TIPS incrediblePBX googlevoice trunks (no audio)

[Merlin]

Am I just one of the lucky ones or is there something I'm missing? I've been using the Oauth with Wazo - Version: 17.17 "Phoenix" and outbound calls work fine, but inbound goes straight to vm after 30 seconds.

 

[HermanMiller]

A New Day: Google Voice GVSIP Now Available for FreePBX

Ok. So great progress and glad to see this moving but for me not quite where I need it as yet. What I really need is something I can apply to my running iFPBX 13-13 enchilada from a few months back. Looks like we're not there yet but also not to far away.
Think we might see this within the next week or 2 or more like a month or more? Just roundly asking. For now GV still working, using with a 'callback' mod I found on DSLR to mitigate the silent ring and one way audio. So barring goog pulling the plug hard I'm good for the moment.

I'd join you guys in the deep end but you'll only end up dragging me out and someone performing cpr on my lifeless corpse...

 

[Timma]

Ok. So great progress and glad to see this moving but for me not quite where I need it as yet. What I really need is something I can apply to my running iFPBX 13-13 enchilada from a few months back. Looks like we're not there yet but also not to far away.
Think we might see this within the next week or 2 or more like a month or more? Just roundly asking. For now GV still working, using with a 'callback' mod I found on DSLR to mitigate the silent ring and one way audio. So barring goog pulling the plug hard I'm good for the moment.

I'd join you guys in the deep end but you'll only end up dragging me out and someone performing cpr on my lifeless corpse...

I think you may have over estimated the complexity.

This was literally as easy as running a bash script. I would think if you have been tinkering in the interim, you would probably want a fresh install.

I don't see why you couldn't back up your extensions and config, fresh install, and after that gvsip-naf install & restore it. It does update SSL, and from what I saw before I took the advice and grabbed a cup of coffee, asterisk also.

I'm a plain asterisk guy, so ipbx is new to me. I'll let someone else advise you on your current config and upgrading to gvsip.

I gotta say, it was a very smooth easy process to get going.

Reach out if you need any help. The script he linked is pretty much ready for general use as far as im concerned. I am no expert with ipbx, but Im pretty seasoned for installs, hardware, asterisk and voip in general.

Regards,

 

[HermanMiller]

I think you may have over estimated the complexity.

This was literally as easy as running a bash script. I would think if you have been tinkering in the interim, you would probably want a fresh install.

I don't see why you couldn't back up your extensions and config, fresh install, and after that gvsip-naf install & restore it. It does update SSL, and from what I saw before I took the advice and grabbed a cup of coffee, asterisk also.

I'm a plain asterisk guy, so ipbx is new to me. I'll let someone else advise you on your current config and upgrading to gvsip.

I gotta say, it was a very smooth easy process to get going.

Reach out if you need any help. The script he linked is pretty much ready for general use as far as im concerned. I am no expert with ipbx, but Im pretty seasoned for installs, hardware, asterisk and voip in general.

Regards,

Thanks for the offer Timma.

I was put off by the fresh install because I just have too many balls up in the air right now to risk breaking something that currently works well enough. Then that turns into yet another 'ball in the air'.

I'll keep watching and maybe I'll jump in when resources loosen up. I'd much rather get this handled before it breaks but for the moment the risk profile is it's in the middle of, rather than the top of, the todo list. Stay tuned though... if that changes I may yet take you up on that offer

 

[HermanMiller]

Spoke too soon I guess... TIL, not only is that Murphy a SoB but he's good at it...

So now I can't make out going calls started sometime yesterday I think. I noted the day before that on my Android phone the GV app updated with a benign but different message than usual. I'm assuming that because xmpp plug was pulled? (or maybe special just for me??)

Ok, so now I guess I try the path with a fresh install. Here goes nothin'.....

 

[HermanMiller]

Ok, so I knew there was a reason this was gonna go the hard way - I'm on CentOS 7 - Naf GV doesn't play nice w/ 7 and no backup (I don't think) from a 7 system to a 6.
Rebuilding on a second, parallel VM w/ 6 following Whole Enchilada (http://nerdvittles.com/?p=23948) but notice a lot of errors in first script run. At the first reboot stage so we'll see.
Just reporting progress so far.

