September 2006 - Xymon - Xymon mailman web

[hobbit] log file monitoring issues
by gmbfly98＠gmail.com 02 Nov '06

02 Nov '06

Maybe I'm just missing something in the documentation, but I can't seem to get the log file monitoring to work properly. In the example below, I'm trying to look at the "messages" and "maillog" files on Linux. Particularly, I'm trying to EXCLUDE the following "messages" lines: Aug 9 21:19:45 www upsd[7860]: Connection from 127.0.0.1 Aug 9 21:19:45 www upsd[7860]: Client on 127.0.0.1 logged out Aug 9 21:19:45 www upsd[7860]: Connection from 127.0.0.1 Aug 9 16:44:01 www crond(pam_unix)[5382]: session opened for user root by (uid=0) Aug 9 16:44:14 www crond(pam_unix)[5382]: session closed for user root Aug 9 16:45:01 www crond(pam_unix)[5484]: session opened for user mailman by (uid=0) Aug 9 16:45:01 www crond(pam_unix)[5484]: session closed for user mailman And EXCLUDE the following "maillog" lines: Aug 6 11:55:02 www sendmail[15076]: k76Ft1pU015076: from=<mailman at HOSTNAME>, size=576, class=0, nrcpts=1, msgid=<200608061555.k76Ft1A2015075 at HOSTNAME>, proto=ESMTP, daemon=MTA, relay=localhost.localdomain [127.0.0.1] Below is the respective lines from the "client-local.cfg" file: log:/var/log/messages:10240 ignore upsd* Client|Connection 127.0.0.1 ignore session opened|closed for user mailman|root log:/var/log/maillog:10240 ignore relay=localhost.localdomain trigger denied And below the specific log entries I'm looking for from "hobbit-clients.cfg": LOG /var/log/maillog "relaying denied" color="yellow" Now, the problem I'm having... The "ignore" line for the /var/log/maillog file appears to be working correctly, as it does indeed ignore such entries as shown above. Also working is the "ignore session opened..." line for the /var/log/messages file. What is NOT working is the "ignore" line for the "upsd*" lines in /var/log/messages. For the life of me, I just can't figure out how to get that to work properly. That is, two of the three "ignore" lines are not working, as those lines still show up in the "full log" output. If anyone has any ideas, let me know. I'm also having problems with some logs not showing up on the messages page. Do you need both a "LOG" entries in the hobbit-clients.cfg AND client-local.cfg, or will an entry in only client-local.cfg be sufficient to have it show up on the messages page?

5 17

Somewhat off-topic - the GPL v2/v3 debate, and Hobbit
by henrik＠hswn.dk 11 Oct '06

11 Oct '06

As You may have heard, there is currently a rather intense debate going on about the proposed new version 3 of the GNU General Public License (the GPL). Hobbit uses the GPL license - and specifically, it uses the current version 2. A change to version 3 of the GPL therefore cannot happen automatically - it has to be a deliberate decision by me (and anyone else who has contributed code to Hobbit) to do so. The proposed GPL v3 adds some rather troubling restrictions to the license, especially when it comes to using GPL-licensed code in systems that include various DRM (Digital Rights Management) mechanisms. It also poses some legal challenges to companies that obtain patents on their work, which could very well mean that they would not be able to use GPLv3 licensed software at all. I fundamentally believe that DRM and patents - especially patents on computer software - is evil, and I would never buy anything that comes with such restrictions on what I can use it for. However, the GPL v3 proposal is too much of a "fight fire with fire" agenda for me. Whether our society should permit patents is a political discussion, not a software-development issue. So it should be handled politically, not in a software license. DRM restrictions can be dealt with simply through normal market mechanisms - if you dislike them, dont buy products that have them. Or make something that is better and release it in an Open Source way. I am not a lawyer, so I may have misunderstood quite what the proposed changes mean - but if I have, then several other software developers whose judgment I respect have also misunderstood them. Most notably, a large collection of the Linux kernel developers - including Linus Torvalds - strongly oppose the GPL v3 changes and have recently published a discussion document detailing why they believe these changes are wrong. http://permalink.gmane.org/gmane.linux.kernel/448894 When the second draft of GPLv3 was published, Linus also participated in a lengthy discussion explaining his views on the Groklaw blog. If you're interested, see the (long) "DRM misunderstood" thread starting at http://www.groklaw.net/article.php?story=20060727140038810 and especially http://www.groklaw.net/comment.php?mode=display&sid=20060727140038810&title… All that just to explain my position: As far as I am concerned there will not be a change of license for Hobbit, at least not with the current wording of the GPLv3. Just in case anyone cared. Regards, Henrik

