October 2006 - Xymon - Xymon mailman web

Passing non-linux data to hobbit for monitoring and graphing.
by Thomas.Kern＠hq.doe.gov 17 Oct '06

17 Oct '06

Hello. I have a non-linux system that I want to supply data to a hobbit server for monitoring. This is a z/VM system on an IBM mainframe. I have a client from Rich Smrcina. Using this client as a template, I want to pass other data to hobbit for monitoring and graphing. I have seen the hobbit webpages and graphs showing data like CPU utilization (System, User, IDLE) and User/Proc counts. Is there a manual that details what fields in what hobbit status messages are recorded and graphed by the default hobbit installation? I would like to know more about what each of the default tests is supposed to show so that I can transform my system's data into the format hobbit is expecting. I would rather not report data to new test columns until I cannot wedge my data into a hobbit format, like a status for my system backups. /Thomas Kern /301-903-2211 FCX148 CPU 2086 SER xxxxx Interval 11:35:10 - 11:36:10 Remote Data CPU Load Vector Facility Status or PROC %CPU %CP %EMU %WT %SYS %SP %SIC %LOGLD %VTOT %VEMU REST ded. User P00 2 1 1 98 0 0 99 2 not installed Master Total SSCH/RSCH 7/s Page rate .0/s Priv. instruct. 29/s Virtual I/O rate 4/s XSTORE paging .0/s Diagnose instr. 18/s Total rel. SHARE 5600 Tot. abs SHARE 0% <-- Device Descr. --> Mdisk Pa- <-Rate/s-> <------- Time (msec) -------> Req. Addr Type Label/ID Links ths I/O Avoid Pend Disc Conn Serv Resp CUWt Qued 1D11 3390-9 VLNX02 1 4 1.0 .0 .4 .0 1.1 1.5 1.5 .0 .00 D936 3390-9 VML905 1 4 .5 .0 .3 .0 .5 .8 .8 .0 .00 D435 3390-9 VML908 1 4 .4 .0 .3 .0 .7 1.0 1.0 .0 .00 D235 3390-9 VML906 1 4 .4 .0 .3 .0 .7 1.0 1.0 .0 .00 CHPID Chan-Group <%Busy> <----- Channel %Busy Distribution 00:01:10-11:3 (Hex) Descr Qual Cur Ave 0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 C4 OSD 00 8 8 100 0 0 0 0 0 0 0 C5 OSD 00 8 8 100 0 0 0 0 0 0 0 Userid %CPU TCPU VCPU Ratio Total DASD Avoid Diag98 UR Pg/s User Status User Class Data: GENERAL .02 .010 .009 1.1 .1 .1 .0 .0 .0 .0 ---,---,--- LINUX .15 .089 .073 1.2 .4 .4 .0 .0 .0 .0 ---,---,--- User Data: NETTRACK .24 .142 .139 1.0 1.5 1.5 .0 .0 .0 .0 EME,CL0,DIS LNXSSS1 .19 .114 .094 1.2 .2 .2 .0 .0 .0 .0 ESA,CL0,DIS SLES9MNT .16 .095 .093 1.0 .9 .9 .0 .0 .0 .0 EME,CL0,DIS SLESXMNT .14 .084 .063 1.3 .0 .0 .0 .0 .0 .0 EME,CL0,DIS LINUX1 .10 .062 .043 1.4 .3 .3 .0 .0 .0 .0 ESA,CL0,DIS SSLSERV .09 .056 .037 1.5 .0 .0 .0 .0 .0 .0 ESA,CL0,DIS PERFSVM .08 .046 .045 1.0 .0 .0 .0 .0 .0 .0 ESA,---,DOR ORASRV1 .06 .034 .033 1.0 .0 .0 .0 .0 .0 .0 EME,---,DOR DIRMAINT .04 .025 .024 1.0 .2 .2 .2 .0 .0 .0 ESA,---,DOR DATAMOVE .03 .020 .019 1.1 .3 .2 .2 .0 .0 .0 ESA,---,DOR TCPIP .01 .004 .002 2.0 .0 .0 .0 .0 .0 .0 ESA,---,DOR

2 1

bb-cfg2html v1.0
by fduranti＠q8.it 17 Oct '06

17 Oct '06

Well.. If anyone need this... You can get a client side script to create configuration report of linux machine with cfg2html and send them to hobbit via notes protocol to hobbit server from http://sourceforge.net/projects/hobbit-perl-cl/ Francesco

1 0

Proper alert escalations question
by charles.jones＠ciscolearning.org 17 Oct '06

17 Oct '06

The goal here is for the secondary on-call person to have a restful nights sleep, assuming that the primary oncall person fixes the problem that caused the alert. * The primary should be alerted immediately upon a problem * The primary should get a RECOVERY page when the alert recovers * The secondary should ONLY get an alert if something is red for 30+ minutes * The secondary should get a RECOVERY page, ONLY if they were alerted in the first place (for a +30min event). * If the primary ACKs an alert, the secondary will not be emailed unless the ACK expires and the service is still red. If I use a ruleset of: HOST=www.foo.com SERVICE=http MAIL primary_oncall at foo.com FORMAT=sms COLOR=red RECOVERED MAIL secondary_oncall at foo.com FORMAT=sms COLOR=red DURATION>30 RECOVERED I do not believe this will work? Because the secondary will still get woke up by the RECOVERED message? Or is Hobbit smart enough that it only sends RECOVERED messages to only MAIL recipients that have previously received an alert?

