[Home]

Summary:ASTERISK-02574: [patch] set verbose and set debug still conflict with each other
Reporter:Christopher L. Wade (clwade)Labels:
Date Opened:2004-10-09 10:11:03Date Closed:2008-01-15 16:03:34.000-0600
Priority:TrivialRegression?No
Status:Closed/CompleteComponents:Core/General
Versions:Frequency of
Occurrence
Related
Issues:
Environment:Attachments:( 0) set_debug_above_zero.txt
( 1) set_debug_zero.txt
( 2) debug_verbose.patch
( 3) sav_2617_r32798.diff
Description:When set debug is executed with any value other than 0, it continues to suppress verbose output on the CLI.  This bug is somewhat related to bug ASTERISK-2559.

****** ADDITIONAL INFORMATION ******

This is with 'set verbose 25' having already been run.

=======================================================
asterisk*CLI> set debug 0
Core debug is now OFF
asterisk*CLI> extensions reload
 == Parsing '/etc/asterisk/extensions.conf': Found
[snip 11 more 'Parsing' lines...]
   -- Setting global variable 'PROTO' to 'SIP'
[snip 11 more 'Setting' lines...]
   -- Registered extension context 'default'
   -- Added extension '_80[1-5]' priority 1 to default
[snip 8 more 'Added' lines...]
[snip 154 more mixed 'Registered' and 'Added' lines...]
   -- Including context 'default' in context 'internal'
[snip 2 more 'Including' lines...]
asterisk*CLI> set debug 25
Core debug was 0 and is now 25
asterisk*CLI> extensions reload
   -- Setting global variable 'PROTO' to 'SIP'
[snip 11 more 'Setting' lines...]
   -- Including context 'default' in context 'menu-main'
[snip 3 more 'Including' lines...]
asterisk*CLI>
=======================================================
Comments:By: Christopher L. Wade (clwade) 2004-10-09 10:20:28

The two attached files show the actual output difference between running 'extensions reload' with 'set debug 0' and 'set debug 25'.  As you can see, debug being anything higher than zero greatly reduces the verbosity of the output.

This may not be the only place still affected by 'set debug' bug.  I'm looking for others, will keep posted.
By: Mark Spencer (markster) 2004-10-09 10:41:56

There are several places where the presense of DEBUG has historically reduced verbose output.  Since they both used to always go to the same place, it made sense.  If you want to go through and find them all that's fine, I'll be happy to apply the patch.
By: Christopher L. Wade (clwade) 2004-10-09 10:46:05

No problem, I'll see what I can do.  It might be Monday before I can do any real work on it though, I've got an upcoming deadline for another project.
By: Mark Spencer (markster) 2004-10-16 15:51:18

ANything further you want to do here?  If not I'll probably just close this out since it's not *really* a bug.
By: Christopher L. Wade (clwade) 2004-10-18 08:08:44

Go ahead and close it for now.  As for it being a bug or not, I guess it could be considered a feature request instead.  Thanks Mark.
By: Christopher L. Wade (clwade) 2005-11-29 16:53:22.000-0600

Finally, over a year later, get around to actually fixing this, sorry it took so long.  Regardless patch against head available, should be ported back to 1.2 and even 1.0 as this was a bug, in my humble opinion.
By: Christopher L. Wade (clwade) 2005-11-29 17:38:57.000-0600

extension_reload_debug_verbose.patch just fixes this for 'extensions reload'
debug_verbose.patch attempts to fix this for everything

I'm sure I missed stuff in debug_verbose.patch, but I did my best to check all references to option_debug in the source.

... and summary should be updated to include [patch]


By: Olle Johansson (oej) 2005-12-03 12:54:20.000-0600

Will take a look at this patch.
/O
By: Olle Johansson (oej) 2005-12-03 12:59:27.000-0600

I guess that I only need debug_verbose.patch since it includes everything, right?

Implementing this patch in
https://svn.digium.com/svn/asterisk/team/oej/bug2617
By: Olle Johansson (oej) 2005-12-03 13:48:16.000-0600

I removed double logging. I think there's no sense of logging the same information both to debug and verbose channels. I've made a decision on where to log in most cases.
By: Christopher L. Wade (clwade) 2005-12-03 15:08:50.000-0600

oej: Yes, debug_verbose.patch is the only one needed if you want to fix all of these issues.

As for the double loggin, I saw those, I just wasn't sure what to do in most cases so I figured it was better to allow double logging than remove some 'critical' log.

Regardless, thanks for looking at this oej.


By: Matt O'Gorman (mogorman) 2006-01-10 10:27:50.000-0600

oej can you put this year old bug to rest?
By: Kevin P. Fleming (kpfleming) 2006-02-13 16:52:46.000-0600

I'm trying to review this patch via the bug2617 branch, but it seems to be quite confusing... the merge properties show it being based on branches/1.2, but it contains bits that are not in 1.2... but it's also missing bits that are in the trunk.
By: Serge Vecher (serge-v) 2006-05-26 15:16:13

I've also tried to make the diff and it shows that the branch is seriously messed up. clwade: can you please make one last patch against the latest trunk?

Thank you!
By: Serge Vecher (serge-v) 2006-06-07 13:28:12

alright, debug_verbose.patch had only one hunk failure in pbx.c which was easy to fix. Uploading a patch for the latest trunk, disclaimer sent on 06/07/06
By: Serge Vecher (serge-v) 2006-08-25 14:23:55

*ping*
By: Matt O'Gorman (mogorman) 2006-09-19 11:24:11

committed into trunk and 1.2
By: Digium Subversion (svnbot) 2008-01-15 16:03:34.000-0600

Repository: asterisk
Revision: 7303

A   team/oej/bug2617/

------------------------------------------------------------------------
r7303 | oej | 2008-01-15 16:03:33 -0600 (Tue, 15 Jan 2008) | 2 lines

Temporary branch for bug ASTERISK-2574

------------------------------------------------------------------------

http://svn.digium.com/view/asterisk?view=rev&revision=7303