4 5

Running logfetch 'by hand'
by Andrew.Jackson＠ordnancesurvey.co.uk 10 Oct '06

10 Oct '06

Hi, First, thanks for hobbit. It is making my life much easier. I'm trying to run logfetch by hand as it were. It takes two files logfetch.<machine>.cfg and logfetch.<machine>.status as arguments. I can see that logfetch.<machine>.cfg is generated when you submit data to the hobbit server but I can't find out where the logfetch.<machine>.status file is generated. Any help gratefully received! Thank Andy J. . This email is only intended for the person to whom it is addressed and may contain confidential information. If you have received this email in error, please notify the sender and delete this email which must not be copied, distributed or disclosed to any other person. Unless stated otherwise, the contents of this email are personal to the writer and do not represent the official view of Ordnance Survey. Nor can any contract be formed on Ordnance Survey's behalf via email. We reserve the right to monitor emails and attachments without prior notice. Thank you for your cooperation. Ordnance Survey Romsey Road Southampton SO16 4GU Tel: 023 8079 2000 http://www.ordnancesurvey.co.uk

2 3

Small enhancement to Hobbit JavaScript menu
by frederic.mangeant＠steria.com 05 Oct '06

05 Oct '06

Hi Henrik I didn't find bb-datepage.sh the first time I set up Hobbit, maybe this small modification to ~/server/www/menu/menu_items.js would be helpful : --- menu_items.js.orig 2005-11-18 15:46:58.000000000 +0100 +++ menu_items.js 2005-11-18 15:53:50.000000000 +0100 @@ -6,7 +6,12 @@ ], ['Reports', null, null, ['Event log Report', '/hobbit-cgi/bb-eventlog.sh'], - ['Availability Report', '/hobbit-cgi/bb-rep.sh'], + ['Availability Reports', null, null, + ['Daily Report', '/hobbit-cgi/bb-datepage.sh?type=day'], + ['Weekly Report', '/hobbit-cgi/bb-datepage.sh?type=week'], + ['Monthly Report', '/hobbit-cgi/bb-datepage.sh?type=month'], + ['Custom Report', '/hobbit-cgi/bb-rep.sh'], + ], ['Snapshot Report', '/hobbit-cgi/bb-snapshot.sh'], ['Config Report', '/hobbit-cgi/hobbit-confreport.sh'], ], Regards, -- Frédéric Mangeant Steria EDC Sophia-Antipolis

3 2

Hobbitd_client crash
by dunne＠accuweather.com 03 Oct '06

03 Oct '06

Greetings, The hobbitd_client module has been crashing on and off all day. Below is the gdb output. Any advice would be greatly appreciated. This GDB was configured as "powerpc-apple-darwin". Reading symbols for shared libraries .. done #0 0x9004312c in kill () (gdb) bt #0 0x9004312c in kill () #1 0x9009fb9c in abort () #2 0x00016bdc in receive_samples () #3 <signal handler called> #4 0x90007260 in strlen () #5 0x0000ec68 in error_message () #6 0x00002dc8 in main (argc=1192252, argv=0xa0005290) at bb.c:138 (gdb) quit Thanks, Mike Michael Dunne Network Support Specialist AccuWeather, Inc. ~HISTORY, n. An account mostly false, of events mostly unimportant, which are brought about by rulers mostly knaves, and soldiers mostly fools. (Ambrose Bierce, "The Devil's Dictionary")~