3 2

rrd question
by gjohnson＠trantor.org 17 Oct '06

17 Oct '06

Hey, I have the graphing working just fine...however the rrd file is huge...below is a sample of the status message getting sent to hobbit. There are 16 terminal users on winserver1. Of the 16 total users: Active Sessions: 2 Disconnected Sessions: 14 Group1 Users: 7 Group2 Users: 9 USERNAME SESSIONNAME ID STATE IDLE TIME LOGON TIME user01 4 Disc none 9/18/2006 3:52 PM user02 1 Disc none 9/20/2006 3:33 PM user03 rdp-tcp#360 6 Active 6:09 9/22/2006 5:11 PM user04 12 Disc none 9/26/2006 9:15 AM user05 14 Disc none 9/27/2006 10:07 AM user06 15 Disc none 9/27/2006 2:52 PM user07 11 Disc none 9/27/2006 4:55 PM user08 9 Disc none 9/27/2006 5:37 PM user09 16 Disc none 9/28/2006 3:49 PM user10 rdp-tcp#359 2 Active . 10/2/2006 10:57 AM user11 5 Disc none 10/2/2006 2:56 PM user12 7 Disc none 10/6/2006 10:15 AM user13 8 Disc none 10/10/2006 9:21 AM user14 3 Disc none 10/11/2006 2:51 PM user15 18 Disc none 10/11/2006 3:50 PM user16 17 Disc none 10/12/2006 1:56 PM When I looked at the file sizes the termusers.rrd file is huge compared to the other rrd files for this same system, so I did a dump and looked at the content...it created a rrd entry for EVERY userXX as well...broken up in strange ways...regardless...I want it ignore everything except the 4 items above USERNAME...can I do this with NCV (which is what I'm currently using) or will it require a separate script? I'm going to continue digging into the docs and mailing list but thought I'd ask while doing so (in case you could save me some time). =G= Note, this is hobbit 4.2.0 (unpatched) and the client is BBWin with a custom script.

6 18

Notes protocol: working as expected?
by fduranti＠q8.it 17 Oct '06

17 Oct '06

Hi all, I'm trying to get the notes protocol to work to send data with cfg2html to the hobbit server... I don't know if those are known problems or is working as expected: 1) sending bb localhost "notes HOSTNAME.html pippopluto" Will not create any file it seems that the only way to make the notes protocol to work is to use only the HOSTNAME with no extention like bb localhost "notes HOSTNAME pippopluto" , is this right? 2) the file will not contain only the data sent to hobbit but also the "notes HOSTNAME" at the beginning. Is this normal too? Francesco

3 2

[hobbit] Notes protocol: working as expected?
by fduranti＠q8.it 17 Oct '06

17 Oct '06

> You're the first I've heard ever using the "notes" protocol. > So I wouldn't be surprised if there were bugs. I'm trying to use all the features hobbit can provide ;D The notes protocol can be useful to send host information with something like cfg2html Francesco

1 0

Problem with Hobbit graphs
by pierre.ronnefalk＠relacom.com 17 Oct '06

17 Oct '06

Hi, After installation i can´t se the trend graphs,, for example under disk....its only a red X Is there any hobbit forum anywhere ? /Pierre Ronnefalk [#################### RELACOM CORPORATE EMAIL FOOTER ####################] 1. This communication is confidential and is only intended for use of the individual or entity to which it is directed. If you are not the intended recipient of this email, please delete it without forwarding any of it's content. 2. Opinions and thoughts expressed in this email might be personal opinions from the sender and not necessary representing Relacom. 3. For further information please visit: http://security.relacom.com/email/ [############################################################################]

2 2

Menu Question
by shane.presley＠gmail.com 17 Oct '06

17 Oct '06

Hello, I'm customizing my menu (menu_items.js), and one of my categories has a long list of servers. The menu that is creates is longer than my browser length, and there are no scroll bars on the menu...so I can never reach the lower items. Aside from splitting this into multiple sub-menus, is there any way for the menu to scroll, or be automatically split into two rows? Thanks Shane

2 2

Hobbit_perl_client v1.15
by fduranti＠q8.it 17 Oct '06

17 Oct '06

Hi all, new release with some bugfix for beastat and dbcheck and some graph definition for extended statistics of netapp storage in the hobbitgraph.add file (you must patch hobbit server to get the data into rrd file). Changelog: beastat.pl 1.07 - Corrected errors while configuring a domain via "beastat.pl hostname port username password" It got broken some version ago while restructuring the code dbcheck.pl 1.08 - In Custom Check you can hide a column setting the relative column size to 0 - Corrected a problem with displaylegend - Corrected defaultalign value (was incorrectly defined)

1 0

integrating help messages in alert pages
by stephane.caminade＠ias.u-psud.fr 17 Oct '06

17 Oct '06

Hello, Is there a way to configure some text to appear on the alert page, indicating some procedure to be used to solve said alert ? Let's say that we have several administrators in a network, and that not all of them are completely up-to-date on what procedures one of the others uses. Could this other admin set up some kind of pointer on the service which he uses hobbit to monitor, either to link a URL with some documentation, or to add some text explaining briefly what should be done ? Thanks, Stephane -- _____________________________________________________________________________ Stephane Caminade Administrateur Systemes et Reseaux \ <Stephane.Caminade at ias.u-psud.fr> Institut d'Astrophysique Spatiale / tel : (33) (1) 69 85 87 03 Batiment 121, Universite Paris XI \ fax : (33) (1) 69 85 86 75 F-91405 ORSAY Cedex / www : http://www.ias.u-psud.fr/ _____________________________________________________________________________

2 1