Edit 1 - after reboot and login I got the ipbx update prompt. Cancelled, and at command line went to run the installer script again as instructed. Get an error about 1) my terminal size (nonsense, term was larger and tried again full screen on portrait 1080x1920, same error) and 2) telling that IPBX already installed:

****** WARNING WARNING WARNING WARNING ******
Window size is too small: 80 x 24
Resize console window to at least 82 x 30.
Otherwise, Asterisk may fail to compile.
****** WARNING WARNING WARNING WARNING ******

Incredible PBX is already installed.
WARNING: Always run Incredible PBX behind a secure hardware-based firewall.
root@huma:~ $

Not sure if/what, but going to proceed with instructions at next paragraph as there it says after reboot I should see the updater script. Maybe just the tut needs updating to match installer?





Sidebar - that article needs some other fixing.
a) I think gotta make it clear that you cannot choose CentOS 7 if plan to use pjsip/naf GV.
b) fyi Vultr no longer gives a IPv4 address with the $2.50 plan so you need the $5/mo plan minimum (can add a v4 IP to the $2.50 for an add'l $3 which makes no danged sense cuz it's $.50 more for half the ram and bandwidth..)

 

[HermanMiller]

Yeah, so that didn't go as planned at all

[email protected]'s password:
Last login: Sat Jul 21 20:19:56 2018 from 69.115.252.251

[H[2JThis update utility goes to IncrediblePBX.com to retrieve the latest updates.
We test updates before release, but NO WARRANTY EXPRESS OR IMPLIED IS PROVIDED.
The first 10 updates are free. Voluntary calendar year update license is $20.
To sign up and make payment using a credit card, go to: http://nerd.bz/QwQkYO
 
To proceed solely at your own risk, press Enter. Otherwise, Ctrl-C to abort.
 
Retrieving... tar (child): lastupdate7.tar.gz: Cannot open: No such file or directory
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now
rm: cannot remove `lastupdate7.tar.gz': No such file or directory
/root/update-IncrediblePBX: line 23: /root/lastupdate7: No such file or directory
rm: cannot remove `lastupdate7': No such file or directory
-bash: pbxstatus: command not found
]0;root@huma:~[?1034hWARNING: Always run Incredible PBX behind a secure hardware-based firewall.

reading back the session log I see tons of errors during the long part of the install from PYCURL unable to resolve hosts for stuff (stuff that resolves from my dekstop). On closer inspection I can't resolve anything from the command line (simple pings for example) and nslookup doesnt exist on the system. Not sure wth. Gonna wipe and start afresh I guess..

 

[kdthomas]

It works, but yeah the instructions are a little all over the place at the moment. I made my own notes below that I used to build a working Incredible PBX with Full Enchilada and Hylafax using three GV trunks.

Install CentOS 6.9 Minimal (64 bit), Configure Network to be Connected on Power On, and set IP Address

cd /root
yum -y update
yum -y install net-tools nano wget tar
tar zxvf incrediblepbx-13-13-NAF.tar.gz
rm -f incrediblepbx-13-13-NAF.tar.gz
./create-swapfile-DO
./Incredible*
Reboot
./Incredible*
Reboot
/root/admin-pw-change
/root/timezone-setup
./Enchilada-upgrade.sh
Reboot
./incrediblefax13.sh
Reboot
/root/avantfax-pw-change
/root/gvsip-naf/install-gvsip [Repeat for GVSIP2, etc.]
/root/admin-pw-change
/root/timezone-setup

Change PHP Timezone in Settings, Advanced Settings in the Incredible PBX GUI

Add route for Inbound for Fax Number to custom extension HylaFax

In the Incredible PBX GUI, navigate to Connectivity:Outbound Routes and click Add Outbound Route. In the Route Settings tab, give the Outbound Route a name and choose one or more trunks to use for the outbound calls. In the Dial Patterns tab, specify the dial strings that must be matched to use this Outbound Route. NXXNXXXXXX

Add an Outbound Route for each trunk (GVSIP1, GVSIP2, GVSIP3, etc.)

