Last chance!

Well, if nothing popups this week 0.4.0 will be release this weekend! So now's the time to test or it will ship with bugs :)

Michael Medin

Comments

1. Remc0 -- 2012-04-17 15:02

Well, I tested today the latest build for you, but I have problems connecting with check_nt. When I run debug the following lines are created in the nsclient.log: 

2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:51: Timeout reading: 1024
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:53: Timeout <<<-
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:51: Timeout reading: 1024
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:53: Timeout <<<-
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:51: Timeout reading: 1024
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:53: Timeout <<<-
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:51: Timeout reading: 1024
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:53: Timeout <<<-
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:51: Timeout reading: 1024
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:53: Timeout <<<-
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:51: Timeout reading: 1024
2012-04-17 08:57:37: d:..\..\..\..\trunk\include\nrpe\server\connection.cpp:53: Timeout <<<-

My service check gets this output immediately: 

Plugin return code: 2 (status: CRITICAL)
Command: /usr/bin/asmonitor /opt/plugins/check_nt -H 192.168.50.161 -p 1248 -v MEMUSE -w 80 -c 90
Seen command (one argument per line):
  /opt/plugins/check_nt
  -H
  192.168.50.161
  -p
  1248
  -v
  MEMUSE
  -w
  80
  -c
  90
Plugin output:
Connection refused
could not fetch information from server
2. mickem -- 2012-04-17 15:08

Humm, check_nt should not connect to nrpe so something is not right. Check the port configuration?

Also default port is 12489 (for check_nt/NSCLientServer)

Feel free to email config and I can have a look...

Michael Medin

3. Remc0 -- 2012-04-17 18:08

Oke, appearently the port is wrong. How/where can I change this in nsclient.ini?

And where can I find the documentation of the 0.4 release? All I can seem to find is from the 0.3x releases.

4. mickem -- 2012-04-17 19:10

If you go to Documentation you can find information about both 0.4.0 and 0.3.x. Much is the same when it comes to using the various plugins and what not and if you use the "old" configuration even that is the same... But if you ask me the best place to start with "new configuration" is to enable the modules you want and then run nscp settings --generate settings --add-defaults

5. Remc0 -- 2012-04-18 18:30

Oke, I fixed everything, accept check_nt. I changed the port to 1248 in the nsclient.ini like this: 

[/settings/NSClient/server]
allowed hosts = 192.168.50.44
bind to =
cache allowed hosts = true
password =
performance data = true
port = 1248
timeout = 30

When I run it, I get this message: 

Plugin return code: 3 (status: UNKNOWN)
Command: /usr/bin/asmonitor /opt/plugins/check_nt -H 192.168.50.161 -p 1248 -v USEDDISKSPACE -l D -w 85% -c 95%
Seen command (one argument per line):
  /opt/plugins/check_nt
  -H
  192.168.50.161
  -p
  1248
  -v
  USEDDISKSPACE
  -l
  D
  -w
  85%
  -c
  95%
Plugin output:
NSClient - ERROR: Could not complete the request check log file for more information.

I turned on logging (even on trace), but I there is no error or information generated at all in the nsclient.log

6. mickem -- 2012-04-18 18:59

Which version do you have? A check_nt issue was fixed in build 161...

7. Remc0 -- 2012-04-18 20:01

I have version: NSClient++ 0,4,0,160 2012-04-11 x64 booting... I will update to 163 and let you know.

8. Remc0 -- 2012-04-18 20:09

Nope.Same error. Version: NSClient++ 0,4,0,163 2012-04-14 x64 booting...

9. ga15859 -- 2012-04-26 23:26

FYI, post upgrading to 0.4.0.163, I'm getting this on one of my checks - still trying to figure out why...

2012-04-26 15:20:36: c:D:\source\nscp\trunk\include\nscapi/nscapi_plugin_wrapper.hpp:245: Exception in: NSHandleCommand: bad lexical cast: source type value could not be interpreted as target

10. mickem -- 2012-04-27 06:47

You mean you get it after 163 but not before? Could you send me the config/checks (though I guess that is related to NSCA?)

11. ga15859 -- 2012-04-30 20:28

The exception is new with 163, yes. I apologize but I ended up throwing out that check since I found a better third party script anyway, which does work with 163 (I know, last thing you wanted to hear!!). That said, I did find one more issue after pushing 163 to a few more servers. For some reason CheckCounter seems to be returning 0 for pretty much all disk-based counters (why, I have no idea). As an example:

./check_nrpe -H SERVER -t 150 -c CheckCounter -a "Counter=
PhysicalDisk?(_Total)
% Idle Time" MinWarn=50

...will always return zero, as will any on PhysicalDisk?(_Total).

12. ga15859 -- 2012-04-30 20:29

Sorry, I see that was stripped a bit - see here: 

./check_nrpe -H COLOADMIN1 -t 150 -c CheckCounter -a "Counter=\\PhysicalDisk(_Total)\\% Idle Time" MinWarn=50
13. ga15859 -- 2012-04-30 20:34

Same with checks against the processor too. FYI these checks are against the 32-bit version of NSClient on both 2003 and 2008 servers. 

./check_nrpe -H COLOADMIN1 -t 150 -c CheckCounter -a "Counter:proc=\\Processor(_total)\\% Processor Time" MinWarn=50 ShowAll