3 9

[hobbit] pb with hobbit-perl-client and dbcheck
by fduranti＠q8.it 03 Oct '06

03 Oct '06

Some more information from the README of DBD::Oracle 1.18a From: John Scoles <scoles at pythian.com> Date: Fri, 29 Sep 2005 10:48:47 -0700 (EST) Subject: RE: Oracle 10g Instantclient The Makefile.PL will now work for Oracle 10g Instantclient. To have both the Compile and the test.pl to work you must first have the LD_LIBRARY_PATH correctly set to your "instantclient" directory. (http://www.oracle.com/technology/tech/oci/instantclient/instantclient.h tml) The present version of the make creates a link on your "instantclient" directory as follows "ln -s libclntsh.so.10.1 libclntsh.so". It is needed for both the makefile creation and the compile but is not need for the test.pl. It should be removed after the compile. If the Makefile.PL or make fails try creating this link directly in your "instantclient" directory. Hoping this could solve the problem... > -----Original Message----- > From: Olivier Boyaval [mailto:oboyaval at ch-bethune.fr] > Sent: Tuesday, September 19, 2006 10:00 AM > To: hobbit at hswn.dk > Subject: RE: [hobbit] pb with hobbit-perl-client and dbcheck > > Hello, > > > > * /To/: <hobbit (at) hswn.dk> > > * /Subject/: RE: [hobbit] pb with hobbit-perl-client and dbcheck > > * /From/: "Francesco Duranti" <fduranti (at) q8.it> > > * /Date/: Tue, 19 Sep 2006 00:04:57 +0200 > > * /Thread-index/: AcbbXNjiyCoEBWDISL+6vz/9wLlMUwAD3t5g > > * /Thread-topic/: [hobbit] pb with hobbit-perl-client > and dbcheck > > > > > ---------------------------------------------------------------------- > > -- I've not tested the dbcheck.pl with the InstantClient > because I've > > a full 10r2 client installed on the machine but checking on some > > mailing list archive it seems that InstantClient and > DBD::Oracle have > > some problems working together.... > > One suggestion I've found was to set the Oracle_home (in the > > dbcheck.ini file) to the lib directory instead of the root > into which > > the instant client is installed (something like > > /usr/lib/oracle/10.2.0.2/client/lib)... > > > With Instantclient, there's not lib directory, my ORACLE_HOME > and my LD_LIBRARY_PATH are OK > > Another thing is to check that you have the following package > > installed (took from the DBD:Oracle readme) > > * The "Basic" package for the essential Oracle libraries. > > * The "SDK" package for the headers and makefile. > > * The "SQL*Plus" component is optional, but will help you > check your > > configuration and DBD::Oracle determine your Oracle version. > > > it's OK for this package > > Then you can try to define the full > hostname/port/sid/user/password in > > the dbcheck.ini file under the db you want to configure (it should > > skip the tnsnames.ora and go through to the db server). Or > you can try > > to define your db in a tnsnames.ora under > $ORACLE_HOME/network/admin > > and put only sid username and password in the dbconfiguration. > > > I tested with bbcmd and sqlplus, I can connect with the > database I tested with bbcmd and the next perl script and I > can also connect with the database > > ---- > use DBI; > $dbh = > DBI->connect("dbi:Oracle:host=server;sid=SID;port=1521", > "user", "password"); > ---- > > > What version of DBD::Oracle module are you using? Did it > compile/test > > successfully? Can you check if with sqlplus you can connect > correctly > > to that database? > > > DBD::Oracle 1.18 > I have detected a pb with the Makefile.PL of DBD : > > ----------- > Configuring DBD::Oracle for perl 5.008003 on linux > (x86_64-linux-thread-multi) > > Remember to actually *READ* the README file! Especially if > you have any problems. > > Using Oracle in /opt/instantclient > SP2-1503: Unable to initialize Oracle call interface > SP2-0152: ORACLE may not be functioning properly > > If sqlplus failed due to a > linker/symbol/relocation/library error or similar problem > then it's likely that you've not configured your > environment correctly. > Specifically, your LD_LIBRARY_PATH environment variable > set to include the directory containing the > Oracle libraries. > > I'm having trouble finding your Oracle version number... trying harder > > WARNING: I could not determine Oracle client version so I'll > just default to version 8.0.0.0. Some features of DBD::Oracle > may not work. > Oracle version based logic in Makefile.PL may produce > erroneous results. > You can use "perl Makefile.PL -V X.Y.Z" to specify a your > client version. > > Oracle version 8.0.0.0 (8.0) > Looks like an Instant Client installation, okay Your > LD_LIBRARY_PATH env var is set to '/opt/instantclient/' > ----------- > > I will test with "perl Makefile.PL -V 10.2.0" > > Francesco > Cordialement > -- > Olivier Boyaval > > > > To unsubscribe from the hobbit list, send an e-mail to > hobbit-unsubscribe at hswn.dk > > >

