[Review please!] Logging refactor / cleanup #415

cdrage · 2015-11-23T17:21:20Z

This is a work-in-progress.

It currently looks like this:

DONE!

Review away :)

landscape-bot · 2015-11-23T17:25:49Z

Repository health increased by 2% when pulling d6c9d1e on cdrage:new-logging into 5504ddf on projectatomic:master.

9 new problems were found (including 0 errors and 9 code smells).
13 problems were fixed (including 0 errors and 13 code smells).

landscape-bot · 2015-11-23T17:35:53Z

Repository health increased by 2% when pulling 19e2aaf on cdrage:new-logging into 5504ddf on projectatomic:master.

9 new problems were found (including 0 errors and 9 code smells).
13 problems were fixed (including 0 errors and 13 code smells).

dustymabe · 2015-11-23T20:34:23Z

Should the [OK] messages in your picture of example output be [INFO] instead?

landscape-bot · 2015-11-24T00:16:16Z

Repository health increased by 2% when pulling f0f21e2 on cdrage:new-logging into 5504ddf on projectatomic:master.

10 new problems were found (including 0 errors and 9 code smells).
13 problems were fixed (including 0 errors and 13 code smells).

cdrage · 2015-11-24T01:13:19Z

@dustymabe Yeah, it should :), I'll fix that.

landscape-bot · 2015-11-24T02:56:16Z

Repository health increased by 2% when pulling 3434eaa on cdrage:new-logging into 5504ddf on projectatomic:master.

5 new problems were found (including 0 errors and 5 code smells).
13 problems were fixed (including 0 errors and 13 code smells).

sub-mod · 2015-12-11T18:22:59Z

@cdrage
All the display messages cannot be shown on UI. I would like a method where I can add cockpit specific message like display("step 1 completed", cockpit=true) and these messages should show up when i put cockpit flag

also, can journald output be another PR so that we can have some code for me to test with cockpit

cdrage · 2015-12-11T18:43:54Z

Yup. JournalD will be another PR. At the moment its not possible.

I'll update the PR and add a function for cockpit only info messages so you
can at least test it with cockpit.
On Dec 11, 2015 1:23 PM, "sub-mod" [email protected] wrote:

@cdrage https://github.com/cdrage
All the display messages cannot be shown on UI. I would like a method
where I can add cockpit specific message like display("step 1 completed",
cockpit=true) and these messages should show up when i put cockpit flag

also, can journald output be another PR so that we can have some code for
me to test with cockpit

—
Reply to this email directly or view it on GitHub
#415 (comment)
.

dustymabe · 2016-02-17T19:33:27Z

atomicapp/display.py

+            return input
+
+
+def set_logging(args):


Let's modify this to take vars as input. Looks like the few that we use are: verbose, quiet, and logging_output.

That way we aren't coupled to changes in cli/main.py.

cdrage · 2016-02-17T23:28:56Z

Okay so I've gone ahead and pushed all my other commits.

I have one failing error right now that could be due to a bug in the test suite. Otherwise, good for another review!

dustymabe · 2016-02-18T04:24:08Z

atomicapp/display.py

+            return
+
+        if code == LOG_LEVELS['debug']:
+            msg = "[DEBUG] %6d %0.2f %s" % (os.getpid(), time.time(), msg)


After running it a couple of times I feel like the extra output in the debug statements (the pid/time) doesn't really help debug anything any more. We don't run multiprocess so I don't think this is useful. Thoughts?

@dustymabe updated. yeah, too much output!

dustymabe · 2016-02-18T04:27:55Z

So one really nice thing (for me) about the old output is that it included the name of the file in the output:

[root@f23 redis-centos7-atomicapp (upstreammaster *%=)]# aa run ./ --provider kubernetes                                                                                                                                                                                       
2016-02-17 20:37:21,499 - __main__ - INFO - Action/Mode Selected is: run
2016-02-17 20:37:21,532 - atomicapp.plugin - WARNING - Configuration option 'providerconfig' not found
2016-02-17 20:37:21,532 - atomicapp.plugin - WARNING - Configuration option 'providerconfig' not found
2016-02-17 20:37:21,532 - kubernetes - INFO - Using namespace default
2016-02-17 20:37:21,532 - kubernetes - INFO - trying kubectl at /host/usr/bin/kubectl
2016-02-17 20:37:21,533 - kubernetes - INFO - found kubectl at /host/usr/bin/kubectl
2016-02-17 20:37:21,535 - kubernetes - INFO - Deploying to Kubernetes
2016-02-17 20:37:21,808 - atomicapp.plugin - WARNING - Configuration option 'providerconfig' not found
2016-02-17 20:37:21,808 - kubernetes - INFO - Using namespace default
2016-02-17 20:37:21,809 - kubernetes - INFO - trying kubectl at /host/usr/bin/kubectl
2016-02-17 20:37:21,809 - kubernetes - INFO - found kubectl at /host/usr/bin/kubectl
2016-02-17 20:37:21,810 - kubernetes - INFO - Deploying to Kubernetes

Is there a way to preserve that with the new system?

cdrage · 2016-02-18T14:57:25Z

@rtnpro @dustymabe

Hey all!

So I've updated the PR for final review.

One thing to note is that I will be removing the syslog logging output provider until a later date. I want to do more thorough testing with that.

