FOP2 Community Forum

Followup - clicking numbers in Yahoo Mail and Contacts doesn't work for me either.

We are using testing the FOP2 Chrome Extension. We are finding that on most webpages when a phone number is simply displayed it is clickable, but when it is in a text box, or on any Google Mail or Google Contact page, phone numbers are not detected as clickable. On Goggle pages it doesn't matter if outbound voice calling using Google Voice is enabled or disabled. Numbers on a Google Site web page are clickable.

We are using Google Apps for Business so this could be a big deal. Probably the major use would be for dialing numbers in an email or a contact record.

Can anyone confirm these symptoms? Better yet, found a work around?

Will do, thanks.

Our call center manager needs to track time spent on pause by reason. Pause/unpause actions are logged in queue_log, and our reporting system (Queuemetrics) can track and report paused time, but the supervisor needs tracking to be more granular.

FOP2 DND status is noted on the Panel but is not logged anywhere. I know there are probably folks that need to be able to pause without DND, but maybe combining DND with pause logging would be easy and not disruptive. The reason codes are already in place, just need the additional pause directive to be logged and DND does a defacto pause as it is.

BTW, we did look at using QM for the pause reasons, but we are not using the agent page now and think it would be an extra confusion to just use it for pauses.

Thanks

Ok, I think I've got this figured out. When users log in there is a set of debug lines like this:

192.168.20.169 <= <msg data="1|auth|3015|dddbeb9eccdb350328b6ee285c36aa8c" />
-- PROCESS_FLASH_COMMAND origen 1 accion auth destino 3015 password dddbeb9eccdb350328b6ee285c36aa8c
** RECEIVING AUTH REQUEST for 3015
TRYING TO VALIDATE USER 3015
VALIDAR USUARIO 3015
VALIDAR USUARIO 3015 OK con clave regular (192.168.20.169)
(interesting mix of English and - I assume - Spanish)

The system is an Ipiphony with a general login page that links to the FOP2 panel and feeds it the login credentials. The tricky part is that some of my users have bookmarked the URL for the FOP2 panel. The bookmarked page automatically and immediately does a submit with no login entries when it is loaded. When extension is null a debug line like this is generated:

192.168.20.82 <= <msg data="1|auth||c8bca59c92e7dc9123e1f0c8b7fb230a" />
-- PROCESS_FLASH_COMMAND origen 1 accion auth destino password c8bca59c92e7dc9123e1f0c8b7fb230a
** RECEIVING AUTH REQUEST for
ValidateSecret general, no user

Note the missing data after 'auth'.

So mystery solved. I do have a request for the developers though. It took a couple of hours to find a way to link the audit log entry to the debug output. And I had to do it on my live system bouncing FOP2 with debuglevel changes and pissing off my users because I didn't know how to replicate the problem on my test system. It would be a big help if specific audit log entries could be tied to debug output by a timestamp in output.log. Or if the generation of the audit entry was actually in the debug output. In any case I learned a lot!

Thanks! I was about to post just that question. I haven't found this elsewhere in the documentation or forum, just references to a few of the levels by number with no explanation of what is being logged. It might be helpful to have it on the http://www.fop2.com/documentation-insta ... ommandLine page.

BTW, in the --debuglevel description on that page, it mentions the command line option overrides the value in fop2.cfg. I didn't see any comments for debuglevel in my fop2.cfg, and putting in debuglevel = n also didn't seem to work. I'm running version 2.22.

The audit log includes time stamps. I've not seen any in error.log or output.log. Is there a way to enable timestamps in those?

After enabling audit today I've been finding many of the following:

1341965715||GENERAL|FAILED LOGIN

With no identifying data it's hard to troubleshoot. Is there a way to increase the verbosity of the audit log?

Ok, after further research I found what I was looking for. I found two types of logging with examples of how to control them with the command line.

-- audit /path/to/log.file (I used /var/log/fop2/audit.log)

I added this option to /etc/default/fop2 (which is pulled in by /etc/init.d/fop2 on my system). After creating the directory and restarting fop2 audit.log was automatically created and began logging logins and logouts.

