Unregistered and Diabled Trunks still showing on FOP2

Ramblin

I did NOT run the
/usr/local/fop2/generate_override_contexts.pl -write
script, but it says this is not required; that it is an option if you want tighter FreePBX configuration
RIchard

I did run
/usr/local/fop2/generate_override_contexts.pl -write
which did edit the files as described

I then restarted fop2 - no change
I then rebooted the system - no change

I still have:

trunks not reflecting a loss of registration
no icons in the Toolbar

I really like the FOP2 system design (heck I just bought 2 licenses); I'd like to get it fully functional

nicolas

Hi,

We are many hours a day available at the live help, sometimes that is the best way to fix configuration issues.

If you do not see action buttons is because you are logged with an extension that is NOT listed in the extension box below, your extension must be also present as a button in order to perform actions. It is mandatory. If you log with user "johndoe" or an extension number that does not match any "extension" in your button config, then you won't see the action buttons.

Do you see sip registration status for your extensions? for trunks works the same, as long as the trunk channel matches the peerstatus event channel.

Ramblin

We are many hours a day available at the live help, sometimes that is the best way to fix configuration issues.
If you do not see action buttons is because you are logged with an extension that is NOT listed in the extension

With your help, I think I just discovered an undocumented feature
I had manually established a user (I disabled the auto-config of users) in the fop2.cfg file (say johndoe) as
user=<johndoe>:<password>:all
to be the generic user who could monitor all extensions, trunks, ...

This allowed me to have "johndoe" act as a system administrator to diagnose connection status and issues. It might even be an outside contractor who does not have an extension. It really does not matter.

Now that I know about the need to have the user be an actual extension, I realize that the "johndoe" user could be intentionally set up for someone you want to have read-only status to the panel. No control icons, just read-only. Nice feature actually.

So yes, now that I have established, as a user, one of the legitimate extensions (mine), I can see the toolbar with the icons.
I need now to figure out how to do the lsiten and whisper feature so I can coach a new call centre employee, but I'll dig into that.

Thank you.
One item resolved - and yes, the root cause was user issue .

However, I still am not able to get the trunk button to show in the panel when the SIP trunk goes unregistered.
I signed in as the real extension user and manually disabled the trunk.
When I ran

asterisk -rx 'sip show registry'

it did properly show the trunk as Not Registered / No Authentication

Host                                    dnsmgr Username       Refresh State                Reg.Time                 
...
sip06.unlimitel.ca:5060                 N      5195940055         120 No Authentication                             
...
17 SIP registrations.

but the trunk button in the FOP2 panels stayed green througout.
I left the trunk in a Not Registered state for 5 minutes in case there was a lag issue with FOP2 or something, but no change - the button stayed green.

I hear you that the FOP2 panel uses AMI and I am using CLI commands, but the net is that the trunk loses registration so I would have expected the panel button to realize that.
When an extension is not registered, it greys out (regardeless of what user I sign in with), but not the trunks.

So maybe it is time to try a live interaction.
I am in the Eastern Standard Time zone.
Where are you and what would be a good time to try and use the live interaction?

Thanks for your help on this; it is a great piece of software and I'd like to get the configuration working fully.

nicolas

The grayed out buttons are done for SIP Status (sip show peers). Not registry information. A Peer with qualify set will send periodic events via AMI about its status. So, check "sip show peers" for your trunks, not "sip show registry"

Best regards,

Ramblin

check "sip show peers" for your trunks

I disabled the trunk and then ran

asterisk -rx 'sip show registry'

just to make sure it was disabled and yes, it was.

I then went and ran

asterisk =rx 'sip show peers'

and it showed no problems

But the trunk was unusable

I waited for 10 minutes with the trun in a disablled status but it made no difference:

sip show registry continued to show the SIP trunk as not Registered
sip show peers continued to show no problems

The trunk was not usable for the 10 minutes that is was not registered, but FOP2 showed it as fine (using sip show peers)

Is there a way to get sip show peers to show the non-functioning trunK?
If not, is there a way to get the panel to reflect sip show registry as well as sip show peers?

R

Ramblin

The FOP1 panel used to show when the trunk went unregistered

Is there some coed I could take from that and try to integrate with FOP2?