I've also had to move a few things around in main.py in order to get the logging to output correctly. I checked to see if it would do any harm and from the looks if it, it wouldn't.

Otherwise, feel free to review and test!

dustymabe · 2016-02-18T15:18:25Z

atomicapp/display.py

+            custom_level = LOG_LEVELS[logging_output]
+            break
+        else:
+            custom_level = LOG_LEVELS['default']


LOG_LEVELS is a list right? could we do something like this:

if logging_output in LOG_LEVELS: custom_level = LOG_LEVELS[logging_output] else: custom_level = LOG_LEVELS['default']

@dustymabe Ahhhhh, forgot Python had that functinon >.> I'll add it

cdrage · 2016-02-18T16:17:47Z

@dustymabe updated based on your comments for the for loop

dustymabe · 2016-02-19T18:54:17Z

Closing in favor of #565

cdrage force-pushed the new-logging branch 2 times, most recently from 1da2472 to d6c9d1e Compare November 23, 2015 17:24

cdrage closed this Nov 23, 2015

cdrage force-pushed the new-logging branch from d6c9d1e to e362bb8 Compare November 23, 2015 17:27

cdrage reopened this Nov 23, 2015

cdrage added the work in progress label Nov 23, 2015

rtnpro added the GA label Nov 25, 2015

rtnpro added this to the CDK 2 GA milestone Nov 25, 2015

rtnpro removed the GA label Nov 25, 2015

cdrage mentioned this pull request Nov 25, 2015

Non-quiet, non-verbose mode should output pretty strings #186

Closed

cdrage force-pushed the new-logging branch 8 times, most recently from 1ca21f2 to ffd768a Compare December 2, 2015 23:38

sub-mod mentioned this pull request Dec 8, 2015

Fix cockpit integration #432

Closed

goern assigned cdrage Dec 11, 2015

cdrage force-pushed the new-logging branch 2 times, most recently from 0cc39f4 to 9b7b496 Compare December 21, 2015 17:21

cdrage force-pushed the new-logging branch 2 times, most recently from 1bcee57 to a9ebb56 Compare February 17, 2016 18:33

dustymabe reviewed Feb 17, 2016
View reviewed changes

cdrage force-pushed the new-logging branch 2 times, most recently from 8172d95 to 3edb100 Compare February 17, 2016 23:27

dustymabe reviewed Feb 18, 2016
View reviewed changes

cdrage force-pushed the new-logging branch from 067a54b to 87661da Compare February 18, 2016 14:45

cdrage changed the title ~~[WIP] Logging refactor / cleanup!~~ [Review please!] Logging refactor / cleanup Feb 18, 2016

cdrage removed the work in progress label Feb 18, 2016

cdrage force-pushed the new-logging branch 2 times, most recently from e36e4d0 to e5b9b43 Compare February 18, 2016 14:56

dustymabe reviewed Feb 18, 2016
View reviewed changes

cdrage force-pushed the new-logging branch from e5b9b43 to fd2e945 Compare February 18, 2016 16:10

cdrage added 4 commits February 18, 2016 11:17

Add display class integration

4c92a5a

Cockpit logging output support

d47d98b

Convert logging library outputs to display class

66a8e28

Fix test on errornous output

282f07b

cdrage force-pushed the new-logging branch from fd2e945 to 282f07b Compare February 18, 2016 16:19

dustymabe closed this Feb 19, 2016

cdrage deleted the new-logging branch February 21, 2016 01:11

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Review please!] Logging refactor / cleanup #415

[Review please!] Logging refactor / cleanup #415

cdrage commented Nov 23, 2015

landscape-bot commented Nov 23, 2015

landscape-bot commented Nov 23, 2015

dustymabe commented Nov 23, 2015

landscape-bot commented Nov 24, 2015

cdrage commented Nov 24, 2015

landscape-bot commented Nov 24, 2015

sub-mod commented Dec 11, 2015

cdrage commented Dec 11, 2015

dustymabe Feb 17, 2016

cdrage commented Feb 17, 2016

dustymabe Feb 18, 2016

cdrage Feb 18, 2016

dustymabe commented Feb 18, 2016

cdrage commented Feb 18, 2016

dustymabe Feb 18, 2016

cdrage Feb 18, 2016

cdrage commented Feb 18, 2016

dustymabe commented Feb 19, 2016

[Review please!] Logging refactor / cleanup #415

[Review please!] Logging refactor / cleanup #415

Conversation

cdrage commented Nov 23, 2015

DONE!

landscape-bot commented Nov 23, 2015

landscape-bot commented Nov 23, 2015

dustymabe commented Nov 23, 2015

landscape-bot commented Nov 24, 2015

cdrage commented Nov 24, 2015

landscape-bot commented Nov 24, 2015

sub-mod commented Dec 11, 2015

cdrage commented Dec 11, 2015

dustymabe Feb 17, 2016

Choose a reason for hiding this comment

cdrage commented Feb 17, 2016

dustymabe Feb 18, 2016

Choose a reason for hiding this comment

cdrage Feb 18, 2016

Choose a reason for hiding this comment

dustymabe commented Feb 18, 2016

cdrage commented Feb 18, 2016

dustymabe Feb 18, 2016

Choose a reason for hiding this comment

cdrage Feb 18, 2016

Choose a reason for hiding this comment

cdrage commented Feb 18, 2016

dustymabe commented Feb 19, 2016