#47910 allow logconv.pl -S/-E switches to work even when timestamps not present in access log
Closed: wontfix None Opened 9 years ago by rmeggins.

Ticket was cloned from Red Hat Bugzilla (product Fedora): Bug 1145948

Created attachment 940690
access log

Description of problem:
Attached is a small example access log which can be used to reproduce this, I'm
seeing the same on F20 & RHEL6.

$ logconv.pl ./access - reports meaningful statistics

$ logconv.pl -S foo ./access - reports all zeroes but does not complain about
the invalid time string - should probably error out if the time string is
invalid as otherwise the results will be unexpected and it's hard to spot such
mistakes

$ logconv.pl -S '[24/Sep/2014:10:10:00 +0300]' ./access - reports all zeroes
and the printed "Start of Logs" does not match the provided time string

Using -E does not seem to have any effect.

Version-Release number of selected component (if applicable):
F20 / RHEL6

{{{
1632 if ($currentTimeInSecs > $endTimeInSecs) {
}}}

Shouldn't this be $currentTimeInSecs <= $endTimeInSecs ?

Replying to [comment:5 rmeggins]:

{{{
1632 if ($currentTimeInSecs > $endTimeInSecs) {
}}}

Shouldn't this be $currentTimeInSecs <= $endTimeInSecs ?

Nope. We want to stop processing the log after we have passed the end time.

so how much slower is processing now that we have to convert every timestamp?

Replying to [comment:8 rmeggins]:

so how much slower is processing now that we have to convert every timestamp?

This was a concern I had, but I would think that when using a start/stop time you would not be processing gigs of logs, but smaller snipets.

So I tested 3 million lines of logs:

With start/end times: 59 seconds
WIthout start/end times: 47 seconds

Looks like there is about a 25% performance hit.

If someone is using start/end times to troubleshoot a particualr issue, then I think this is fine. But I'm not sure who asked for this feature in the first place, and in what capacity it was going to be used.

bb55a72..b0213c1 master -> master
commit b0213c1
Author: Mark Reynolds mreynolds@redhat.com
Date: Wed Jun 10 10:49:59 2015 -0400

QE found issue with date parsing with invalid values...

2c484cc..8495afa master -> master
commit 8495afa
Author: Mark Reynolds mreynolds@redhat.com
Date: Tue Jul 14 10:09:00 2015 -0400

69fd1f1..3bf1daa 389-ds-base-1.3.4 -> 389-ds-base-1.3.4
commit 3bf1daa

QE found an issue when the start time is greater than the end time. Reopening...

eb3086d..34ffa6c master -> master
commit 34ffa6c
Author: Mark Reynolds mreynolds@redhat.com
Date: Mon Jul 20 11:18:12 2015 -0400

23a3ff6..16b95b1 389-ds-base-1.3.4 -> 389-ds-base-1.3.4
commit 16b95b1

Test case added for - allow logconv.pl -S/-E switches to work even when timestamps not present in access log
0001-Ticket-47910-allow-logconv.pl-S-E-switches-to-work-e.2.patch

391acfc..71987ed master -> master
commit 71987ed
Author: amsharma amsharma@redhat.com
Date: Fri Aug 7 17:55:20 2015 +0530

096b386..73d7dcf 389-ds-base-1.3.4 -> 389-ds-base-1.3.4
commit 73d7dcf
Author: amsharma amsharma@redhat.com
Date: Fri Aug 7 17:55:20 2015 +0530

Metadata Update from @mreynolds:
- Issue assigned to mreynolds
- Issue set to the milestone: 1.3.4.0

7 years ago

389-ds-base is moving from Pagure to Github. This means that new issues and pull requests
will be accepted only in 389-ds-base's github repository.

This issue has been cloned to Github and is available here:
- https://github.com/389ds/389-ds-base/issues/1241

If you want to receive further updates on the issue, please navigate to the github issue
and click on subscribe button.

Thank you for understanding. We apologize for all inconvenience.

Metadata Update from @spichugi:
- Issue close_status updated to: wontfix (was: Fixed)

3 years ago

Login to comment on this ticket.

Metadata