27 Aug
2013
27 Aug
'13
12:15 p.m.
Module: check_mk
Branch: master
Commit: beb8d0f1ba4675f739a81eaad7e542c21bf9ccc3
URL:
http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=beb8d0f1ba4675…
Author: Bernd Stroessenreuther <bs(a)mathias-kettner.de>
Date: Tue Aug 27 12:14:48 2013 +0200
Updated bug entries #1034, #1035, #1036, #1037, #1038
---
.bugs/1034 | 58 ++++++++++++++++++++++++++++++++++++++++++++
.bugs/1035 | 57 +++++++++++++++++++++++++++++++++++++++++++
.bugs/1036 | 37 ++++++++++++++++++++++++++++
.bugs/1037 | 29 ++++++++++++++++++++++
.bugs/1038 | 79 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
5 files changed, 260 insertions(+)
diff --git a/.bugs/1034 b/.bugs/1034
new file mode 100644
index 0000000..7e0c95d
--- /dev/null
+++ b/.bugs/1034
@@ -0,0 +1,58 @@
+Title: modify check_mk_agent.linux - heartbeat_crm test for clusters with pacemaker since
1.1.9
+Component: core
+State: open
+Date: 2013-08-27 10:22:15
+Targetversion: 1.2.3
+Class: todo
+
+came in via feedback@ // original mail still available at feedback/gitbug folder
+
+
+From: Stefan Botter <jsj(a)jsj.dyndns.org>
+To: feedback(a)check-mk.org
+Subject: [PATCH] - modify check_mk_agent.linux - heartbeat_crm test for clusters with
pacemaker since 1.1.9
+Date: Tue, 28 May 2013 09:45:22 +0200
+
+
+From 13d33c76eb02b87e6f816a67dba393b586e9d4bf Mon Sep 17 00:00:00 2001
+From: Stefan Botter <jsj(a)jsj.dyndns.org>
+Date: Tue, 28 May 2013 09:38:06 +0200
+Subject: [PATCH 1/1] change heartbeat_crm: since pacemaker 1.1.9 there
+are no
+ /var/run files, also use crm status instead of crm_mon
+
+
+Signed-off-by: Stefan Botter <jsj(a)jsj.dyndns.org>
+---
+ agents/check_mk_agent.linux | 9 +++++++++
+ 1 file changed, 9 insertions(+)
+
+diff --git a/agents/check_mk_agent.linux b/agents/check_mk_agent.linux
+index 335ba79..118fa41 100755
+--- a/agents/check_mk_agent.linux
++++ b/agents/check_mk_agent.linux
+@@ -371,9 +371,18 @@ fi
+ # Heartbeat monitoring
+ # Different handling for heartbeat clusters with and without CRM
+ # for the resource state
++# pacemaker since 1.1.9 has no /var/run files, ugly detect FTB
++newPM=0
++for n in /dev/shm/qb-cib_rw-response-*-data; do if [ -f $n ]; then
+newPM=1; break; fi; done
+ if [ -S /var/run/heartbeat/crm/cib_ro -o -S /var/run/crm/cib_ro ]; then
+ echo '<<<heartbeat_crm>>>'
+ crm_mon -1 -r | grep -v ^$ | sed 's/^ //; /^\sResource Group:/,$
+s/^\s//; s/^\s/_/g'
++elif [ $newPM != 0 ]; then
++ echo '<<<heartbeat_crm>>>'
++ # pacemaker since 1.1.9
++ # additionally, in that case use crm status, as crm_mon segfaults
++ # also add line for identification of resources in agent output,
+which is missing from crm status
++ crm status | sed -e '/^Online:/{:a;n;/^$/!ba;i\Full list of
+resources:' -e '}' | grep -v ^$ | sed 's/^ //; /^\sResource Group:/,$
+s/^\s//; s/^\s/_/g'
+ fi
+ if type cl_status > /dev/null 2>&1; then
+ echo '<<<heartbeat_rscstatus>>>'
+-- 1.7.10.4
diff --git a/.bugs/1035 b/.bugs/1035
new file mode 100644
index 0000000..765fe9b
--- /dev/null
+++ b/.bugs/1035
@@ -0,0 +1,57 @@
+Title: make nfsexports plugin work with rpcbind too
+Component: checks
+State: open
+Date: 2013-08-27 10:31:53
+Targetversion: 1.2.3
+Class: nastiness
+
+came in via checkmk-en@ // mail still avialable at feedback/gitbug folder
+
+
+Date: Mon, 03 Jun 2013 19:25:04 +0200
+From: morlix <morlix(a)morlix.de>
+To: checkmk-en(a)lists.mathias-kettner.de
+Subject: [Check_mk (english)] [PATCH] FIX nfsexports plugin: enhance daemon
+ process detection
+
+Hello,
+
+today i added the nfsexports plugin to my check_mk_agent and wondered
+why it doesn't find any exports.
