February 2006 - Xymon - Xymon mailman web

Snapshot 19/2 crash
by lars.ebeling＠leopg9.no-ip.org 19 Feb '06

19 Feb '06

Hi all, todays (060219) crashed. This was found in hobbitlaunch.log: $ pg * 2006-02-19 14:54:38 hobbitlaunch starting 2006-02-19 14:54:38 Loading tasklist configuration from /home/hobbit/server/etc/ hobbitlaunch.cfg 2006-02-19 14:54:38 Loading hostnames 2006-02-19 14:54:38 xmalloc: Out of memory! 2006-02-19 14:54:39 Could not connect to bbdÉ192.168.0.2:1984 - Connection refus ed 2006-02-19 14:54:39 Whoops ! bb failed to send message - Connection failed 2006-02-19 14:54:39 Task hobbitd terminated by signal 6 2006-02-19 14:54:46 Loading hostnames 2006-02-19 14:54:46 xmalloc: Out of memory! 2006-02-19 14:54:47 Task hobbitd terminated by signal 6 $ And this is from the coredump: $ ll core -rw------- 1 hobbit users 312908 Feb 19 14:50 core $ file core core: core file from 'hobbitd' - received SIGABRT $ gdb bin/hobbitd ./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 éhobbitd'. 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 hobbitd referenced in dld_list. 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 0x00011cc0 in xmalloc (size=1073749888) at memory.c:122 #5 0x00003688 in load_clientconfig () at hobbitd.c:328 #6 0x0000c914 in main (argc=14, argv=0x18000) at hobbitd.c:3755 (gdb) q $ 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

2 2

sendmail graphs
by great.dilla＠gmail.com 19 Feb '06

19 Feb '06

How does one enable [sendmail] graphs? I have searched the mailing list archives but cannot find any hints. I'm using version 4.1.2p1. Things I have done so far: on solaris clients, edited "hobbitclient-sunos.sh" inserted lines: ***** echo "[sendmail]" mailstats ***** on linux clients, edited "hobbitclient-linux.sh" inserted lines: ***** echo "[sendmail]" mailstats ***** on the hobbit server, added "TRENDS:*,sendmail:sendmail|sendmail1" to each host monitored. I don't see any graphs for sendmail on the hobbit webpage. Any inputs?

2 3

Monitoring NetWare - Changing between purple & green frequently
by TMoore＠rooneyholdings.com 17 Feb '06

17 Feb '06

Hello all, I'm using the Perl/MRTGEXT.NLM combination designed for BigBrother to monitor NetWare systems (it's all I could find). My problem is that the system on which I'm testing frequently changes between "purple" and "green/yellow" statuses for the items it's tracking. This happens about every five minutes, which is to say it spends 5 minutes in the "purple" status for the MRTGEXT.NLM based tests, and 5 minutes or so in the normally reporting states. I've tried playing around with the interval settings, to no real avail, but I'm not sure I know 100% what I'm doing. Any help is greatly appreciated...Also, for what it's worth, here's some possibly relevant config entries: BB-HOSTS: ... 192.168.26.7 mdfwfp16.tamu.rbc.app # ncp dns nwstats ... HOBBITLAUNCH [bbdisplay] ENVFILE /usr/lib/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd GROUP generators CMD bbgen $BBGENOPTS --report LOGFILE $BBSERVERLOGS/bb-display.log INTERVAL 1m [bbnet] ENVFILE /usr/lib/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd CMD bbtest-net --report --ping --checkresponse LOGFILE $BBSERVERLOGS/bb-network.log INTERVAL 90 [nwstats] ENVFILE /usr/lib/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd CMD $BBHOME/ext/nwstats2bb_wrap.sh LOGFILE $BBSERVERLOGS/nwstats.log INTERVAL 1m

