August 2007 - Xymon - Xymon mailman web

custom script help
by arosen＠oleh.com 03 Aug '07

03 Aug '07

Hi, I'm writing a custom script that parses a mailbox and sends the contents to the hobbit server. The problem is some emails are large, so I would like the status page to show a list of email subjects which, when one is clicked, would display the contents of that message. Is that possible? If so, how would I go about sending the body of each message and generating a link in the status page. Thanx. Avi R.

4 9

bbgen coredumps todays (3/8) snapshot
by lars.ebeling＠leopg9.no-ip.org 03 Aug '07

03 Aug '07

$ gdb bin/bbgen tmp/core GNU gdb 5.3 Copyright 2002 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 "hppa2.0n-hp-hpux11.00"... Core was generated by `bbgen'. Program terminated with signal 6, Aborted. warning: The shared libraries were not privately mapped; setting a breakpoint in a shared library will not work until you rerun the program. warning: Can't find file bbgen referenced in dld_list. Reading symbols from /usr/local/lib/libpcre.sl...done. Reading symbols from /usr/lib/libnsl.1...done. Reading symbols from /usr/lib/libxti.2...done. Reading symbols from /usr/lib/libc.2...done. Reading symbols from /usr/lib/libdld.2...done. #0 0xc020bad0 in kill () from /usr/lib/libc.2 (gdb) bt #0 0xc020bad0 in kill () from /usr/lib/libc.2 #1 0xc01a655c in raise () from /usr/lib/libc.2 #2 0xc01e69a8 in abort_C () from /usr/lib/libc.2 #3 0xc01e6a04 in abort () from /usr/lib/libc.2 #4 0x00027498 in sigsegv_handler (signum=16432) at sig.c:57 #5 <signal handler called> #6 do_eventlog (output=0x7b031468, maxcount=100, maxminutes=240, fromtime=0x0, totime=0x0, pageregex=0x0, expageregex=0x0, hostregex=0x0, exhostregex=0x0, testregex=0x0, extestregex=0x0, colrregex=0x0, ignoredialups=0, hostcheck=0x400027e2 <::hexchars_(void)+1066>, eventlist=0x0, hostcounts=0x0, servicecounts=0x0, counttype=COUNT_NONE, sumtype=S_NONE, periodstring=0x0) at eventlog.c:760 #7 0x0000c080 in do_bb2ext (output=0x7b031468, extenv=0x7c000b58 "", family=0x32ee0 "mkbb") at pagegen.c:932 #8 0x0000c780 in do_bb2_page (nssidebarfilename=0x0, summarytype=1) at pagegen.c:1110 #9 0x00005a4c in main (argc=6, argv=0x31800) at bbgen.c:643 (gdb) -- 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

Frank Jakobsen is out of the office.
by fjakobsen＠csc.com 03 Aug '07

03 Aug '07

I will be out of the office starting 03.08.2007 and will not return until 27.08.2007. For EMEA UNIX support create a Monet ticket for EMEA-MCS-UNIX-SRV.

1 0

Snapshot issue
by ralphmitchell＠gmail.com 03 Aug '07

03 Aug '07

Henrik, Either I've done something silly, or something you changed after the 2007-07-25 snapshot may have broken "bin/bb". I'm getting this in my hobbitclient.log: Hobbit version 4.3.0-0.20070725 Usage: /home/hobbit/client/bin/bb [--debug] [--proxy= http://ip.of.the.proxy:port/] RECIPIENT DATA RECIPIENT: IP-address, hostname or URL DATA: Message to send, or "-" to read from stdin and purple dots for my Hobbit server's cpu, disk, memory, etc. To make sure it wasn't anything old left behind from the previous install, I moved the server & client directories to another place and reran the 'make install', then copied back server/etc/bb-hosts. I'm still getting the above usage message and no client reports for the server itself. I get the same kind of usage message if I try: server/bin/bb 0.0.0.0 "drop xxxf501 conn" When I copy the client/bb & common/bb binaries from the 2007-07-25 snapshot build directory over to client/bin/bb & server/bin/bb, I now get the server's cpu, memory, disk, &c reports. This is with gcc-4.1.2 & glibc-2.5 on a Gentoo Linux system. Ralph Mitchell

