May 2005 - Xymon - Xymon mailman web

Comments about the Debian packaging of Hobbit
by Christian.Perrier＠onera.fr 11 May '05

11 May '05

Henrik, I'm in the process of installing Hobbit (besides a working BB install, on a Debian sarge server) and I decided to do this with the Debian packages you have built. My intent, indeed, is getting Hobbit in the official Debian distribution by finding a Debian developer wanting to maintain it, or sponsor uploads for you. Even if I'm am a Debian developer myself (bubulle at debian.org) I can't really manage to take this over...at least at the moment. As I'm am installing the software, let's make comments about what I find in the Debian packaging which could be "improved". (indeed, this is mostly nitpicking...) Debconf templates ----------------- Make them translatable. Some rewrite below to follow the Debconf Templates Style Guide we apply in Debian (I'm the author if it..:-))) This needs using dh_installdebconf in debian/rules, which is recommended anyway. Template: hobbit/BBSERVERHOSTNAME Type: string Default: localhost _Description: Hobbit server hostname: Please enter the hostname of the Hobbit server. This is usually the fully-qualified DNS name for this host. It should be resolvable via DNS, or you may run into problems with your network tests. Template: hobbit/BBSERVERIP Type: string Default: 127.0.0.1 _Description: IP address of the Hobbit server: Please enter the IP address which clients will use to access your Hobbit server. Template: hobbit/BBSERVERWWWNAME Type: string Default: localhost _Description: Hobbit webserver hostname: Webservers sometimes use a different hostname than the name of the server they are running on - e.g. www.sample.com is the web hostname, but the server might be known as web01.sample.com. . Please enter the host name used to contact the web interface of the Hobbit server. PID file location ----------------- Can't be in /var/log. should be in /var/run/hobbit Conffiles: ---------- /var/lib/hobbit/www/menu/menu_items.js /var/lib/hobbit/www/menu/menu_tpl.js /var/lib/hobbit/www/menu/menu.css Should be moved in /etc by some mean.... Ext scripts ----------- Currently, adding custom ext scripts requires putting them in /usr/lib/hobbit/server/ext This slightly does against Debian Policy where the /usr directory is aimed at hosting files coming from Debian packages, but no locally modified files. The logical place for these extensions seems to be /usr/local/lib/hobbit/ext or something similar. I would here recommend adding the option for hobbit to look in more than one directory for extension scripts. This would allow future "plugin" packages for some Hobbit extension scripts to put their files in /usr/lib/hobbit/server/ext while the local admin still have the opportunity to use theirs in /usr/local. Apache config ------------- The example file could be put as "hobbit" in /etc/apache/conf.d. However, this may need to check the specific policy for apache-dependent apps More to come.... Do you have the diff.gz file available somewhere ? --

2 1

Big Sister client with Hobbit Server ?
by mark.simon＠mci.com 11 May '05

11 May '05

Does anyone use the Big Sister client with the Hobbit Server for a complete open source replacement for Big Brother? Mark Simon MCI SysAdmin (972)729-5674 aol marksimon4u

1 0

[hobbit] Some thoughts on clustered hobbit
by KauffmanT＠nibco.com 11 May '05

11 May '05

I'm planning to have (on the failover system) two hobbitlaunch.cfg files - a hobbitlaunch.stby and a hobbitlaunch.run. The .stby file has bbpage and bbnet disabled, while .run does not. At failover time, stop hobbit, copy the .run to .cfg, and start hobbit. At recovery time, stop hobbit, copy .stby to .cfg, start hobbit. I'm already syncing the config files from the primary to the standby whenever changes occur (I'm the only one making changes, so far). What I *don't* have in place is a dedicated IP address; right now, if the BB server (now primary hobbit server) goes down, you need to know which system is running the fallover display. And I'm trying to decide if I want to sync the history and the rrds BACK to the primary on recovery, or just leave a hole in the data. Right now, I'm leaning toward doing this manually after the fact. Tom -----Original Message----- From: Brian Lynch [mailto:brianlynch at gmail.com] Sent: Tuesday, May 10, 2005 2:22 AM To: hobbit at hswn.dk Subject: Re: [hobbit] Some thoughts on clustered hobbit I agree with your assessment, but chose the model for a few reasons (note that I'm basing my experience on about 2 1/2 years running a dual big brother failover setup): 1. There is always one repository for both configuration and data that are kept reasonably identical on both systems (within the synch delay). 2. There is only one ip address accepting BB reports cutting down on both network traffic and firewall rules (for hosts in locked down vlans). 3. The other system can be dedicated to another purpose (it currently hosts our documentation site that fails over in the opposite direction). 4. No redundant work is done. Indeed, no load is being 'shared' across the systems unless you host the web server on the other box. There is a risk to this based on the possibility of complete machine failure in between synchronizations. Hence, Hobbit may come up without all the updates for hosts or alerts. Based on my current model, I will lose about a day of historical data. These synch rates can be changed and a gigabit crossover between machines cuts down on any traffic imposed by multiple synch's. Note that you could very easily turn off the hobbit alerts with the same clustering software by truncating and restoring the hobbit-alerts.cfg file. Not sure how to disable the network tests, so that may require some custom coding... Once complete, you could use the same cluster resource sw to accomplish a 'hot' standby.