1 0

[hobbit] Intervals and graphs
by greg.hubbard＠eds.com 17 Feb '06

17 Feb '06

Doug, The RRD definitions have a "valid time" -- default is 600 seconds, or 10 minutes. So if your values are not updated within 10 minutes they are considered invalid. Problem is that you may have to fiddle in the source to change these default settings to accommodate non-standard polling intervals... GLH -----Original Message----- From: Doug Ledbetter [mailto:dougl at mybetter.com] Sent: Friday, February 17, 2006 12:46 PM To: hobbit at hswn.dk Subject: [hobbit] Intervals and graphs Hello, I'm running a Hobbit server and have one other client connecting to it. All are running Linux. So far I've been pleased with it except for one issue. I changed the interval from 5m to 15m for several settings (bbcombotest, bbnet, hobbitclient) in the "hobbitlaunch.cfg" config file (on both the server & client), restarted the Hobbit server and client and then noticed that it stopped graphing. It would still update the current status board, etc. but it stopped adding to the graphs. I let it go like this for a few days and then recently set the intervals back to 5m, restarted the server and client and then it started adding to the graphs again. Is this normal? Shouldn't I be able to run the tests every 15 minutes instead of every 5 and still get graphs? Am I missing something? Thanks, -dougl ____________________________________________________________ Doug Ledbetter --> dougl at mybetter.com MyBetter Web Hosting - Budget Web Hosting http://www.mybetter.com/ PGP Public Key: http://www.mybetter.com/public_key.html To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

1 0

query message : hobbit protocol question
by etienne.grignon＠gmail.com 17 Feb '06

17 Feb '06

Henrik, I'm having problems to implement query message. I'm sending the message bellow to the socket : query myserver.cpu Then, on the recv call, I get nothing. Then, the socket seems to be closed after somes seconds. Is it the right syntax ? or is there a way to find what's going on on the server at this moment ? It is done against hobbit 4.1.1 Thank you for your help, Etienne

2 2

Potential patches for Darwin and FreeBSD
by glarkin＠sourcehosting.net 17 Feb '06

17 Feb '06

Hi folks, I am using Mac OS X 10.3.9 for my Hobbit server, and I'm monitoring some RH9 Linux boxes and FreeBSD 4.11 boxes. The FreeBSD box reported disk problems as soon as I got the client installed. The filesystem type "linprocfs" is not ignored during the disk checks, so I patched hobbitclient-freebsd.sh to omit it, since it always reports 100% usage. I also patched hobbitclient-darwin.sh a couple of days ago to fix an argument problem with "df" that resulted in this error message over and over in my hobbitclient.log file: df: illegal option -- T usage: df [-b | -H | -h | -k | -m | -P] [-ailn] [-t type] [file | filesystem ...] Henrik (or anyone else), if you want these patches, email me separately and I'll send them to you or place them somewhere for download. If there's a better way to submit patches, let me know. Regards, Greg, SourceHosting.net, LLC Ready. Set. Code. http://www.sourcehosting.net/

1 0

[hobbit] Swamped with e-mails
by neil.camp＠deca.mil 17 Feb '06

17 Feb '06

Sorry for filling up your mailbox. See, if Hobbit was not so useful, we would not want to e-mail you so much ;-) --neil ManTech DeCA Operations Systems Administrator (804) 734-8353 neil.camp at deca.mil -----Original Message----- From: Henrik Stoerner [mailto:henrik at hswn.dk] Sent: Friday, February 17, 2006 6:51 AM To: hobbit at hswn.dk Subject: [hobbit] Swamped with e-mails Guys & gals, my mailbox is currently overflowing - I have about 2000 hobbit-related e-mails that I need to read through. I'll probably spend most of the week-end clearing the back-log. So please - don't mail me directly unless you've got something that is *really* urgent (or has nothing to do with Hobbit). Regards, Henrik To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

1 0

Swamped with e-mails
by henrik＠hswn.dk 17 Feb '06

17 Feb '06

Guys & gals, my mailbox is currently overflowing - I have about 2000 hobbit-related e-mails that I need to read through. I'll probably spend most of the week-end clearing the back-log. So please - don't mail me directly unless you've got something that is *really* urgent (or has nothing to do with Hobbit). Regards, Henrik

