VR: fix logging is not working and logs are not appended to /var/log/cloud.log

[ustcweizhou]

Description

There are no logs in /var/log/cloud.log when something are changed in VRs (for example start vms, add/remove rules).

Types of changes

• Breaking change (fix or feature that would cause existing functionality to change)
• New feature (non-breaking change which adds functionality)
• Bug fix (non-breaking change which fixes an issue)
• Enhancement (improves an existing feature and functionality)
• Cleanup (Code refactoring and cleanup, that may add test cases)

Screenshots (if appropriate):

How Has This Been Tested?

[DaanHoogland]

code looks good @weizhouapache . Are we to understand that any modules imported don't get the changed logging config if it is set after importing?

[weizhouapache]

code looks good @weizhouapache . Are we to understand that any modules imported don't get the changed logging config if it is set after importing?

@DaanHoogland yes. actually all python scripts in /opt/cloud/bin/cs does not have logging settings, so there is no log appended to /var/log/cloud.log
update_config.py is the entry to update VR configurations (including iptables rules/nics/configs,etc).
we have similar code in elder versions which work fine. After systemvm refactoring by @rhtyd , it does not work any more. I have no idea what caused it.

[yadvr]

@blueorangutan package

[blueorangutan]

@rhtyd a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress.

[DaanHoogland]

pylint doesn't approve of this, @weizhouapache @rhtyd (see 1st travis job) maybe we can mashup the order some more

[DaanHoogland]

not really helpfull, maybe. we can solve the pylint issue by tickering the order, something like:

Suggested change

	import logging
	logging.basicConfig(filename='/var/log/cloud.log', level=logging.INFO, format='%(asctime)s %(filename)s %(funcName)s:%(lineno)d %(message)s')
	import sys
	import logging
	import subprocess
	import os
	import os.path
	import configure
	import json
	logging.basicConfig(filename='/var/log/cloud.log', level=logging.INFO, format='%(asctime)s %(filename)s %(funcName)s:%(lineno)d %(message)s')
	from subprocess import PIPE, STDOUT

Maybe with more from x import y's. Or have some option on pylint.

[blueorangutan]

Packaging result: ✔centos7 ✖centos8 ✔debian. JID-2397

[weizhouapache]

pylint doesn't approve of this, @weizhouapache @rhtyd (see 1st travis job) maybe we can mashup the order some more

@DaanHoogland @rhtyd found another fix, tested ok. please review and test it

[yadvr]

@blueorangutan package

[yadvr]

Lgtm, but needs checking if we need to specify explicit log file path

[blueorangutan]

@rhtyd a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress.

[blueorangutan]

Packaging result: ✔centos7 ✖centos8 ✔debian. JID-2403

[yadvr]

@blueorangutan test

[blueorangutan]

@rhtyd a Trillian-Jenkins test job (centos7 mgmt + kvm-centos7) has been kicked to run smoke tests

[DaanHoogland]

great @weizhouapache more elegant and the travis errors so far are unrelated. it's easy to overlook the package init possibility, thanks for this.

[blueorangutan]

Trillian test result (tid-3206)
Environment: kvm-centos7 (x2), Advanced Networking with Mgmt server 7
Total time taken: 31599 seconds
Marvin logs: https://github.com/blueorangutan/acs-prs/releases/download/trillian/pr4466-t3206-kvm-centos7.zip
Smoke tests completed. 83 look OK, 0 have error(s)
Only failed tests results shown below:

Test	Result	Time (s)	Test File

[DaanHoogland]

marking this as major as it stops operators from being able to do proper diagnosis on the system/network

[davidjumani]

LGTM
Able to see logs when rules added