2 2

Hobbit_perl_client 1.13
by fduranti＠q8.it 03 Oct '06

03 Oct '06

New version is out with 3 nice new features: 1) DB2 connection and tablespace report. 2) Report of what the script did in a column called like the script on the host where the script is running (all errors and warning will also be reported here). 3) Custom query engine to write your own query (If you want to share them you can put them on the hobbit_perl_client forum on sourceforge or send them to me and I'll try to put them there for everyone to use). As always it can be downloaded http://sourceforge.net/projects/hobbit-perl-cl/ or from http://www.deadcat.net The custom query engine in a bit experimental so be sure to try your query and check all the parameters. The documentation is not so good but you can find how it work in the example ini files (I think I'll finish to write some documentation before starting to put more functionality into the scripts :D) so If you need any info feel free to contact me via mail/forum. Francesco Changes to this version: Hobbit_fd_lib.pm 1.13 (general library) - now the scripts will send a status report to hobbit into a column on the server where the scripts are running to do a summary of what was done (total and host running time, total and color/event of messages sent, total host checked and test done for each host, a list of warning and error got from the scripts). For dbcheck if you use the fast/slow kind of check you will have 2 columns one called dbcheck_fast and one dbcheck_slow. - new general (for the 3 scripts) option warningout that can be equal to "log", "send", "both" and will influence how warning are treated. If it's "log" it will simply out the warning to the hobbit log file, "send" will send the warning into the status report of the test, "both" will do both the things (check ini files to see hot this work). Beastat 1.05 - Added an option to check version of program and modules used (beastat.pl -v) - Corrected some bugs Dbcheck 1.06 - Added a custom query engine to enable the user to create new tests for all databases - Added an option to check version of program and modules used (dbcheck.pl -v) - Added test for IBM-DB2 Database: connection and tablespace check (see dbcheck.ini for configuration information) - Added uptime check for all database types - Changed the locks checking code to get more data about the sql command that are blocked and the process number and os command name of the blocker/blocked statement - Corrected a bug with informix table checks - Corrected a problem with mysql that could not be used as a default database type without defining the hostname of every server - Various bug fix Netapp 1.09 - Some fixes on the code to execute the sysstat command reporting data about cpu and system statistics. - Added an option to check version of program and modules used (netapp.pl -v)

2 2

LOG test fails to IGNORE ?
by jjj863＠gmail.com 02 Oct '06

02 Oct '06

hi, I have one single Hobbit server running 4.2RC1-20060712 on CentOS 4. Clients have some harmless SELinux warning messages in /var/log/messages. I configured Hobbit to ignore such lines (conf entries appended). However, the LOG test still generates alerts. Below is excerpted from an email alert. Any idea why? I brought up before 4.2 final version is released, I believe, and was told it should not be the case. &yellow Sep 28 15:58:20 saturn su[4070]: Warning! Could not relabel with user_u:object_r:devpts_t, not relabeling. ---cut---x------ hobbit-clients.cfg -----x-------cut----- HOST=saturn log /var/log/messages %WARNING|NOTICE|ERROR COLOR=yellow IGNORE=relabeling ---cut---x------ clients-local.cfg -----x-------cut----- [saturn] log:/var/log/messages:10240 ignore MARK

2 6

Change request: FNPATTERN matching string
by fduranti＠q8.it 30 Sep '06

30 Sep '06

Hi Henrik, just another little request. The FNPATTERN is matched as a regular expression and this could is some situation cause problems matching the right file. For example I'm creating some rrd named xtradisk,diskname.rrd and they get matched with the normal disk test and I start to not get any disk graphs from disk and trend column. The simple solution is to put a "^" in front of the disk in the hobbitgraph.cfg. This is good for this test until the next one that will try to get wrong data. >From how the rrd are organized it make (at least to me) sense to have the FNPATTERN match directly against ^somedefinition because all graphs defined until now have the pattern in the form of testname_followed_by_something_else. The line responsible for the FNPATTERN is line 286 of web/hobbitgraph.c and changing it from: newitem->fnpat = strdup(p); To newitem->fnpat=malloc(strlen(p)+2); strcpy(newitem->fnpat,"^"); strcat(newitem->fnpat,p); Will do the trick (at least it seem to work). If in the future there's the necessity to define a pattern in the middle of the file it will be simply matched with a FNPATTER .*something so it should not create problems to other tests/graphs but will definitely create less problem with disk or data having something like disk inside of them. Francesco

1 0

Bug in rrd/do_disk.c while using /GROUP option of bb command
by fduranti＠q8.it 30 Sep '06

30 Sep '06

Hi Henrik, I got some problems with RRD creation using the "/group:groupname" directive. The do_disk.c will skip every line that don't contain a "/" and the string " red " or " yellow ". This will work if the test are in error but it will not work if the test are not red or yellow. The first line of the message sent with the group directive is something like status+60/GROUP:netappadmin VITROMFS65.disk green Sat Sep 30 16:59:17 2006 And this will be carried on generating a rrd file with a strange name. This can be avoided skipping the first like changing line 54 of hobbitd/rrd/do_disk.c from curline = msg; To curline = strchr(msg,'\n'); if (curline) curline++; Just to skip the first line. Francesco

1 0