Module: check_mk
Branch: master
Commit: 3139a69a455a7738a6810c9c8609996b674bce23
URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=3139a69a455a77…
Author: Andreas Boesl <ab(a)mathias-kettner.de>
Date: Wed Dec 4 11:17:49 2013 +0100
vsphere monitoring: new option to skip placeholder vms in agent output
When using a Site Recovery Manager (SRM) some of the virtualmachines are only
used as fallback. The new agent_vsphere option -p allows to suppress any output
regarding placeholder vms.
Right now this option only works when querying a host system.
Placeholder vms are identified by not having an assigned VirtualDisk Device.
The option is also configurable via the WATO rule "Check state of VMWare ESX via vSphere"
---
.werks/238 | 13 +++++++
ChangeLog | 1 +
agents/special/agent_vsphere | 58 +++++++++++++++++++------------
checks/agent_vsphere | 2 ++
web/plugins/wato/datasource_programs.py | 10 +++++-
5 files changed, 61 insertions(+), 23 deletions(-)
Diff: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commitdiff;h=3139a69a45…
Module: check_mk
Branch: master
Commit: f1e8a65ab8664acdaee4347ff30523486e40ca45
URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=f1e8a65ab8664a…
Author: Lars Michelsen <lm(a)mathias-kettner.de>
Date: Wed Dec 4 10:12:34 2013 +0100
Updated bug entries #2094
---
.bugs/2094 | 23 +++++++++++++++++++++++
1 file changed, 23 insertions(+)
diff --git a/.bugs/2094 b/.bugs/2094
new file mode 100644
index 0000000..f8fe841
--- /dev/null
+++ b/.bugs/2094
@@ -0,0 +1,23 @@
+Title: BI-Cache vs. Apache-Memory - make debugging easier
+Component: docu
+State: open
+Date: 2013-12-04 09:54:12
+Targetversion: 1.2.3i6
+Class: todo
+
+It is hard to diagnose memory problems caused by the BI-Cache which increases
+the memory of the sites apache processes dramatically. While building the cache,
+there are often opened a lot of other apache processes, which might lead into
+system memory issues, swapping etc.
+
+There is no single value/check to point the user to this issue/solution.
+
+It would be nice to have a clearer view on the number of apache processes and
+their consumed memory.
+
+At the moment it is neccessary to install the apache_status agent plugin
+and create a ps.perf check which monitors the site apaches processes to get
+all information.
+
+Another soultion would be to reduce the number of maximum allowed apache
+processes within OMD.
Module: check_mk
Branch: master
Commit: 6a8fd59512a095d90ad39ae12b683d543031ace9
URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=6a8fd59512a095…
Author: Mathias Kettner <mk(a)mathias-kettner.de>
Date: Tue Dec 3 10:13:19 2013 +0100
Updated bug entries #1086
---
.bugs/1086 | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/.bugs/1086 b/.bugs/1086
new file mode 100644
index 0000000..b248248
--- /dev/null
+++ b/.bugs/1086
@@ -0,0 +1,13 @@
+Title: Exception in views when EC views are missing
+Component: multisite
+State: open
+Date: 2013-12-03 10:12:05
+Targetversion: future
+Class: bug
+
+When you enabled the Event Console, modify an EC view and afterwards
+disabled the EC again, then the views the modified views base on
+are missing - and also the columns and filters. This leads to exceptions
+in other views.
+
+Solution: validate views, catch exceptions.
Module: check_mk
Branch: master
Commit: 9ece55eda82a354da7b5e39bf8d99de8a7b830dc
URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=9ece55eda82a35…
Author: Mathias Kettner <mk(a)mathias-kettner.de>
Date: Mon Dec 2 17:17:27 2013 +0100
FIX Fixed summary computation in availability when grouping is used
In the availability view the summary line was computed wrong in cased
you had more than one group (e.g. by grouping after the service
group). That has been fixed.
---
.werks/15 | 11 +++++++++++
ChangeLog | 1 +
web/plugins/views/availability.py | 4 ++--
3 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/.werks/15 b/.werks/15
new file mode 100644
index 0000000..9e70185
--- /dev/null
+++ b/.werks/15
@@ -0,0 +1,11 @@
+Title: Fixed summary computation in availability when grouping is used
+Level: 1
+Component: reporting
+Version: 1.2.3i7
+Date: 1386000955
+Class: fix
+
+In the availability view the summary line was computed wrong in cased
+you had more than one group (e.g. by grouping after the service
+group). That has been fixed.