+
+After looking into the script i knew the problem.
+The script just looks for portmap but one some systems this daemons has
+the name rpcbind.
+
+So i created this patch to check for both variants.
+
+Kind regards,
+
+morlix
+
+
+0001-FIX-nfsexports-plugin-enhance-daemon-process-detecti.patch
+
+From d917306e23396baf7ae4f7bfc94c95a1d3fc2c15 Mon Sep 17 00:00:00 2001
+From: morlix <morlix(a)morlix.de>
+Date: Mon, 3 Jun 2013 19:19:53 +0200
+Subject: [PATCH] FIX: nfsexports plugin: enhance daemon process detection
+
+---
+ agents/plugins/nfsexports | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/agents/plugins/nfsexports b/agents/plugins/nfsexports
+index 185de07..d5ca96d 100755
+--- a/agents/plugins/nfsexports
++++ b/agents/plugins/nfsexports
+@@ -7,7 +7,7 @@
+ if [ -r /etc/exports ]; then
+ EXPORTS=$(grep -v -e ^# -e ^$ /etc/exports)
+ fi
+-pgrep portmap 1>/dev/null && pgrep rpc.mountd 1>/dev/null &&
DAEMONS="ok"
++pgrep '(portmap|rpcbind)' 1>/dev/null && pgrep rpc.mountd
1>/dev/null && DAEMONS="ok"
+
+ # any exports or have running daemons? then look for registered exports
+ if [[ $EXPORTS ]]; then
+-- 1.8.1.2
diff --git a/.bugs/1036 b/.bugs/1036
new file mode 100644
index 0000000..43329c7
--- /dev/null
+++ b/.bugs/1036
@@ -0,0 +1,37 @@
+Title: df_check_filesystem_single returns a 2-tuple when check_esx_vsphere_datastores
expects 3
+Component: checks
+State: open
+Date: 2013-08-27 10:50:42
+Targetversion: 1.2.3
+Class: bug
+
+came in via feedback@ // mail is still available at feedback/gitbug folder
+
+
+From: Clynt Heikes <cheikes(a)peak6.com>
+To: "'feedback(a)check-mk.org'" <feedback(a)check-mk.org>
+Subject: [Bug]:df_check_filesystem_single returns a 2-tuple when
+ check_esx_vsphere_datastores expects 3
+
+
+def df_check_filesystem_single(hostname, mountpoint, size_mb, avail_mb, params):
+
+ if size_mb == 0:
+
+ return (1, "size of filesystem is 0 MB")
+
+
+
+
+
+def check_esx_vsphere_datastores(item, params, info):
+
+…
+
+…
+
+…
+
+ state, infotext, perfdata = df_check_filesystem_single(
+
+ g_hostname, item, size_mb, avail_mb, params)
diff --git a/.bugs/1037 b/.bugs/1037
new file mode 100644
index 0000000..1061495
--- /dev/null
+++ b/.bugs/1037
@@ -0,0 +1,29 @@
+Title: check_mk_agent.exe unable to restart after a crash: logstate.txt 0 bytes
+Component: checks
+State: open
+Date: 2013-08-27 10:58:29
+Targetversion: 1.2.3
+Class: bug
+
+came in via feedback@ // original mail still available in feedback/gitbug folder
+
+
+Date: Fri, 07 Jun 2013 08:56:13 -0400
+From: "Peter Rogers" <peterrogers(a)email.com>
+Message-ID: <20130607125613.34790(a)gmx.com>
+MIME-Version: 1.0
+Subject: check_mk_agent.exe unable to restart after a crash
+To: feedback(a)check-mk.org
+
+
+Host = Windows 7
+CMK Agent = Agent version 1.2.2p1
+
+I have found a problem where the check_mk_agent.exe is unable to restart after a crash
because the logstate.txt file becomes currupt.
+
+Attached is an example currupt logstate.txt file.
+
+To work around the problem I remove the logstate.txt and then start the
check_mk_agent.exe successfully.