1 0

Difficulty with CPU/disk/procs columns
by glarkin＠sourcehosting.net 16 Feb '06

16 Feb '06

Hi there, First of all, let me say - Hobbit is great! I set aside some hours yesterday to get things set up, and it's working really well. Here's my setup: Hobbit version 4.1.2p1 BBDISPLAY/BBNET/BBPAGER: Mac OS X 10.3.9 machine Hostname: Gregory-Larkins-Computer.local Client: RedHat 9 (2.4.20-31.9.progeny.8 i686 i686 i386 GNU/Linux) Hostname: patches The one problem I'm having is that I can't get the CPU/disk/procs columns to display for the client I'm monitoring. I double- and triple-checked by bb-hosts, hobbit-clients.cfg, firewall, etc. So far, everything looks good, but there must be a problem somewhere. I've followed the advice here (http://hobbitmon.sourceforge.net/docs/hobbit-tips.html#noclient) but I haven't fixed the problem yet. Here are the relevant parts of my config files: **** **** bb-hosts **** **** 192.168.1.240 Gregory-Larkins-Computer.local # BBDISPLAY BBPAGER BBNET bbd http://localhost/ ssh smtp ftp 204.8.45.68 www.sourcehosting.net # CLIENTNAME:patches http://www.sourcehosting.net/ **** **** ******** **** **** **** **** hobbit-clients.cfg **** **** HOST=Gregory-Larkins-Computer.local PROC named 1 1 PROC /usr/local/sbin/dyndnsd 1 1 PROC ntpd 1 1 HOST=patches LOAD 3.0 5.0 DEFAULT # These are the built-in defaults. UP 1h LOAD 5.0 10.0 DISK * 90 95 MEMPHYS 100 101 MEMSWAP 50 80 MEMACT 90 97 **** **** ****************** **** **** I've checked that messages are getting through the firewall around the BBDISPLAY machine by running strace on the client machine. I also checked the msg.txt file on the client to make sure that the client name is correct: *** *** *** *** [hobbit at patches tmp]$ head -5 msg.txt client patches.linux [date] Thu Feb 16 14:26:26 EST 2006 [uname] Linux patches 2.4.20-31.9.progeny.8 #1 Thu Jun 9 22:44:19 EDT 2005 i686 i686 i386 GNU/Linux *** *** *** *** The client name looks OK to me, and the messages are getting through the firewall as far as I know. I ran ktrace on the Mac OS X machine to make sure that the client messages are received, and it looks like they are, but what file should the msg.txt data be written to? I've looked in the data subdirectory, but haven't found it yet. I would guess that's related to my problem. What other things should I do to troubleshoot this problem? Thanks very much for a great tool, Greg SourceHosting.net, LLC Ready. Set. Code. http://www.sourcehosting.net/

2 2

MRTG for Switches - OK why?
by Todd.Costa＠state.ma.us 16 Feb '06

16 Feb '06

Hello again, After I sent the email below. I figured I would continue my reading and also go back to how I used BB to see if I could put the pieces together in Hobbit 4.1.2p1. I currently use Debian (Sarge) 3.1 as my platform of choice and needed to add another lib to get things moving. Eventually, I figured out that I needed to use cfgmaker with the appropriate options and switches. Good so far. I had removed my previous host names from bb-hosts and dropped the names with the bb command. Waited and cleaned up any residual left of those names in the Hobbit file structure. I commented out the mrtg in hobbitlaunch.cfg before I did the previous steps and finished reading up on cfgmaker. OK, so I decided on what I wanted in cfgmaker and went with the naming convention Henrik provided in his docs. I ran cfgmaker though a script and generated the mrtg.cfg file. Went throught it and mod'd it some more to my liking. Cool so far was what I was thinking. Now I am ready to put the ip and hostname back in bbhosts and to wait a few minutes later for the basic columns to appear. I quickly went through the hobbit logs and notice a couple of things like creating directories, yada yada..., then I noticed in the trend column the MRTG graphs were showing up. OK stop I thought. I didn't uncomment the mrtg in hobbitlaunch.cfg. What gives? So now I am confused. Why is this happening and what would be the purpose of enabling it in hobbitlaunch if hobbit already does it automatically. I am guessing the RRDtool is automatically picking up that I want stats from mrtg.cfg I just created. I think this is what I meant when hobbit makes thing alot easier. So does anyone understand what I am describing? Any discussion on this would help me understand what I am seeing. Thanks in advance TMC > -----Original Message----- > From: Costa, Todd (DMH) > Sent: Wednesday, February 15, 2006 4:39 PM > To: Hobbit (E-mail) > Subject: MRTG for Switches > > > Hello Hob-sters, > > I am wondering were I can find some mrtg cfg examples > that will let me track port traffic activity on switches? I > recently followed the integrating steps for MRTG data into > Hobbit. And I now see MRTG in the trends link which is cool. > But I now want to watch by ports and generate graphs per > port. Can this still be done? I once did this through BB and > I remember it being tricky and cumbersome to do. But so far, > everything I have tried with hobbit has gotten easier to deal > with. I hope its the same with what I am asking now. Any help > and input is appreciated. > > > Todd M. Costa > Commonwealth of Massachusetts >

1 0