February 2008 - Xymon - Xymon mailman web

hobbit color-changing issue
by georgi.lalev＠zdv.uni-tuebingen.de 17 Mar '08

17 Mar '08

Hello, I have a small hobbit issue. I want hobbit to go red every time the word "/evicted/" shows up in /var/log/messages on some hosts. So the part in /hobbit-clients.cfg/ is as follows: HOST=%cn4[3-6]*-hp PROC "%syslogd" 1 1 red PROC "%/usr/sbin/nscd" 1 red PROC "%iscsid" 2 2 red LOAD 16.5 20.0 LOG %/var/log/messages %(?-i)evicted|Oops|ALERT|NOTICE COLOR=red LOG %/var/log/messages %(?-i)LustreError|WARNING|FAILED|kernel|failing|ILLEGAL COLOR=yellow The Part in client-local.cfg: [cn44-hp] log:/var/log/messages:10240 ignore MARK trigger ALERT trigger .*(dumping log to)|(evicted).* file:/var/log/messages The Issue: Every time I put the word "/evicted/" manually in the messages file, hobbit will go red for about 15 minutes. That's fine. If I put a second line with the word "/kernel/" in the messages file, a second (yellow) line appears in the mesgs column. That's also fine. Hobbit is still red, but stays that way only for about 15 minutes. After the 15 minutes, it changes to yellow and the "/evicted/" warning is gone. It stays yellow for another 15 minutes, and then it turns green. The messages file is really small (under 10 KByte) and not much is being written in it. That's why hobbit must stay red all the time!?! I searched for the number 900 in server/etc, but nothing matches (exept some proxy configuration). Does anyone have an Idea what or where the problem could be? Greetings Georgi

1 1

[hobbit] Sending Hobbit data from one server to another
by gumby3203＠gmail.com 10 Mar '08

10 Mar '08

Here's the situation. I have one Hobbit server that is on our DMZ segment, and another that is on our internal LAN segment. There are several hosts that are monitored from the DMZ hobbit server, which are unreachable (due to network policy) by the internal hobbit server. What would be the best way (if any) to merge the status of the DMZ hobbit to the internal hobbit? That is, I would like to create a sub-page on the internal hobbit server that shows essentially a mirror of the hobbit status page for the DMZ server.

10 18

ntpdate question
by Dirk.Kastens＠uni-osnabrueck.de 09 Mar '08

09 Mar '08

Hi, before updating our hobbit server to version 4.2.0, I tested the ntp configuration with the bb-ntp-1_4.sh script on the client nodes. When I now configure the server to use the ntp network test instead, I realize that this doesn't work with most of our RedHat clients. I get the response ntpdate -u -q -p 2 a.b.c.d server a.b.c.d, stratum 0, offset 0.000000, delay 0.00000 29 Aug 10:53:26 ntpdate[3509]: no server suitable for synchronization found The network test works with all AIX clients and with older RedHat clients. The manual page of ntpdate on a RedHat machine says, that "the ntpdate program is to be retired from this distribution". Is there another way of testing the ntp synchronisation from the server side or do I still have to use the bb-ntp.sh script? Best wishes, Dirk Kastens

9 10

alerts won't stop
by lebarber＠gmail.com 03 Mar '08

03 Mar '08

I have removed some machines from hobbit, taken all mention of them from the configuration files and issued the "drop" command using bb. I am still getting notified of their "purple" status every day, and it has been several weeks since I removed them from Hobbit. How can I stop these alerts from being generated? Thanks, Larry Barber

5 14

Hobbit in the News !
by s_aiello＠comcast.net 29 Feb '08

29 Feb '08

I got my Linux Magazine (http://www.linux-mag.com) March 2008 issue yesterday. And imagine my surprise in the DIY section that there is an article on Hobbit !! Here's to a tremendous piece of software actually getting recognition. Just figured I would share, ~Steve

4 4

bbcmd --area option
by s_aiello＠comcast.net 29 Feb '08

29 Feb '08

Should bbcmd honor the --area option ? per the hobbitserver.cfg: areas are invoked by using the "--area" option for all tools, or via the ENVAREA setting in the hobbitlaunch.cfg(5) file. Presently it doesn't seem to hobbit 4.2.0 (all-in-one patch): $ bin/bbcmd --area DB 2008-02-27 16:48:59 Using default environment file /export/home/bb4/client/etc/hobbitclient.cfg 2008-02-27 16:48:59 execvp() failed: No such file or directory If it did.. would make debugging ext script a touch easier for me. ~Steve

