April 2008 - Xymon - Xymon mailman web

cupsd
by stoner-lists＠comcast.net 04 Jun '08

04 Jun '08

For those who want to monitor a cups print server, here's a definition you can put in your bb-services file: [cupsd] send "GET /printers\r\n" expect "HTTP/1.1 200 OK" port 631 Alternatively, you could just use http://host:631/printers in your bb-hosts file. The cupsd is, for monitoring purposes, a web server. -Jeff "Fun with Users #1: ln -s /dev/random /etc/motd"

2 1

BBWin 0.11 Centralised Config problems
by Matthew.Roberts＠nt.gov.au 18 May '08

18 May '08

Hi All, I need a bit of assistance in setting up BBWin0.11 with Hobbit 4.3 in Centralised config mode. My problem lies with the eventlog monitoring of my Windows hosts. When running BBWin in local config mode, I can setup filtering so that only specific event log types get sent to the hobbit server and displayed on the msgs page. However when I configure BBWin to use a central config, I cant filter which event logs get sent to the hobbit server and displayed on the msgs page. As a result my msgs page is full of eventlog entries which I dont need to know about. Below is a copy of my BBWin.cfg and the client-local.cfg on the hobbit server <?xml version="1.0" encoding="utf-8" ?> <configuration> <bbwin> <setting name="bbdisplay" value="hobbit-server" />     <setting name="mode" value="central" /> <setting name="configclass" value="win32" /> <setting name="autoreload" value="true" /> <setting name="timer" value="5m" /> <load name="cpu" value="cpu.dll"/> <load name="disk" value="disk.dll"/> <load name="externals" value="externals.dll"/> <load name="memory" value="memory.dll"/> <load name="msgs" value="msgs.dll"/> <load name="procs" value="procs.dll"/> <load name="stats" value="stats.dll"/> <load name="svcs" value="svcs.dll"/> <load name="uptime" value="uptime.dll"/> <load name="who" value="who.dll"/> <setting name="loglevel" value="3" /> <setting name="logpath" value="C:\Program Files\BBWin\logs\BBWin.log"/>  <setting name="logreportfailure" value="false" /> </bbwin> client-local.cfg: (dcstest2 is the name of one of my hosts I am having problems with) [dcstest2] eventlog:security ignore Success eventlog:application ignore information eventlog:system ignore information Also something else I noticed is that the msgs page for my hobbit server which I also monitor displays the eventlog entries of my Win32 hosts. The hobbit server is running on SUSE Server 10.1 and I have even tried a complete rebuild in case there was a problem during the install phase. Assistance would be greatly appreciated. Thanks

4 6

Greetings and a question...
by baugust＠stanford.edu 15 May '08

15 May '08

First off, I love Hobbit. A lot. I've been running it as a test in our department for almost 9 months now and I have convinced the boss to not renew the contract for our proprietary monitoring package (from a certain other Scandinavian country) in favor of using Hobbit. I just need to see if I can put donations on my corporate card. :) But I do have one last question after all of that effusion. Is anyone out there running dbcheck.pl and had this error: "Wed Jan 23 10:21:40 2008:ERROR on sid_name: Timeout connecting to DBI:Oracle:sid_name!" I know it isn't a network problem because the hobbit user can connect using sqlplus just fine and the oratns check is OK. The required Perl modules installed OK... I think. (no, sid_name is not our SID, name has been changed to protect the innocent) Thanks! -- --Ben Benjamin P. August System Administrator - VPUE Stanford University

4 6

Hobbit to IPPhone
by jonescr＠cisco.com 05 May '08

05 May '08

I just cobbled together a script that puts a hobbit display on a Cisco IPPhone (7960). The display is pretty spartan since we only have half a dozen lines to put text on :) The script is run out of cron on my linux workstation, and requires access to an http server, and knowing your phones login and password (same as your user login to call manager). If anyone is interested in this script I will post it here or send it to the Shire. I just did it mainly for fun, but with some effort you could easily even acknowledge alerts right from your phone :) -Charles

