Module: check_mk
Branch: master
Commit: 6deae9e85c6687bbb9d4eb9656384b3d22ea119f
URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=6deae9e85c6687…
Author: Marcel Arentz <ma(a)mathias-kettner.de>
Date: Tue Feb 14 09:54:42 2017 +0100
4194 FIX esx_vsphere_hostsystem.multipath: Fixed grouping of paths
The changes in vSphere6 made it impossible to display the data in the known way. Now it's necessary to display the information as L#. Additionally there is a suffix to identify the type of the LUN correctly. Otherwise the path to a pseudo-logical device like USB sticks could be accidentally added to the paths of a SAN-LUN.
A rediscovery will be necessary since this fix will change the Service Description!
Change-Id: If5e2ae3e1c8e54948f10fb1e16c1e5b776fc61b5
---
.werks/4194 | 11 +++++++++++
checks/esx_vsphere_hostsystem | 17 ++++++++++++++++-
2 files changed, 27 insertions(+), 1 deletion(-)
diff --git a/.werks/4194 b/.werks/4194
new file mode 100644
index 0000000..6ac9baf
--- /dev/null
+++ b/.werks/4194
@@ -0,0 +1,11 @@
+Title: esx_vsphere_hostsystem.multipath: Fixed grouping of paths
+Level: 2
+Component: checks
+Compatible: incomp
+Version: 1.4.0i4
+Date: 1487062070
+Class: fix
+
+The changes in vSphere6 made it impossible to display the data in the known way. Now it's necessary to display the information as L#. Additionally there is a suffix to identify the type of the LUN correctly. Otherwise the path to a pseudo-logical device like USB sticks could be accidentally added to the paths of a SAN-LUN.
+
+A rediscovery will be necessary since this fix will change the Service Description!
diff --git a/checks/esx_vsphere_hostsystem b/checks/esx_vsphere_hostsystem
index 881a4e1..eba4699 100644
--- a/checks/esx_vsphere_hostsystem
+++ b/checks/esx_vsphere_hostsystem
@@ -399,7 +399,22 @@ def esx_vsphere_multipath_convert(info):
for path, state in data:
# ESX 6.0 uses a different format, e.g "vmhba32:C0:T0:L0 active"
if len(path.split(":")) == 4:
- real_path = ":".join(path.split(":")[:3])
+ hba_num = int(path.split(":")[0].replace("vmhba", ""))
+ real_path = path.split(":")[-1]
+
+ # vmware specificies three different types of LUNs. Physical LUNs will
+ # reserves vmhba number 0-31. The other two reserve numbers of base64.
+ # The lower 32 range is reserved for pseudo logical LUNs. The upper 32
+ # for logical LUNs. e.g. 32-63 for pseudo logical LUNs like USB and
+ # 64-95 for logical LUNs like iSCSI.
+ # These should be added as different multipath LUNs.
+ if hba_num >= 32 and hba_num % 64 < 33:
+ real_path = "%s pseudo-logical" % real_path
+ elif hba_num >= 32 and hba_num % 64 > 32:
+ real_path = "%s logical" % real_path
+ else:
+ real_path = "%s physical" % real_path
+
paths.setdefault(real_path, [])
# Note: There is no path type information available in these esx paths
paths[real_path].append((state, None))
Module: check_mk
Branch: master
Commit: 0ebcf7bf769694a2a7e84c3687bf43d630b9d72d
URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=0ebcf7bf769694…
Author: Sven Rueß <sr(a)mathias-kettner.de>
Date: Wed Mar 22 12:12:30 2017 +0100
New bug for agent updater
Change-Id: I139cafbf199edc99a9b188323094ef88dc0dca29
---
.bugs/2900 | 28 ++++++++++++++++++++++++++++
1 file changed, 28 insertions(+)
diff --git a/.bugs/2900 b/.bugs/2900
new file mode 100644
index 0000000..b436de4
--- /dev/null
+++ b/.bugs/2900
@@ -0,0 +1,28 @@
+Title: cmk-update-agent: Better configuration of updater plugin to prevent problems during updates
+Component: checks
+State: open
+Date: 2017-03-22 11:58:23
+Targetversion: 1.2.8
+Class: bug
+
+The current standard setings of our cmk-update-agent:
+- Timeout for plugin: 60s
+- Execution is synchron to agent
+
+This settings do not take care about several situations on customer side.
+Updates in the local network will not be a problem, because most systems do not need
+more time than 60 seconds for it.
+
+If the updater is used on foreign hosts, like other locations or countries, then
+the 60 seconds are a problem. The download will take much more time to end successfully.
+That will result in the situation, that download perhaps will be finished, but the updater
+does not have enough time to finish. The agent will not work and no further information
+can be queried.
+
+A workaround for it is following (manual setting of specific rules):
+- Set execution mode for plugins and local checks: as value use cmk-update-agent.exe
+- Set timeouts for plugins and local checks: as value use cmk-upadte-agent.exe and 300 seconds (5 Minutes)
+
+It will not solve the problem of long running agent udpates. It will only take care about,
+that the update will successful ended in more situations. For a good solution, the updater
+should handle those thing itself.
Module: check_mk
Branch: master
Commit: ad3acb4b39cf95b233b1d36e82a224557fe79729
URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=ad3acb4b39cf95…
Author: Lars Michelsen <lm(a)mathias-kettner.de>
Date: Wed Mar 22 11:50:51 2017 +0100
Fixed typo
Change-Id: I01d6c4fbaa0d116183c845b546e9bc6fa051598b
---
web/plugins/wato/check_parameters.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/web/plugins/wato/check_parameters.py b/web/plugins/wato/check_parameters.py
index 9cedef1..a3192fa 100644
--- a/web/plugins/wato/check_parameters.py
+++ b/web/plugins/wato/check_parameters.py
@@ -12676,7 +12676,7 @@ register_check_parameters(subgroup_applications,
)),
("spool", Dictionary(
title = _("Spool messages that could not be sent"),
- help = _("Messages that can now be forwarded, e.g. when the target Event Console is "
+ help = _("Messages that can not be forwarded, e.g. when the target Event Console is "
"not running, can temporarily be stored locally. Forwarding is tried again "
"on next execution. When messages are spooled, the check will go into WARNING "
"state. In case messages are dropped by the rules below, the check will shortly "