+
+
+hint by bs: the attached logstate.txt is 0 bytes
diff --git a/.bugs/1038 b/.bugs/1038
new file mode 100644
index 0000000..99e9d80
--- /dev/null
+++ b/.bugs/1038
@@ -0,0 +1,79 @@
+Title: problem with mrpe if lable has white space characters
+Component: checks
+State: open
+Date: 2013-08-27 12:12:33
+Targetversion: 1.2.3
+Class: bug
+
+came in via feedback@ // original mail still available in feedback/gitbug folder
+
+
+Date: Wed, 26 Jun 2013 15:03:06 +0200
+Message-ID: <CACA1nGEzOmECJufVGOgUqZm6tAxGH+vHoXoEh7SGbx7uw+3ZwQ(a)mail.gmail.com>
+Subject: Fwd: [BUG]: MRPE Checks mit Performance Daten bringen Fehler
+From: =?ISO-8859-1?Q?Andreas_D=F6hler?= <andreas.doehler(a)gmail.com>
+To: Check MK Feedback <feedback(a)check-mk.org>
+
+
+
+Hallo zusammen,
+
+habe heute eine seltsame Sache mit einigen MRPE Checks gehabt.
+Falls die Checks Performance Daten liefern und das Label der Performance Daten
Leerzeichen enthält so wir der Check mit einer Exception abgebrochen.
+
+Traceback (most recent call last):
+ File "/usr/share/check_mk/modules/check_mk.py", line 5442, in <module>
+ do_check(hostname, ipaddress, check_types)
+ File "/usr/share/check_mk/modules/check_mk_base.py", line 888, in do_check
+ agent_version, num_success, error_sections, problems =
do_all_checks_on_host(hostname, ipaddress, only_check_types)
+ File "/usr/share/check_mk/modules/check_mk_base.py", line 1197, in
do_all_checks_on_host
+ result = check_function(item, params, info)
+ File "/usr/share/check_mk/checks/mrpe", line 89, in check_mrpe
+ perf_parsed = map(mrpe_parse_perfdata, perfdata)
+ File "/usr/share/check_mk/checks/mrpe", line 41, in mrpe_parse_perfdata
+ varname, valuetxt = perfinfo.split("=", 1)
+ValueError: need more than 1 value to unpack
+
+Bei genauerer Untersuchung hab ich festgestellt das dieses Verhalten seit dem Umbau des
MRPE am 22.05. (Commit 7705017d0d5db5093d9bdaa1538eecac56832dfe) auftritt.
+Das Problem besteht aus zwei Punkten.
+Einmal diese Zeile
+perfdata = parts[1].strip().split()
+Hierbei werden nicht nur vorn und hinten die Leerzeichen abgeschnitten sondern auch der
String vom Performancedaten Label zerschnitten.
+Beispiel:
+'Cluster Groups on This Node'=0;0;0 Perfdaten noch ok
+["'Cluster", 'Groups', 'on', 'This',
"Node'=0;0;0"] nach der "Behandlung"
+
+Der zweite Punkt tritt dann auf wenn ich den String hier im ganzen lasse.
+'Cluster Groups on This Node'=0;0;0
+und es nur einen Perfomance Daten Wert gibt dann wird perfdata nicht zur Liste und bringt
einen Fehler dann beim Aufruf der map Funktion in Zeile 90
+Deshalb hab ich extra noch vor der ersten perfdata Zuweisung in Zeile 70 eine Zeile
eingefügt mit perfdata = [] und Zeile 70 auf perfdata += parts[1].strip() geändert.
+Danach funktionieren wieder alle meine MRPE Checks mit Performance Daten ohne Fehler.
+
+Einen MRPE Check mit mehreren Werten hatte ich leider nicht an der Hand zum testen.
+Damit ist der Abschnitt ab # Further lines noch offen ob das alles so klappt. :)
+
+Gruß
+Andreas
+
+
+
+mrpe.diff
+
+--- a/mrpe 2013-06-26 10:05:15.000000000 +0200
++++ b/mrpe 2013-06-26 14:59:54.000000000 +0200
+@@ -57,6 +57,7 @@
+ line = line[1:]
+ else:
+ check_name = None
++ perfdata = []
+ if line[0] == item:
+ state = int(line[1])
+ # convert to original format by joining and replacing \1 back with \n
+@@ -67,7 +68,7 @@
+ parts = lines[0].split("|", 1)
+ output = [parts[0].strip()]
+ if len(parts) > 1:
+- perfdata = parts[1].strip().split()
++ perfdata.append(parts[1].strip())
+ else:
+ perfdata = []