[checkmk-commits] Check_MK Git: check_mk: Updated bug entries #0736, #0733, #0734, #0735

Module: check_mk Branch: master Commit: cdd4c2efd50bd25c66cd2cbfdffb57b6d14ff9cd URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=cdd4c2efd50bd2… Author: Bastian Kuhn &lt;bk(a)mathias-kettner.de&gt; Date: Wed Aug 8 11:33:58 2012 +0200 Updated bug entries #0736, #0733, #0734, #0735 --- .bugs/733 | 9 +++++++++ .bugs/734 | 45 +++++++++++++++++++++++++++++++++++++++++++++ .bugs/735 | 10 ++++++++++ .bugs/736 | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 110 insertions(+), 0 deletions(-) diff --git a/.bugs/733 b/.bugs/733 new file mode 100644 index 0000000..09f290a --- /dev/null +++ b/.bugs/733 @@ -0,0 +1,9 @@ +Title: Turnoff context in logwatch +Component: checks +State: open +Date: 2012-08-06 16:02:17 +Targetversion: future +Class: feature + +It should be possible to turn the saving of context logentries from logwatch off. +In some cases are there hunderts of not needed entries and the 500kb limit are full with only one error line diff --git a/.bugs/734 b/.bugs/734 new file mode 100644 index 0000000..a61c8a9 --- /dev/null +++ b/.bugs/734 @@ -0,0 +1,45 @@ +Title: Logwatch, logfile names with date +Component: checks +State: open +Date: 2012-08-08 11:15:14 +Targetversion: future +Class: feature + +We have a requirement from our developers to monitor log files in the format: /path/logfile.20120719 with the date changing every day. +To do this, we added code to the parse and change the name in the config file: + +# cat /etc/check_mk/logwatch.cfg +/var/log/test.DATE +C test + +Now to interpret it, we changed the mk_logwatch + +[root@temolate2 ~]# diff mk_logwatch.newest /usr/lib/check_mk_agent/plugins/mk_logwatch +30,35d29 +< from datetime import date +< dd=date.today().strftime("%Y%m%d") +< yyear = date.today().strftime("%Y") +< year = date.today().strftime("%y") +< month = date.today().strftime("%m") +< day = date.today().strftime("%d") +278,284d270 +< gsave = glob +< glob=glob.replace('DATE',dd) +< glob=glob.replace('YYYY',yyear) +< glob=glob.replace('YY',year) +< glob=glob.replace('MM',month) +< glob=glob.replace('DD',day) +< +294c280 +< process_logfile(logfile, patterns, gsave) +--- +> process_logfile(logfile, patterns) + +Could you please add this feature to the next version of the logwatch plugin. + +Thanks! + +Best Regards, + +Roman Lazarev +Information Technology Division diff --git a/.bugs/735 b/.bugs/735 new file mode 100644 index 0000000..2ca4f76 --- /dev/null +++ b/.bugs/735 @@ -0,0 +1,10 @@ +Title: Notes field for Activate Changes in Wato +Component: wato +State: open +Date: 2012-08-08 11:18:29 +Targetversion: future +Class: feature + +* Add a notes field associated with an activate changes (auditing) - we would like to be able to say such and such a change was related to our ticketing system ticket xxxx, for instance + +o Optional - make it a choice to make the field required diff --git a/.bugs/736 b/.bugs/736 new file mode 100644 index 0000000..f575c61 --- /dev/null +++ b/.bugs/736 @@ -0,0 +1,46 @@ +Title: Livechecks goes zoombie +Component: livecheck +State: open +Date: 2012-08-08 11:32:12 +Targetversion: future +Class: bug + +Have been using livecheck and came across this rather nasty bug. We +use livecheck for a whole bunch of http checks (check_http) and after +a while, one of the checks will "hang" and the livecheck process +becomes a "Zombie" process. + +# ps faux +nagios 10880 12.4 2.0 573000 12380 ? Ssl 12:16 16:19 +/usr/bin/nagios -d /etc/nagios/nagios.cfg +nagios 10889 0.0 0.1 11848 1036 ? S 12:16 0:00 \_ +livecheck /var/spool/nagios/checkresults 30 60 +nagios 4100 0.0 0.0 0 0 ? Z 13:44 0:00 | +\_ [livecheck] +nagios 4120 0.0 0.5 138664 3504 ? S 13:44 0:01 +/usr/lib64/nagios/plugins/check_http -H xxx -t 15 -f follow -S + +# strace -p 10880 +Process 10880 attached - interrupt to quit +read(15, + +# strace -p 10889 +Process 10889 attached - interrupt to quit +read(4, + +# strace -p 4120 +Process 4120 attached - interrupt to quit +futex(0x7f996e8bde80, FUTEX_WAIT_PRIVATE, 2, NULL + +Nagios completely stops doing anything, when you kill the check_http +process, Nagios comes back to life: + +nagios.log: +[1343918059] Warning: A system time change of 0d 0h 47m 11s (forwards +in time) has been detected. Compensating... + + + + + +Reference in Ticket System: #110644