Refactor a bunch of the logging config files

RESOLVED FIXED

Status

()

RESOLVED FIXED
a year ago
11 months ago

People

(Reporter: dylan, Assigned: dylan)

Tracking

Production

Firefox Tracking Flags

(Not tracked)

Details

Attachments

(1 attachment)

I think we want a log4perl log file in the vagrant directory on the developer VM.

What format should it be in? Plain text or the JSON log format we use in production?
(Assignee)

Comment 1

a year ago
Thoughts on logfile format?

I could do both -- it's not any harder.
Flags: needinfo?(imadueme)
Flags: needinfo?(dkl)
(Assignee)

Comment 2

a year ago
The JSON logs contain a lot more data, so I'm going to just do both.
Flags: needinfo?(imadueme)
Flags: needinfo?(dkl)
Plain text would probably be easier to read, but, having logs in the same format as production also makes sense and could help correlate the same problems. Tough decision, I vote for plaintext but that is the lazy side of me speaking.
(Assignee)

Comment 4

11 months ago
- provide vagrant logging config
- provide docker-compose logging config
- provide production (json) logging config that logs json to STDERR when cereal.pl is not available
- various other fixes
Summary: Log::Log4perl in the developer VM must log messages to some file → Refactor a bunch of the logging config files
(Assignee)

Comment 5

11 months ago
Posted file PR
(Assignee)

Comment 6

11 months ago
- also no longer logging timings from the apache handler. We have this information from a lot of other sources already
and it makes logs hard to read.
(Assignee)

Updated

11 months ago
Attachment #8966446 - Flags: review+
(Assignee)

Updated

11 months ago
Status: NEW → RESOLVED
Last Resolved: 11 months ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.