2 2

.sh files downloading
by mail＠johankooijman.com 29 Feb '08

29 Feb '08

Goodmorning all, I've just installed hobbit for about the 20th time. The former 19 tries were all succesfull. This time however, I have a problem with the webpages. The machine is a CentOS 5.1 machine, running Apache 2.2.3 (from standard repo). I've installed the RPM, but also tried compiling, Hobbit. The monitor itself works, it generates pages fine. The problem happens when I click one of the icons, for example the one that leads to http://hostname/hobbit-cgi/bb-hostsvc.sh?HOST=hostname&SERVICE=bbtest. My browser wants to download the file, instead of displaying it. I've copied the hobbit-apache.conf from the installdirectory directly into the apache conf.d directory. I can't find anything in that file related to this problem. Any help is much appreciated! -- Met vriendelijke groeten / With kind regards, Johan Kooijman T +31(0) 6 43 44 45 27 F +31(0) 76 201 1179 E mail at johankooijman.com

3 2

ipmi digestion script w/ graph?
by baugust＠stanford.edu 29 Feb '08

29 Feb '08

I don't want to reinvent the wheel, but if this wheel doesn't exist, I guess I'll go invent it. :) Does anyone have a script to take the system temperature data from the IPMI interface (say, on a Dell box), report it to Hobbit, and pass the data to RRD so we get lovely temperature graphs? -- --Ben Benjamin P. August System Administrator - VPUE Stanford University

4 6

Bug in bbgen coredumping
by tlewick＠tradebotsystems.com 29 Feb '08

29 Feb '08

I posted something on this earlier, but just now found the steps on how to report a bug. Details: Installed hobbit 4.2.0 on Suse10.1 server, installed fine. Later applied allinone patch, ran make, make install, and that succeeded, but when restarting hobbit I noticed bbgen was crashing. Here is the gbd output: gdb bin/bbgen.coredumping tmp/core GNU gdb 6.6 Copyright (C) 2006 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "x86_64-suse-linux"... Using host libthread_db library "/lib64/libthread_db.so.1". warning: core file may not match specified executable file. Reading symbols from /usr/lib64/libpcre.so.0...done. Loaded symbols for /usr/lib64/libpcre.so.0 Reading symbols from /lib64/libc.so.6...done. Loaded symbols for /lib64/libc.so.6 Reading symbols from /lib64/ld-linux-x86-64.so.2...done. Loaded symbols for /lib64/ld-linux-x86-64.so.2 Core was generated by `./bbgen.coredumping'. Program terminated with signal 6, Aborted. #0 0x00002b7342165b95 in raise () from /lib64/libc.so.6 (gdb) bt #0 0x00002b7342165b95 in raise () from /lib64/libc.so.6 #1 0x00002b7342166f90 in abort () from /lib64/libc.so.6 #2 0x000000000041f353 in sigsegv_handler (signum=<value optimized out>) at sig.c:57 #3 <signal handler called> #4 0x000000000040a82f in calc_pagecolors (phead=0x5417b0) at process.c:96 #5 0x0000000000402b1f in main (argc=1, argv=0x7fff68bb7dc8) at bbgen.c:594

1 0

hobbit-clients.cfg include directive
by s_aiello＠comcast.net 28 Feb '08

28 Feb '08

I was curious if the hobbit-clients.cfg honored include directives. The reason I ask, is that our standard for defining LA thresholds is based on the numbers of CPUs. I edited my client's hobbitclient-<os>.sh to add a new section, [cpucount]. I figured I could write a server side script that could process that data & set the correct threshold accordingly. I just don't want to write a script that writes directly to the etc/hobbitclients.cfg file, and prefer writing to an external file that I can then include. So far only my linux & sun servers have the cpucount section, and these are the commands I used: Linux: $GREP processor /proc/cpuinfo | $WC SunOS: /usr/sbin/psrinfo -p ~Steve

3 3