I then set the log directory with "-l /var/log/fop2/"

Again, I added this to /etc/default/fop2. Entries started accumulating in error.log and output.log. I found the logging level in these responded to changes to the debug level. The default level on my system appears to be 0, but I added the following option as a placeholder to make it easier for future debugging:

--debuglevel 0

Increasing to examples found in the forums for command line option -X n creates increasingly detailed entries as n increases. On a general unused test system I accumulated over 4MB in output.log over a couple of hours.

Hope this helps someone!

Ok, that didn't get me any further than my forum searching and I can't find the answer in the on-line documentation so I'll just ask the question - where in a config file can I find where FOP2 logs errors?

I'm trying to search for the FOP2 log file and searched for "where is the fop log file". All the words were rejected as too common and the search failed. Any suggestions?

I don't need to change from the default, but I would like to get rid of the errors. Would a touch to create an empty file cause a problem? Or would parsing an empty file just cause more problems?

Doing a general sweep through logs on my server I found we have been getting the following error in /var/log/apache2/error.log for every workstation every day going back to when the system was built:

[Fri Nov 25 08:33:03 2011] [error] [client ipaddress] File does not exist: /var/www/fpbx/fop2/fop2-variablesGeneral.txt, referer: http://pbx/advanced/fop2

We aren't experiencing any problems I could trace to these, but unexplained errors bug me

There is no file on my system anywhere with this name. Is it an optional configuration file, or perhaps a typo in the code somewhere? I also can't find it greping files in fop2 directories.

To be of any help I'll need to understand what is triggering the state changes. Unfortunately I have no time to dig into the code. Can you provide a description of what is supposed to be happening and what you suspect? I always have at least one agent with the wrong visual state in the queue button, usually more, so if you can point me in the right direction I'll see what I can find.

We can use that as a work around, but it would be good to understand why it's not correct.

We are running asterisk 1.6.2.18, Freepbx 2.8.1.1, and FOP2 2.22. All our agents are configured in the queue configuration as dynamic members and add/remove/pause themselves as needed.

The color mismatch on the queue buttons continues to happen for agents on both incoming and outgoing calls.

The extension buttons do work. The point is that the agents (and supervisors) are trying to use the Queue buttons to show who is currently added to each queue and of those who is available, and we can't trust the display.

I typed it into one text box, then copied it out to the clipboard and pasted it into the list in the appropriate alphabetical spot. I needed to hit enter after the paste. Maybe that was it?

[quote]Tried again this morning to add preferences to individual users. Some I'd tried twice with no luck. This time I selected preferences from the Permissions Quick Pick dropdown instead of pasting it in and now the users are seeing the Preferences icon. Problem solved, no time to figure out why!

Most probably a typo ?!?[/quote]

Possibly, but I did check the spelling several times. I'm wondering if cutting and pasting the word into the text box was somehow different that (cr lf wise) that using the drop down.

Whatever. It's working now!

I am talking about fop2. We have occasionally seen the clock in the agent button showing they are on pause status but the lines for the agent in the queue buttons on the right side of the display did not change to the clock.

More commonly, the agent is on a call but the agents shirt color on the icon on the queue buttons is still green. Or they are off the call and the agent button is green, but the queue button shows an orange shirt. I'm looking a queue with six agents right now. Three are on calls and all are wearing green shirts on the queue button.

I'd like to understand why the status color is changing on one part of the panel and not on the other.

The bigger concern is the agent button not resetting to it's 'available' display after a call completes. That is what other agents are looking at when deciding who is available to take a transfer.

Agents are asking for two options on their Action Submenu:

Add to All My Queues
Remove from All My Queues

Most of our agents are dynamically assigned to multiple queues and this would be a handy (and globally appreciated) time-saver.

There is also confusion with the Action submenu displaying "Add to queuename" and "Remove from queuename" lines for every queue being displayed on their panel whether the agent is allowed to join them or not. If it is not possible to suppress those lines it would be helpful to have a different icon for the queues they can actually join.

Thanks!