4 4

BBWin: cluster.exe external
by s_aiello＠comcast.net 02 Aug '07

02 Aug '07

All, I am still in the process of upgrading from bbgen to Hobbit. After the server upgrade I plan on upgrading all the clients to hobbit and bbwin clients. Since I close to the end of this upgrade project I see no need to install the old bbclients on windows servers, so I have been performing bbwin installs. Today I had the opportunity to use the cluster.exe external module. The module works great when reporting to a Hobbit Server. A problem occurs when it is reporting to a BB/BBGEN server. It seems when the report is sent in BB/BBGEN see the first line as " green" not "green". Looking at the bbwin/tmp/cluster file before it is sent to the server in notepad and wordpad, the first line is "green". But looking at the bbvar/logs/server.cluster file it is " green". So somewhere in the communication of BBWin client --> bbproxy --> BBGen something weird is going on. Since " green" is an unknown color status, BBGen has issue displaying the status icon & background color of the cluster report. I believe the problem is limited to the cluster.exe module and not in the bbwin client since I have written a vbs external script, and it reports fine into BBGen. I do not see this as a high priority bug, especially since I am moving to hobbit soon. Figured it just needed to be reported on. ~Steve

1 0

hobbit and lighttpd
by rodolfo＠pilas.net 02 Aug '07

02 Aug '07

I have lighttpd serving hobbit web pages. I am not a expert lighttpd user but this is the configuration that works to me, perhaps anybody can improve it and developers can distribute the configuration file (like hobbit-apache.conf) ################################ ## hobbit configuration based on ## http://www.lighttpd.net/documentation/cgi.html server.modules += ( "mod_cgi", "mod_auth" ) alias.url += ( "/hobbit/" => "/var/lib/hobbit/www/", "/hobbit-cgi/" => "/usr/lib/hobbit/cgi-bin/", "/hobbit-seccgi/" => "/usr/lib/hobbit/cgi-secure/" ) $HTTP["url"] =~ "^/hobbit-cgi/|^/hobbit-seccgi/" { dir-listing.activate = "disable" cgi.assign = ( "" => "" ) } auth.backend = "htpasswd" auth.backend.htpasswd.userfile = "/etc/hobbit/hobbitpasswd" auth.require = ( "/hobbit-seccgi/" => ( "method" => "basic", "realm" => "Hobbit Administration", "require" => "user=admin" ) ) --- Regards, Rodolfo Pilas

2 1

How to if_* graph from Devmon to Hobbit
by xbgmsharp＠gmail.com 01 Aug '07

01 Aug '07

How to if_* graph from Devmon to Hobbit This mail describe how to graph all if_* data from Devmon to Hobbit. I am currently running hobbit 4.2.0/all-in-one patch and devmon 0.3-beta2 1) First you need a curently working hobbit and devmon. 2) Then make sure, than the file devmon/templates/modele/if_load/message TABLE:rrd() This will show data in a htmltable and add rrd information in the status page. In order to check, you should see RRD data information in the status page of the service corresponding to your devmon modele when showing source html. 3) Now make parse the data to create a rrd file (one par interfaces), this is done via extra-rrd.pl. In etc/hobbitlaunch.cfg [rrdstatus] ENVFILE /data/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd CMD hobbitd_channel --channel=status \ --log=$BBSERVERLOGS/rrd-status.log hobbitd_rrd \ --rrddir=$BBVAR/rrd \ --extra-script=$BBHOME/ext/extra-rrd.pl \ --extra-tests=if_load In ext/extra-rrd.pl if ( $TESTNAME eq "if_load" ) { # Analyze the message we got foreach $line (@input) { if ($line =~ /(^[A-Z].*) (\d+):(\d+)$/ ) { # The RRD dataset definitions print "DS:in:DERIVE:600:0:U\n"; print "DS:out:DERIVE:600:0:U\n"; # The filename print "if_load_$1.rrd\n"; # The data print "$2:$3\n"; } } } 4) Modify etc/hobbitgraph.cfg [if_load] FNPATTERN if_load(.*).rrd TITLE Network Traffic YAXIS Bits/second DEF:in at RRDIDX@=@RRDFN@:in:AVERAGE CDEF:inbytes at RRDIDX@=in at RRDIDX@,8,* DEF:out at RRDIDX@=@RRDFN@:out:AVERAGE CDEF:outbytes at RRDIDX@=out at RRDIDX@,8,* LINE1:inbytes at RRDIDX@#@COLOR@:@RRDPARAM@ In GPRINT:inbytes at RRDIDX@:LAST: %8.2lf %s (cur) GPRINT:inbytes at RRDIDX@:MAX: %8.2lf %s (max) GPRINT:inbytes at RRDIDX@:MIN: %8.2lf %s (min) GPRINT:inbytes at RRDIDX@:AVERAGE: %8.2lf %s (avg)\n LINE1:outbytes at RRDIDX@#@COLOR@:@RRDPARAM@ Out GPRINT:outbytes at RRDIDX@:LAST: %8.2lf %s (cur) GPRINT:outbytes at RRDIDX@:MAX: %8.2lf %s (max) GPRINT:outbytes at RRDIDX@:MIN: %8.2lf %s (avg) GPRINT:outbytes at RRDIDX@:AVERAGE: %8.2lf %s (avg)\n 4) In etc/hobbitserver.cfg Modify the following. See help/howtograph.html TEST2RRD="cpu=la,disk,...,if_load" GRAPHS="la,disk,if_load::1" The ::1 set the maxcount in order to make one graph form each rrd file create. 5) In etc/hobbitcgi.cfg Modify the following. See hobbitsvc.cgi man CGI_SVC_OPTS="--env=/data/hobbit/server/etc/hobbitserver.cfg --no-svcid --history=top --multi-grahs=if_load,if_err..." 6) The hardest part. The hobbit parsing doesn't remove html table table so you will see many graph link. In order to fix than, you need to modify the hobbit source with the patch: lib/htmllog.c $cd lib/ $patch -p0 < htmllog.path Here is the path, this only set a new counter for all if_* services. This is allready working for the trends colunms because, it is checking all rrd file in the directory. Henrik, do you have any tips to get the nice counter without editing the source code of htmllog.c file ? That's all folks.