I know I am harping on one bit of funcitonality here, but it was the ability of FOP1 to show when trunks went unregistered that allowed me to resolve an issue with a trunk service provider. They initially denied any issue at their end but by using FOP1's ability to dispaly trunk status, I was able to track the status and get the provider to fix (although never admit to ) the error on their end.

So I am hoping to retain that functionality in FOP2

FOP2 is a great piece of software and does many things FOP1 did not do. I am hoping to keep the one thing FOP1 did do that at this point it appears FOP2 does not do.

nicolas

FOP2 shows sip peer status, including trunks. If a "sip show peers" shows a trunk as lagged or unreachable, it will show as such in the fop2 display. (if the manager event is properly received).

Best regards,

Ramblin

FOP2 shows sip peer status, including trunks.
Best regards,

Nicolas,

I get the fact that FOP2 shows sip peer status , NOT sip registry status.

What I do not understand is why FOP1 showed the trunk when it went unregistered but FOP2 does not.
Did something change between fop1 and FOP2?

I literally copied/pasted the trunk (and extension) settings from the old (64-bit CentOS 6.2, Asterisk 1.8, FreePBX 2.9, fop1) system to the new (64-bit CentOS 6.4, Asterisk 11, FreePBX 2.11, FOP2) system and on the fop1 system, when the trunk went out of registration, the trunk button "dimmed" but on the FOP2 system it does not change colour.

It may be that the same peer settings for the trunk or the same AMI settings are no longer sending the same status messages out, but I do not know how to check this.
Can you advise?

I have experimented on numerous trunks, but I will share with you the settings for one trunk (private info replaced by {content}) that I got by running

asterisk -rx 'sip show peer Mktg_5-out'

which produced

  * Name       : Mktg_5-out
  Description  :
  Secret       : <Set>
  MD5Secret    : <Not set>
  Remote Secret: <Not set>
  Context      : from-pstn
  Record On feature : automon
  Record Off feature : automon
  Subscr.Cont. : <Not set>
  Language     :
  Tonezone     : <Not set>
  AMA flags    : Unknown
  Transfer mode: open
  CallingPres  : Presentation Allowed, Not Screened
  Callgroup    :
  Pickupgroup  :
  Named Callgr :
  Nam. Pickupgr:
  MOH Suggest  :
  Mailbox      :
  VM Extension : *97
  LastMsgsSent : 0/0
  Call limit   : 0
  Max forwards : 0
  Dynamic      : No
  Callerid     : "" <>
  MaxCallBR    : 150 kbps
  Expire       : -1
  Insecure     : port,invite
  Force rport  : Yes
  Symmetric RTP: Yes
  ACL          : No
  DirectMedACL : No
  T.38 support : No
  T.38 EC mode : Unknown
  T.38 MaxDtgrm: -1
  DirectMedia  : No
  PromiscRedir : No
  User=Phone   : No
  Video Support: Yes
  Text Support : No
  Ign SDP ver  : No
  Trust RPID   : No
  Send RPID    : No
  Subscriptions: Yes
  Overlap dial : Yes
  DTMFmode     : rfc2833
  Timer T1     : 500
  Timer B      : 32000
  ToHost       : {provider_FQDN}
  Addr->IP     : {privider_IP_address}:5060
  Defaddr->IP  : (null)
  Prim.Transp. : UDP
  Allowed.Trsp : UDP
  Def. Username: {phone number}
  SIP Options  : (none)
  Codecs       : (ulaw)
  Codec Order  : (ulaw:20)
  Auto-Framing :  No
  Status       : OK (18 ms)
  Useragent    :
  Reg. Contact :
  Qualify Freq : 60000 ms
  Keepalive    : 0 ms
  Sess-Timers  : Accept
  Sess-Refresh : uas
  Sess-Expires : 1800 secs
  Min-Sess     : 90 secs
  RTP Engine   : asterisk
  Parkinglot   :
  Use Reason   : No
  Encryption   : No

Do you see anything in this setup that would prevent "sip show peers" from getting a failed trunk through to FOP2?

nicolas

I see this in your post:

asterisk -rx 'sip show peer Mktg_5-out'

Name of the peer is "Mktg_5-out", what is the name of the fop2 channel for that trunk button? If you have TWO separate names, one for inbound, one for outbound, you will have to specify the other/missing one in the channel field in the fop2 buttons freepbx admin page. Or add it as an extra channel field in your button definition if configuring manually. Otherwise, any event related to that extra channel name will not match your button.

