Page 1 of 1

Pine A64+ issue with monitoring engine not running

Posted: Thu Oct 01, 2020 1:51 pm
by czer0
I apologize for cross-posting this issue as Help and Support may have not been the correct place as this seems more of a bug than anything else.

Platform: A64+
NEMS Linux version: 1.5.2
NEMS Build number: Build 1

This occurs on a clean install.
Have verified that the checksum is correct for the image to avoid possible corruption being cause.
Have swapped sdcards to rule out storage being the issue.

I am able to build and deploy a config from nconf but logging into Adagios I am presented with the following error:

Oh no, something went wrong ☹

Is our monitoring engine even running ? If so we cant detect it.

Don't worry you can still configure stuff but you might want to start it so you can start monitoring stuff.

Please verify your settings and check all paths are valid.
check that nagios is in fact running and sudoers.d is correctly configured.
Try starting it by clicking here and see if you are getting some sort of errors

ParserError: "[Errno 2] No such file or directory while connecting to '/usr/local/nagios/var/rw/live.sock'. Make sure nagios is running and mk_livestatus loaded."


Show debug Information produces additional insights:
Traceback (most recent call last):
File "/usr/local/lib/python2.7/dist-packages/adagios/views.py", line 43, in wrapper
result = view_func(request, *args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/adagios/status/views.py", line 650, in status_index
c = adagios.status.utils.get_statistics(request)
File "/usr/local/lib/python2.7/dist-packages/adagios/status/utils.py", line 460, in get_statistics
*arguments
File "/usr/local/lib/python2.7/dist-packages/pynag/Parsers/livestatus.py", line 1116, in get_services
return self.query('GET services', *args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/pynag/Parsers/multisite.py", line 80, in query
query_result = backend_instance.query(query, *args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/pynag/Parsers/livestatus.py", line 973, in query
livestatus_response = self.write(livestatus_query.get_query())
File "/usr/local/lib/python2.7/dist-packages/pynag/Parsers/livestatus.py", line 820, in write
livestatus_socket = self._get_socket()
File "/usr/local/lib/python2.7/dist-packages/pynag/Parsers/livestatus.py", line 804, in _get_socket
raise ParserError(msg % (e, self.livestatus_socket_path))
ParserError: "[Errno 2] No such file or directory while connecting to '/usr/local/nagios/var/rw/live.sock'. Make sure nagios is running and mk_livestatus loaded."

Re: Pine A64+ issue with monitoring engine not running

Posted: Thu Dec 10, 2020 2:08 pm
by Robbie Ferguson
Hi czer0. Sorry you encountered this. It was finally reported to me a couple weeks back and was patched. But just seeing your post here on the community forum now... quite late.

So... it's fixed. Something like this though should have been reported directly to me 'cause I feel bad I didn't see it ... and it's not something the community would be likely to be able to help with (since it required patching the distro).

All the best!
Robbie

Re: Pine A64+ issue with monitoring engine not running

Posted: Thu Dec 10, 2020 3:00 pm
by czer0
Hey Robbie,

I actually spoke to you awhile back on the discord about this and had forgotten that I still had this posted.

Re: Pine A64+ issue with monitoring engine not running

Posted: Thu Dec 10, 2020 4:25 pm
by Robbie Ferguson
You're all good then?

Cheers,
Robbie

Re: Pine A64+ issue with monitoring engine not running

Posted: Thu Dec 10, 2020 7:04 pm
by czer0
looks like the issue has been resolved according to your post on the #pine64 channel on discord. Just need the details on the patch, but i can message on discord about that.

Re: Pine A64+ issue with monitoring engine not running

Posted: Thu Dec 10, 2020 7:31 pm
by Robbie Ferguson
Great. Here's the info for anyone finding this later:

rookieone pointed out an issue with the A64+ (and possibly other #pine64 boards) where it was mis-detecting the kernel headers and therefore unable to compile the updated socket, resulting in a failure to start Nagios Core.

This has been fixed.

To run the patch manually you can run this on your NEMS Server:

Code: Select all

sudo nems-update && sudo /root/nems/nems-admin/build/055-check_mk
Please note, this is ONLY VALID for NEMS 1.5.x for PINE64 boards (A64+ tested, others possibly affected). DO NOT run this on legacy NEMS Linux systems nor 1.6+ without first consulting me.

WARNING: This will take a LONG TIME. Set aside 2 hours for the update.

This has already been fixed in 1.6.

Robbie // The Bald Nerd