1 0

bbdisplay problems after adding some new hosts
by stefan_loos＠hotmail.com 11 May '05

11 May '05

Hi, yesterday I add some new hosts to my hobbit-server and short after that hobbit had some problems. Here is what hobbitlauch.log says: 2005-05-11 09:11:18 Heartbeat lost for task hobbitd, bouncing it 2005-05-11 09:11:18 Task bbretest started with PID 4523 2005-05-11 09:11:23 Heartbeat lost for task hobbitd, killing it 2005-05-11 09:11:23 Task bbdisplay started with PID 4524 2005-05-11 09:11:23 Task hobbitd terminated by signal 9 2005-05-11 09:11:23 Task hobbitd started with PID 4525 2005-05-11 09:11:23 Loading hostnames 2005-05-11 09:11:23 Loading saved state 2005-05-11 09:11:23 Setting up network listener on 0.0.0.0:1984 2005-05-11 09:11:23 Setting up signal handlers 2005-05-11 09:11:23 Setting up hobbitd channels 2005-05-11 09:11:23 Setting up logfiles 2005-05-11 09:11:28 Task bbhistory started with PID 4527 2005-05-11 09:11:28 Task bbenadis started with PID 4528 2005-05-11 09:11:28 Task bbpage started with PID 4530 2005-05-11 09:11:28 Task larrdstatus started with PID 4532 2005-05-11 09:11:28 Task larrddata started with PID 4534 2005-05-11 09:12:18 Task bbretest started with PID 4541 2005-05-11 09:12:23 Task bbdisplay started with PID 4542 2005-05-11 09:12:43 Heartbeat lost for task hobbitd, bouncing it 2005-05-11 09:12:48 Heartbeat lost for task hobbitd, killing it 2005-05-11 09:12:48 Task hobbitd terminated by signal 9 2005-05-11 09:12:48 Task bbdisplay terminated by signal 15 So I tried to find out which component causes the problem and disabled everything in hobbitlauch.cfg and reenabled one by one. I found out that everytime I enabled bbdisplay those errors occour. The bb-display.log looks like this: 2005-05-11 09:09:48 Whoops ! bb failed to send message - timeout 2005-05-11 09:09:48 hobbitd status-board not available 2005-05-11 09:09:53 Whoops ! bb failed to send message - timeout 2005-05-11 09:10:53 Whoops ! bb failed to send message - timeout 2005-05-11 09:10:53 hobbitd status-board not available 2005-05-11 09:11:23 Could not connect to bbd at 10.207.193.41:1984 - Connection refused 2005-05-11 09:11:23 Whoops ! bb failed to send message - Connection failed 2005-05-11 09:11:23 hobbitd status-board not available 2005-05-11 09:11:23 Could not connect to bbd at 10.207.193.41:1984 - Connection refused 2005-05-11 09:11:23 Whoops ! bb failed to send message - Connection failed I also found some core files in ~server/tmp but I'm pretty shure they came from killing hobbit - nevertheless I've run the gdb util: GNU gdb Red Hat Linux (6.1post-1.20040607.52rh) Copyright 2004 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 "i386-redhat-linux-gnu"...Using host libthread_db library "/lib/tls/libthread_db.so.1". Core was generated by `hobbitd --debug --pidfile=/var/log/hobbit/hobbitd.pid --restart=/usr/local/hobb'. Program terminated with signal 6, Aborted. Reading symbols from /lib/tls/libc.so.6...done. Loaded symbols for /lib/tls/libc.so.6 Reading symbols from /lib/ld-linux.so.2...done. Loaded symbols for /lib/ld-linux.so.2 #0 0x00df4cef in raise () from /lib/tls/libc.so.6 (gdb) bt #0 0x00df4cef in raise () from /lib/tls/libc.so.6 #1 0x00df64f5 in abort () from /lib/tls/libc.so.6 #2 0x08054126 in sigsegv_handler (signum=11) at sig.c:57 #3 <signal handler called> #4 0x00e46cac in mempcpy () from /lib/tls/libc.so.6 #5 0x00e3a4d2 in _IO_default_xsputn_internal () from /lib/tls/libc.so.6 #6 0x00e13527 in vfprintf () from /lib/tls/libc.so.6 #7 0x00e2f3dc in vsprintf () from /lib/tls/libc.so.6 #8 0x00e1a03d in sprintf () from /lib/tls/libc.so.6 #9 0x0804d7a4 in do_message (msg=0x9e0b3f8, origin=0x80554bb "") at hobbitd.c:1903 #10 0x0804fcb5 in main (argc=8, argv=0xbfff9084) at hobbitd.c:2944 (gdb) Now I try to find out which of the new hosts - and what test causes the problems... Regards, Stefan Loos