Example in fop2 buttons.cfg:

[SIP/Mktg_5]
type=trunk
...
...

If you have events received with SIP/Mktg_5-out, they will not match as they are in fact a different channel. So, add an extra channel, like:

[SIP/Mktg_5]
type=trunk
channel=SIP/Mktg_5-out
...
...

nicolas

By the way, just to be sure I tried with my own installation for a sip trunk I have. If I block the ip address via iptables, in a couple of minutes I received the proper unreachable event and the button is grayed out in fop2.

If you want to log the manager output for debug, you can set the options line in /etc/sysconfig/fop2 like this

OPTIONS="-d -X 15 -l /tmp"

That will log output into tmp (output.log and error log in fop2 2.26 or prior, fop2_debug log in fop2 2.27 beta). You will see, among other things, the events being received in AMI, like:

127.0.0.1            <- Event: PeerStatus
127.0.0.1            <- Privilege: system,all
127.0.0.1            <- ChannelType: SIP
127.0.0.1            <- Peer: SIP/trunkprovider
127.0.0.1            <- PeerStatus: Unreachable
127.0.0.1            <- Time: -1

Best regards,

Ramblin

If you have TWO separate names, one for inbound, one for outbound, you will have to specify the other/missing one in the channel field in the fop2 buttons freepbx admin page.
Example in fop2 buttons.cfg:

[SIP/Mktg_5]
type=trunk
...
...

If you have events received with SIP/Mktg_5-out, they will not match as they are in fact a different channel. So, add an extra channel, like:

[SIP/Mktg_5]
type=trunk
channel=SIP/Mktg_5-out
...
...

I think you may be on to something, but I am not sure I fully understand the solution (sorry for being dense)

Here is my FreePBX setup for trunk Mktg 5
http://docs.ayuda.ca/Mkgt_5-Trunk.jpg

When I tried to make the Trunk Name (Outgoing settings) and the Context Name (Incoming Settings) the same, I got an error message saying

Trunk Name and User Context cannot be set to the same value

so Ieft them as is, with the
Trunk Name (Top setting): Mktg 5
Outgoing Settings / Trunk Name: Mktg_5-out
Incoming Settings / User Conttext : Mktg_5-in

I added, using the FreePBX -> Other -> FOP2 Buttons, the additional Mktg_5-in Channel to my SIP?Mktg_5-out trunk and then

rebuilt the buttons.cfg file using /usr/local/fop2/autoconfig-buttons-freepbx.sh > /usr/local/fop2/buttons.cfg
restarted fop2 using service fop2 restart
and I now have, in my buttons.cfg file the following entry for SIP/Mktg_5-out

[SIP/Mktg_5-out]
type=trunk
label=SIP/Mktg_5-out
queuecontext=from-internal
privacy=none
channel=SIP/Mktg_5-in
rtmp=0

I entered a bad secret in the registration settings for the Mktg 5 trunk and within a minute, the registration was reflected as bad, but the Trunk button did not change. I also noticed that I did not have any extra buttons in the panel for the Inbound trunk (should I have extra buttons for the additional channel?).

I noticed in your example, you used SIP/Mktg_5 not SIP/Mktgt_5-out in your trunk definition. I have "Mktg 5" as my (free-text) Trunk name but Mktg_5-out and Mktg_5-in as my outbound and inbound setting names. When I run the autoconfigure with

/usr/local/fop2/autoconfig-buttons-freepbx.sh > /usr/local/fop2/buttons.cfg

it automatically picks up the Mktg_5-out name.
Is that an issue?

I am guessing we are close and I appreciate you sticking with me on this; I hope I can "get it" quickly.

Ramblin

If you want to log the manager output for debug, you can set the options line in /etc/sysconfig/fop2 like this
OPTIONS="-d -X 15 -l /tmp"

I changed (temporarily, that quickly started adding up to quite a log file) the settings in /etc/sysconfig/fop2 to the above and restarted fop2

The output.log populated very quickly but the error.log got no entries.