4 3

[hobbit] Feature requests; filter out MVFS
by me2unix＠gmail.com 01 Aug '07

01 Aug '07

I solved this Issue by adding this to hobbit-clients.cfg : HOST=%cc.* DISK %^/vobs/.* IGNORE where HOST= should be your clearcase hostname(s) --Hezki ------------------------------ - *To*: hobbit (at) hswn.dk - *Subject*: RE: [hobbit] Feature requests; filter out MVFS - *From*: "T.J. Yang" <tj_yang (at) hotmail.com> - *Date*: Thu, 26 Jul 2007 17:08:51 -0500 - *Bcc*: - *Cc*: shea_greg (at) emc.com ------------------------------ From: shea_greg (at) emc.com Reply-To: hobbit (at) hswn.dk To: <hobbit (at) hswn.dk> CC: <shea_greg (at) emc.com> Subject: RE: [hobbit] Feature requests; filter out MVFS Date: Thu, 26 Jul 2007 16:28:52 -0400 Hi Henrik, A software versioning tool called ClearCase creates this pseudo filesystem on whichever server you install it on. We have it on a bunch of servers here as the software development environment is spread across continents. Same here, beside it is cross continents. it is also cross OS platform. I need to customize the df script to exclude mvfs on Linux,hpux and solaris before I can deploy hb client. tj yang

2 1

[hobbit] Feature requests; filter out MVFS
by michael.nemeth＠lmco.com 01 Aug '07

01 Aug '07

Might be fine for server with a few /vob s but Ive systems with 100+ vobs ; That a lot of lines to be sending. I'd like it to be added hobbitclient-sunos.sh in the "egrep -v" and any other client script that work like sun. Cant see where that hurts! The DISK %^/vobs/.* IGNORE OR " customize the df script" for the client (Im going to do this for my one hp-ux system since its one that has many vobs) for others ------------------------------ I solved this Issue by adding this to hobbit-clients.cfg : HOST=%cc.* DISK %^/vobs/.* IGNORE where HOST= should be your clearcase hostname(s) --Hezki ------------------------------ - *To*: hobbit (at) hswn.dk - *Subject*: RE: [hobbit] Feature requests; filter out MVFS - *From*: "T.J. Yang" <tj_yang (at) hotmail.com> - *Date*: Thu, 26 Jul 2007 17:08:51 -0500 - *Bcc*: - *Cc*: shea_greg (at) emc.com ------------------------------ From: shea_greg (at) emc.com Reply-To: hobbit (at) hswn.dk To: <hobbit (at) hswn.dk> CC: <shea_greg (at) emc.com> Subject: RE: [hobbit] Feature requests; filter out MVFS Date: Thu, 26 Jul 2007 16:28:52 -0400 Hi Henrik, A software versioning tool called ClearCase creates this pseudo filesystem on whichever server you install it on. We have it on a bunch of servers here as the software development environment is spread across continents. Same here, beside it is cross continents. it is also cross OS platform. I need to customize the df script to exclude mvfs on Linux,hpux and solaris before I can deploy hb client. tj yang -- +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ | _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/ |++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

1 0

Grouping alerts for multiple hosts. (purple)
by gwege＠infomanage.net 01 Aug '07

01 Aug '07

Hello, I'm in the process of upgrading from bb to hobbit. In BB I had written a few custom tests that would grep through the bb status logs for a group of hosts matching color purple, it would then report what it found and bb would send out an alert based on the results. This allowed me to monitor the status of our remote offices, if it went down I would receive a sms page. Basically I want to receive one single alert if this remote office goes offline, not a page for every device. Is there a way to handle this natively in hobbit? I cannot do conn based testing. I know I will not be able to use this, however, I started messing around with the conn tests and the route: flag within bb-hosts but I was unsuccessful with getting it to work in my test environment. The servers would not change to clear status if the gateway reported as red (down). Thanks Greg

2 1