2 2

[hobbit] bb-histlog.sh
by ivan.verbeeck＠vmma.be 10 May '05

10 May '05

Henrik, Looks very good. Thanx, Ivan > On Sat, May 07, 2005 at 07:16:41PM +0200, Ivan Verbeeck wrote: > > > Sorry, I didn't mean the status off the non-green view, but > the link > > to the historical views on the same page in the section "events > > received in the past x minutes". > > So when a test fails and recovers soon after, it is show in that > > section, but if you will check the history for that test (example, > > when you see the pages first thing in the morning), you > need to browse > > to the main view, to the exact subpage, to the test and > than click on > > the history button. > > OK, now I understand. I've put a "Full history" button on the > historical statuslog display. > > Regards, > Henrik > > > To unsubscribe from the hobbit list, send an e-mail to > hobbit-unsubscribe at hswn.dk > > >

1 0

Hobbit 4.0.3 rc2
by jonbjorn＠hi.is 10 May '05

10 May '05

Just installed fresh installation of rc2 and noticed that I´m missing disk1 rrd graph. Tried to include it in hobbitserver.cfg under GRAPHS but it wont show up. Any idea ? best regards -- Jon Bjorn Njalsson <jonbjorn at hi.is>

2 1

[hobbit] bb-histlog.sh
by ivan.verbeeck＠vmma.be 10 May '05

10 May '05

>> Is it possible to have a history button (and eventualy the link to the >> current-rrd) on a page generated by bb-histlog.sh as it is on a page >> generated by bb-hostsvc.sh? > >Why not just use your browsers' "Back" button ? > >> It's nice to have if you click to a status page from the non-green view. > >I don't follow what you mean. The non-green view doesn't link directly to the historical status views, it >links to the current status-view. > Sorry, I didn't mean the status off the non-green view, but the link to the historical views on the same page in the section "events received in the past x minutes". So when a test fails and recovers soon after, it is show in that section, but if you will check the history for that test (example, when you see the pages first thing in the morning), you need to browse to the main view, to the exact subpage, to the test and than click on the history button. > >Henrik > > >To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

2 1

[hobbit] Some thoughts on clustered hobbit
by ajeffco＠orhs.org 10 May '05

10 May '05

Aha... I haven't had the pleasure of downloading and testing 4.0.3rc2, so that explains it :) __________________________________ Al Jeffcoat IBM Certified Support Specialist, AIX Enterprise SAN and Storage Administrator System Programmer II (321)843-1051 ajeffco at orhs.org -----Original Message----- From: Henrik Stoerner [mailto:henrik at hswn.dk] Sent: Tuesday, May 10, 2005 1:22 AM To: hobbit at hswn.dk Subject: Re: [hobbit] Some thoughts on clustered hobbit Brian Lynch wrote up a description of how to cluster two Hobbit servers - it's in the "contrib" directory in the hobbit-4.0.3rc2 sources. On Mon, May 09, 2005 at 08:41:37PM -0400, Jeffcoat, Al wrote: > Did I miss something? I don't see anything regarding a clustered hobbit > setup in the last couple of days, but would be greatly interested in > clustering 2 AIX servers together across locations to handle hobbit. > Henrik To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk This e-mail message and any attached files are confidential and are intended solely for the use of the addressee(s) named above. If you are not the intended recipient, any review, use, or distribution of this e-mail message and any attached files is strictly prohibited. This communication may contain material protected by Federal privacy regulations, attorney-client work product, or other privileges. If you have received this confidential communication in error, please notify the sender immediately by reply e-mail message and permanently delete the original message. To reply to our email administrator directly, send an email to: postmaster at orlandoregional.org . If this e-mail message concerns a contract matter, be advised that no employee or agent is authorized to conclude any binding agreement on behalf of Orlando Regional Healthcare by e-mail without express written confirmation by an officer of the corporation. Any views or opinions presented in this e-mail are solely those of the author and do not necessarily represent those of Orlando Regional Healthcare.

