This bug was originally filed in Launchpad as LP: #1876323
Launchpad details
affected_projects = []
assignee = oddbloke
assignee_name = Dan Watkins
date_closed = 2020-08-25T19:31:21.919428+00:00
date_created = 2020-05-01T14:59:47.385669+00:00
date_fix_committed = 2020-05-06T17:18:11.451975+00:00
date_fix_released = 2020-08-25T19:31:21.919428+00:00
id = 1876323
importance = high
is_complete = True
lp_url = https://bugs.launchpad.net/cloud-init/+bug/1876323
milestone = None
owner = oddbloke
owner_name = Dan Watkins
private = False
status = fix_released
submitter = oddbloke
submitter_name = Dan Watkins
tags = []
duplicates = []
Launchpad user Dan Watkins(oddbloke) wrote on 2020-05-01T14:59:47.385669+00:00
When running cloud-init analyze show against a log file from an Amazon Linux 2 instance (e.g. https://pastebin.com/uhJNysgm), no useful output is produced:
$ cloud-init analyze show -i ~/Downloads/uhJNysgm.txt
-- Boot Record 01 --
The total time elapsed since completing an event is printed after the "@" character.
The time the event takes is printed after the "+" character.
Total Time: 0.00000 seconds
1 boot records analyzed
This is because the log configuration on Amazon Linux 2 (https://pastebin.com/x6NZ4NFJ) configures this for arg0Formatter:
format=%(asctime)s cloud-init[%(process)d]: %(filename)s[%(levelname)s]: %(message)s
which differs from the upstream definition:
format=%(asctime)s - %(filename)s[%(levelname)s]: %(message)s
This is a problem, because the cloud-init analyze commands depend on the " - " part of the line's presence as a separator (see https://github.com/canonical/cloud-init/blob/master/cloudinit/analyze/dump.py#L78), so when we run analyze against an Amazon Linux 2 log file, we incorrectly fail to identify any of the lines in it as being cloud-init log lines.
(As an aside, this is the only non-whitespace difference between the two configuration files.)
This bug was originally filed in Launchpad as LP: #1876323
Launchpad details
Launchpad user Dan Watkins(oddbloke) wrote on 2020-05-01T14:59:47.385669+00:00
When running
cloud-init analyze showagainst a log file from an Amazon Linux 2 instance (e.g. https://pastebin.com/uhJNysgm), no useful output is produced:$ cloud-init analyze show -i ~/Downloads/uhJNysgm.txt
-- Boot Record 01 --
The total time elapsed since completing an event is printed after the "@" character.
The time the event takes is printed after the "+" character.
Total Time: 0.00000 seconds
1 boot records analyzed
This is because the log configuration on Amazon Linux 2 (https://pastebin.com/x6NZ4NFJ) configures this for arg0Formatter:
format=%(asctime)s cloud-init[%(process)d]: %(filename)s[%(levelname)s]: %(message)s
which differs from the upstream definition:
format=%(asctime)s - %(filename)s[%(levelname)s]: %(message)s
This is a problem, because the
cloud-init analyzecommands depend on the " - " part of the line's presence as a separator (see https://github.com/canonical/cloud-init/blob/master/cloudinit/analyze/dump.py#L78), so when we run analyze against an Amazon Linux 2 log file, we incorrectly fail to identify any of the lines in it as being cloud-init log lines.(As an aside, this is the only non-whitespace difference between the two configuration files.)