The entries in the output.log for Mktg_5 were

127.0.0.1       <- Event: PeerEntry
127.0.0.1       <- Channeltype: SIP
127.0.0.1       <- ObjectName: Mktg_5-in
127.0.0.1       <- ChanObjectType: peer
127.0.0.1       <- IPaddress: {IP_address_of_provider}
127.0.0.1       <- IPport: 5060
127.0.0.1       <- Dynamic: no
127.0.0.1       <- AutoForcerport: no
127.0.0.1       <- Forcerport: yes
127.0.0.1       <- AutoComedia: no
127.0.0.1       <- Comedia: yes
127.0.0.1       <- VideoSupport: yes
127.0.0.1       <- TextSupport: no
127.0.0.1       <- ACL: no
127.0.0.1       <- Status: OK (17 ms)
127.0.0.1       <- RealtimeDevice: no
127.0.0.1       <- Description: 
127.0.0.1       <- Server: 0

127.0.0.1       <- Event: PeerEntry
127.0.0.1       <- Channeltype: SIP
127.0.0.1       <- ObjectName: Mktg_5-out
127.0.0.1       <- ChanObjectType: peer
127.0.0.1       <- IPaddress: {IP_address_of_provider}
127.0.0.1       <- IPport: 5060
127.0.0.1       <- Dynamic: no
127.0.0.1       <- AutoForcerport: no
127.0.0.1       <- Forcerport: yes
127.0.0.1       <- AutoComedia: no
127.0.0.1       <- Comedia: yes
127.0.0.1       <- VideoSupport: yes
127.0.0.1       <- TextSupport: no
127.0.0.1       <- ACL: no
127.0.0.1       <- Status: OK (17 ms)
127.0.0.1       <- RealtimeDevice: no
127.0.0.1       <- Description: 
127.0.0.1       <- Server: 0

This was the same as the entries for the other channels (for which I had NOT yet added the extra channel in the fop2 admin) and no errors/outages were reported even though the registry did show No Authentication.

I don't know where to go now.

I was wondering if the timing settings were such that the Peer never realized the trunk was out, but:

I had the same settings in the previous (fop1) system that did show trunk registration failures
the registry settings are immmediately being seen
I had left the (non-registered) state as is for 15 minutes so I thought that would allow any delays to be recognized

Something is different between ours systems though, since you are getting yours to work, so FOP2 is certainly capable of it. It just is not doing it with my configuration. All the other functinality is working (Extensions grey out when not connected, Conference, Park, Trunks show properly when being used, Toollbar Icon commands work, ...)

One thing I did try: On Mktg_8-out (another trunk, that the outbound routes are configured to use as the first one for local calls) I entered a bad secret in the Rgistration string and the Sip Registry recognized the trunk as Not Authenticated, yet, when I made a local call, Mktg_8-out was still used and the call was completed.

Then I entered a bad secret in the Oubound Settings, the Inbound settings and the Registration string and the Mktg_8-out trunk did NOT process the call; it passed to Mktg_7-out (as it is supposed to).

However, in neither case, did the error.log get any entries.

Is this helping to zero in on the issue?

nicolas

An emtpy error_log is good news.. and not related at all with your issue. The event that FOP2 monitors for registration status is "PeerStatus", I posted a sample event before, here it is again:

127.0.0.1            <- Event: PeerStatus
127.0.0.1            <- Privilege: system,all
127.0.0.1            <- ChannelType: SIP
127.0.0.1            <- Peer: SIP/trunkprovider
127.0.0.1            <- PeerStatus: Unreachable
127.0.0.1            <- Time: -1

Your Asterisk server must broadcast an event like that when you have qualify problems with a trunk.

Again. fop2 monitors qualify events, not Registry events. A bad password does not make a sip peer unreachable, it just fails a registration event. fop2 only monitors PeerStatus events. The PeerEntries are read at startup, but the meat of realtime sip status information is actually the PeerStatus event.

FOP2 Community Forum

Unregistered and Diabled Trunks still showing on FOP2