+
diff --git a/ChangeLog b/ChangeLog
index 976b175..2c1ecd9 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -107,6 +107,7 @@
* 0056 New concept of service periods in availability reporting...
* 0002 You can now annotate events in the availability reporting...
* 0014 FIX: Fix styling of tables: always use complete width...
+ * 0015 FIX: Fixed summary computation in availability when grouping is used...
Event Console:
* 0026 FIX: snmptd_mkevent.py: fixed crash on startup
diff --git a/web/plugins/views/availability.py b/web/plugins/views/availability.py
index 540deea..3e0b68c 100644
--- a/web/plugins/views/availability.py
+++ b/web/plugins/views/availability.py
@@ -1240,7 +1240,7 @@ def render_availability_group(group_title, range_title, group_id, availability,
continue
number = summary.get(sid, 0)
if show_summary == "average" or avoptions["timeformat"].startswith("percentage"):
- number /= len(availability)
+ number /= len(group_availability)
if avoptions["timeformat"].startswith("percentage"):
number *= considered_duration
if not number:
@@ -1256,7 +1256,7 @@ def render_availability_group(group_title, range_title, group_id, availability,
if aggr == "cnt":
count = summary_counts.get(sid, 0)
if show_summary == "average":
- count = float(count) / len(availability)
+ count = float(count) / len(group_availability)
text = "%.2f" % count
else:
text = str(count)
Module: check_mk
Branch: master
Commit: bc25ccb41e8e906f200925ceb65c05c3a7930e99
URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=bc25ccb41e8e90…
Author: Lars Michelsen <lm(a)mathias-kettner.de>
Date: Mon Dec 2 16:32:18 2013 +0100
FIX mk-job is now also usable on CentOS 5+
The mk-job script failed with an error on CentOS 5+ systems,
because the /usr/bin/time command does not support the "--quiet"
option. This option seems to be useless when writing to a file,
therefor we removed this option now.
---
.werks/297 | 11 +++++++++++
ChangeLog | 1 +
agents/mk-job | 2 +-
3 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/.werks/297 b/.werks/297
new file mode 100644
index 0000000..5e67b30
--- /dev/null
+++ b/.werks/297
@@ -0,0 +1,11 @@
+Title: mk-job is now also usable on CentOS 5+
+Level: 1
+Component: checks
+Version: 1.2.3i7
+Date: 1385998226
+Class: fix
+
+The mk-job script failed with an error on CentOS 5+ systems,
+because the /usr/bin/time command does not support the "--quiet"
+option. This option seems to be useless when writing to a file,
+therefor we removed this option now.
diff --git a/ChangeLog b/ChangeLog
index 1e84fab..55d4232 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -46,6 +46,7 @@
* 0290 FIX: winperf_processor mem.win: Handling no/empty agent responses correctly now...
* 0293 FIX: esx_vsphere_counters_ramdisk_sizes: Handles ram disk "ibmscratch" by default now
* 0012 FIX: Solaris Agent: fixed broken fileinfo section...
+ * 0297 FIX: mk-job is now also usable on CentOS 5+...
Multisite:
* 0001 New filters for selecting several host/service-groups at once...
diff --git a/agents/mk-job b/agents/mk-job
index 22c1914..e1df542 100755
--- a/agents/mk-job
+++ b/agents/mk-job
@@ -54,7 +54,7 @@ if ! type $1 >/dev/null 2>&1; then
fi
date +"start_time %s" > "$OUTPUT_PATH/.$IDENT.running"
-/usr/bin/time -o "$OUTPUT_PATH/.$IDENT.running" --quiet --append \
+/usr/bin/time -o "$OUTPUT_PATH/.$IDENT.running" --append \
-f "exit_code %x\nreal_time %E\nuser_time %U\nsystem_time %S\nreads %I\nwrites %O\nmax_res_kbytes %M\navg_mem_kbytes %K\ninvol_context_switches %c\nvol_context_switches %w" $@
RC=$?
mv "$OUTPUT_PATH/.$IDENT.running" "$OUTPUT_PATH/$IDENT"