And I don't want to do something poorly - then I'd rather not do it at all. So - commercial support from me won't happen at the moment. If anybody else would like to offer it, then I would very much encourage it. Regards, Henrik

3 3

How to add custom service?
by martin.lyberg＠gmail.com 19 Jan '07

19 Jan '07

Hi, Is it possible to add custom services within hobbitmon? I want to check that certain processes are running like Mailscanner, Postgrey etc. Is there any example available to set this up? Thank you. / Martin

2 2

How to define yellow AND red for a process count
by grsamy＠yahoo.com 19 Jan '07

19 Jan '07

How to define yellow and red for a process count? For example: http 0 = red http 1 to 15 = yellow http >15 red I think it is possible in BB by : hostname:http;<=15:http;>15 Thanks Ram ____________________________________________________________________________________ Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. http://farechase.yahoo.com/promo-generic-14795097

2 1

Snapshot gives empty Status pages
by lars.ebeling＠leopg9.no-ip.org 19 Jan '07

19 Jan '07

It has been like this for a long time. Regards Lars Ebeling http://leopg9.no-ip.org Hobbithobbyist "It is better to keep your mouth shut and appear stupid than to open it and remove all doubt." -- Mark Twain

1 0

After Server Restart get an error and all tests are purple
by stefan.freisler＠aspect-online.de 18 Jan '07

18 Jan '07

Hi, After Server Restart i get an error and all tests are purple. (except conn, http) i get the following error: 2007-01-18 08:28:37 Whoops ! bb failed to send message - Connection failed 2007-01-18 08:28:39 Could not connect to bbd at xxx.de:1984 - Connection refused what can i do? thx Stefan

5 8

Custom graphs
by Thomas.Kern＠hq.doe.gov 18 Jan '07

18 Jan '07

I have two non-linux systems that I have a specialized client on. On one of these systems, I have a special test for the CPU utilization of logical partitions. I used the NCV method of getting the data (LPARname:CPUvalue) into the RRD files supported by Hobbit. I have been able to add a graph to the status page for the LPAR column with a graph definition that lists each of the known LPARs in this system and the total value that is also sent. This has been working for several weeks. Now I am trying to add the second system. It has a different set of LPAR names. These LPAR names must be unique across our organization. I asked our Hobbit admin to update the hobbitgraph.cfg with a modified [lpar] definition that included the second system's LPAR names in the hope that RRD would simply say that those entries from the other system were 'Not Available' or simply not display anything for them. I had hoped that I could use ONE graph definition for the [lpar] column no matter which system was supplying the data. Both graphs had an error about one of the OTHER's DS entries not being present. Is it possible to have two graph definitions for the same column, such as [vm1.lpar] and [vmhost.lpar] versus a single definition [lpar]? Here are copies of the two status pages: Thu Jan 18 10:05:06 EST 2007 General Processor LPAR utilization within thresholds LEGACY:1.0 ZOSEPROD:26.7 ZOSPROD:2.4 Total:30.1 Thu Jan 18 10:05:13 EST 2007 IFL Processor LPAR utilization within thresholds IFLPROD:1.3 Total:1.3 Here is the combined graph definition: [lpar] TITLE z890 LPAR CPU Utilization YAXIS Percent FNPATTERN lpar.rrd -u 100 -l 0 DEF:LEGACY=lpar.rrd:LEGACY:AVERAGE DEF:IFLPROD=lpar.rrd:IFLPROD:AVERAGE DEF:ZOSPROD=lpar.rrd:ZOSPROD:AVERAGE DEF:IFLTEST=lpar.rrd:IFLTEST:AVERAGE DEF:ZOSEPROD=lpar.rrd:ZOSEPROD:AVERAGE DEF:Total=lpar.rrd:Total:AVERAGE LINE2:LEGACY#00cc00:Legacy LINE2:IFLPROD#00cc00:IFL-Production LINE2:ZOSPROD#ff0000:ZOSProd LINE2:IFLTEST#00cc00:IFL-Test LINE2:ZOSEPROD#0000ff:ZOSEProd LINE2:Total#ff00ff:Total COMMENT: \n GPRINT:LEGACY:LAST:LEGACY \: %5.1lf (cur) GPRINT:LEGACY:MAX: \: %5.1lf (max) GPRINT:LEGACY:MIN: \: %5.1lf (min) GPRINT:IFLPROD:AVERAGE: \: %5.1lf (avg) \n GPRINT:IFLPROD:LAST:LEGACY \: %5.1lf (cur) GPRINT:IFLPROD:MAX: \: %5.1lf (max) GPRINT:IFLPROD:MIN: \: %5.1lf (min) GPRINT:IFLPROD:AVERAGE: \: %5.1lf (avg) \n GPRINT:ZOSPROD:LAST:ZOSPROD \: %5.1lf (cur) GPRINT:ZOSPROD:MAX: \: %5.1lf (max) GPRINT:ZOSPROD:MIN: \: %5.1lf (min) GPRINT:ZOSPROD:AVERAGE: \: %5.1lf (avg) \n GPRINT:IFLTEST:LAST:LEGACY \: %5.1lf (cur) GPRINT:IFLTEST:MAX: \: %5.1lf (max) GPRINT:IFLTEST:MIN: \: %5.1lf (min) GPRINT:IFLTEST:AVERAGE: \: %5.1lf (avg) \n GPRINT:ZOSEPROD:LAST:ZOSEPROD \: %5.1lf (cur) GPRINT:ZOSEPROD:MAX: \: %5.1lf (max) GPRINT:ZOSEPROD:MIN: \: %5.1lf (min) GPRINT:ZOSEPROD:AVERAGE: \: %5.1lf (avg) \n GPRINT:Total:LAST:Total \: %5.1lf (cur) GPRINT:Total:MAX: \: %5.1lf (max) GPRINT:Total:MIN: \: %5.1lf (min) GPRINT:Total:AVERAGE: \: %5.1lf (avg) \n /Thomas Kern /301-903-2211

1 0