9 37

Checkpoint High Availability Monitoring
by linuxmasterjedi＠free.fr 05 May '08

05 May '08

Hi, once again ;) I would like to monitor the HA cluster between 2 Checkpoint Firewall. Several time a month, the cluster failed down, sound like one of the numerous Ethernet card seems to be off during a couple of second, thus the HA cluster switch to safe mode. I would like to detect this nasty state from the Hobbit server to broadcast an alarm. I don't know how the 2 Firewalls exchange health information (maybe via Heartbeat). Anyone already figure out how to monitor this issue? Thanks by advance, LMJ

4 10

Depends feature
by David.Stuffle＠adesa.com 02 May '08

02 May '08

I'm using the "depends" feature heavily for HTTP tests and it is doing some weird things. When a HTTP test fails, the status turns clear saying the depended on test has failed, but it has not failed. But then the status turns red, then clear, then red, then clear... See the history below. Each clear status increases by 5 seconds. And I am positive that the depended on test has not failed. Mon Apr 21 20:24:37 2008 0:04:36 Mon Apr 21 20:24:11 2008 0:00:26 Mon Apr 21 20:19:30 2008 0:04:41 Mon Apr 21 20:19:09 2008 0:00:21 Mon Apr 21 20:14:23 2008 0:04:46 Mon Apr 21 20:14:07 2008 0:00:16 Mon Apr 21 20:09:16 2008 0:04:51 Mon Apr 21 20:09:06 2008 0:00:10 Mon Apr 21 20:04:09 2008 0:04:57 Mon Apr 21 20:04:04 2008 0:00:05 Mon Apr 21 20:00:04 2008 0:04:00 Syntax is like this: 0.0.0.0 http-a-chryslerfirst # noconn http://www.chryslerfirst.com=192.168.254.94 depends=(http:was-b-chryslerfirst/http,was-c-chryslerfirst/http) Anyone experience the same thing or have any idea how to troubleshoot? Using 20080103 snapshot. Thanks! D. Stuffle

2 1

admin-senders
by matthew.epp＠us.army.mil 30 Apr '08

30 Apr '08

Are enable/disable commands included in the information forwarded by a proxy as per the admin-senders parameter? I have one main server that receives events from a proxied secondary, all data comes through normally, but when I blue something out on the secondary, the command does not get automatically applied to the master. If this is not the case, has anyone attempted to implement such a system? Also, has anyone attempted setting up a multi-master environment, where each server would run a proxy on port 1984, forwarding to every other server on their display port 1985 (to avoid infinite looping of data)? If so, any problems encountered?

1 0

What's the syntax to display multiple graphs on a status page?
by john.d.burk＠gmail.com 30 Apr '08

30 Apr '08

All data is in the same rrd file. The 2nd graph [mysqlrunningthreads] works if I swap out its name for the working [mysql] graph. Is this possible, or is it one graph per status page, and only show multiples on the trends page? The trends page for the mysql server is getting pretty long. john burk

2 2

Todays (20080430) snapshot
by lars.ebeling＠leopg9.no-ip.org 30 Apr '08

30 Apr '08

gcc -g -O -D_REENTRANT -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -DHPUX -I. -I /home/hobbit/snapshot/include -I/usr/local/include -o bbhostgrep bbhostgrep.o .. /lib/libhobbit.a -L/usr/local/lib -lz -lnsl /usr/ccs/bin/ld: Unsatisfied symbols: SSL_write (code) SSL_CTX_use_PrivateKey_file (code) SSL_get_error (code) process_receive (code) SSL_CTX_ctrl (code) ERR_print_errors_fp (code) SSL_CTX_use_certificate_chain_file (code) SSLv23_client_method (code) SSL_load_error_strings (code) SSL_shutdown (code) SSL_read (code) SSL_CTX_check_private_key (code) SSL_do_handshake (code) SSL_CTX_new (code) SSL_free (code) SSL_library_init (code) collect2: ld returned 1 exit status make[1]: *** [bbhostgrep] Error 1 make[1]: Leaving directory `/home/hobbit/snapshot/common' make: *** [common-build] Error 2 $ -- Regards Lars Ebeling http://leopg9.no-ip.org Hobbithobbyist "I am not young enough to know everything." -- Oscar Wilde

2 1

All stock Hobbit tests work on remote host, except for 'clock'
by john.d.burk＠gmail.com 30 Apr '08

30 Apr '08

it works on the localhost, but not on the (only monitoring 1 for now) remote host. Every item in the remote host's clock.rrd is 'nan'. Any hints where to start looking? john burk

1 0