1 0

[hobbit] Custom RRDs
by dstuffle＠deltafaucet.com 10 May '05

10 May '05

Andy France wrote: > "Stuffle, David" wrote on 10/05/2005 02:15:48: > >>> "Stuffle, David" wrote on 07/05/2005 03:35:47: >>> >>>> I have created a custom RRD script and it been working >>> fine. Any host >>> that >>>> sends a "license" column status with a line containing >>>> "In Use: x" will >>> get >>>> the license usage (x) graphed. But, for some reason >>>> it won't work on a >>> host >>>> I have that has a hostname of "license". RRD file >>>> gets created, but it's all "nan" values. Could having >>>> the exact same hostname, Dataset name, >>> and >>>> column name cause it not to work? The only other >>>> difference in this host >>> is >>>> it's a Windows machine instead of Unix, but the status >>>> sent to Hobbit is >>> the >>>> same so I can't see that it would matter. Thanks! >>> >>>> ~~~~~~~~~~~~~~ >>>> David Stuffle >>>> Delta Faucet Company >>> >> Andy France wrote: >>> It sounds to me like you are picking up the status line >>> rather than the data line. >>> >>> Check your RRD script again... if you are >>> grepping/awking for a line containing license you may >>> need to narrow your criteria :-) >>> >>> It may be as simple as adding "tail -1" to make sure >>> you only get the last occurence. >>> >>> HTH, >>> Andy. >>> >>> >> I have verified I am getting the right data from the >> status by redirecting the script to a file. Somewhere >> between grepping the license count and putting it in the >> RRD file it gets lost. I'm figuring it must be because >> the hostname is the same as the DS name. Any other host >> I plug in to it works fine. Thanks > > If you post the script and message content either here or > directly to me, I'd be happy to take a look :-) > Thanks! Here is the script: (if the format gets messed up, I also attached a file with the same contents) --------------- #!/bin/sh # Input parameters: Hostname, testname (column), and messagefile HOSTNAME="$1" TESTNAME="$2" FNAME="$3" if [ "$TESTNAME" = "license" ] then # Analyze the message we got InUse=`grep "In Use" $FNAME | cut -d : -f 2 | sed "s/ //"` # The RRD dataset definitions echo "DS:license:GAUGE:600:0:U" # The filename echo "license.rrd" # The data echo "$InUse" # Added for debugging purposes if [ "$HOSTNAME" = "license" ] then echo $HOSTNAME, $TESTNAME, $FNAME >>license_test.txt echo "$InUse" >>license_test.txt cat $FNAME >>license_test.txt fi fi The output from the debugging part (license_test.txt) looks like this: -------------- license, license, /usr/local/hobbit/server/tmp/rrd_msg_27159 8 status license.license green Tue 05/10/2005- 7:37:41.18 Pro/Engineer Wildfire 2 Licenses -------------------------------- Total licenses available: 31 In Use: 8 And here's the RRD definition: -------------- [license] TITLE License Utilization YAXIS Used Licenses DEF:avg=license.rrd:license:AVERAGE LINE2:avg#@COLOR@:Used Licenses GPRINT:avg:LAST: \: %5.0lf (cur) GPRINT:avg:MAX: \: %5.0lf (max) GPRINT:avg:MIN: \: %5.0lf (min) GPRINT:avg:AVERAGE: \: %5.0lf (avg)\n Two other hosts are sending license statuses and they're graphing fine. Just this "license" hostname is not. The graph shows up and the rrd file is getting updated, but with no values. Thanks again!! ~~~~~~~~~~~~~~ David Stuffle dstuffle at deltafaucet.com Delta Faucet Company (317) 571-4301 This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. Please note that any views or opinions presented in this email are solely those of the author and do not necessarily represent those of the company. Finally, the recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email.

1 0

bug in enadis a group of systems, one of which is already disabled?
by tgeorgoulias＠mcclatchy.com 10 May '05

10 May '05

We just noticed something strange with enadis. Say we have a system whose procs alert is already disabled. If that system is selected along with others and ALL tests are disabled, only the system previously disabled has ALL tests disabled. The rest do not get disabled. Anyone notice this? Tom

2 2