To delete trunks: /root/gvsip-naf/del-trunk

To refresh the patched version of Asterisk, copy /etc/asterisk/pjsip_custom.conf to a safe place, delete the contents of pjsip_custom.conf, rerun the installer, and then copy your version of pjsip_custom.conf back to /etc/asterisk and restart Asterisk. That way you won't lose any of your previously configured GVSIP trunks.

 

[HermanMiller]

Yeah smth was definitely screwy the first time through. I blew away the vps and started fresh and things ran as expected. After taking IPBX backups and downloading all my call monitor recordings I restored the IPBX backup to the target and realize now that it restored my GV Motif settings (duh!).
Not sure how I should proceed. I could:

1. I took an IPBX backup of the virgin setup from the target before I restored so I have the option to roll back (but I really don't want to set up all my phone extensions and groups, etc. by hand again and that's where I'd be.)
2. I can delete all the motif stuff in the gui. Obviously I never ran the corresponding cli scripts on the target so maybe that's safe? and then just proceed with the naf-gv setup as though nothing happened? I'd like to think so.

grrr. decisions decisions...

And then ultimately I need to rsync this whole server back over to the original vm (to keep my plan.)
So not quite out of the woods yet.

 

[HermanMiller]

So just an update, finally got everything working.
What I did was after I restored the ipbx backup on the target I went in the gui and deleted the trunks. I could not delete them at the cli (I recall a cli utility called del-trunk or smth like that and it no longer exists, neither on the target nor on the original source box so I'm assuming got del'd in recent ipbx update?)
Anyway, after del the trunks I ran the naf install to set up the trunks. It set up but the trunk kept showing unregistered. I had to grope around a lot to figure out what goes on but ultimately what got me running was removing the naf trunks, going to goog oath pages to the oauth management, revoking the previous token, then getting a new token and using that. For at least one of my trunks I still had a problem and discovered that at the mgmt page (for whatever reason) it did not show that I'd granted IPBX access so if anyone is having trouble it might be worth checking there to see if G believes you've granted access.
After that the rest was straightforward per instructions. Set up my outbound routes and back in business.

Thanks to naf and pjsip contribs everywhere for getting this working and thanks forum and wm for infos here which helped me grope my way through figuring it out.

Next steps are how to get it back to my original VPS on vultr (rsync I guess but I have very little relevant exp; Target partition is smaller than source so vultr snapshot or backups won't work) and then convert my extensions to PJSIP (saw a thread on DSLR and one of the claims was better sound qual iirc.)

Thanks all.

 

[wardmundy]

@HermanMiller: After rereading this thread, I would only offer this suggestion. YOU ARE MAKING THIS TOO HARD.

If you're using a cloud-based server already, then I see little reason you wouldn't have tried the cloud-based VPS deal we set up for everyone at HiFormance. At about $1 a month, it provides equivalent performance to Digital Ocean or Vultr at 20% of the cost. Setup takes less than one minute. Once you had a working platform there, you then could have run the GVSIP installer to bring the platform up to speed. That takes about 20 minutes. Then you could have ported your settings from any other platform using cut-and-paste in under an hour and been done.

Just a final heads up, when you start trying to image a VPS platform with rsync and move it somewhere else, your odds of failure are astronomical. Every one of these cloud platforms tends to be proprietary and, when you interject code from another provider, you will be lucky to get it to run at all. And, if it runs, the chances of regular system failures because of incompatibilities are HUGE. Then you'll be writing dozens of posts that your server is unstable and wondering why. So... DON'T DO IT!

 

[HermanMiller]

Ty.

@HermanMiller: After rereading this thread, I would only offer this suggestion. YOU ARE MAKING THIS TOO HARD.

I guess that happens when you are learning. Much of this is new to me. While I'm somewhat Linux-capable, I wouldn't call myself (even proficient) by any stretch. So sometimes you grope around in the dark till you find a light switch. Hopefully you learn something for next time too.

If you're using a cloud-based server already, then I see little reason you wouldn't have tried the cloud-based VPS deal we set up for everyone at HiFormance. At about $1 a month, it provides equivalent performance to Digital Ocean or Vultr at 20% of the cost.

Maybe I mis-read or maybe it's changed but I thought hi-formance was touted as a cheap 'sandbox' and not really to be relied on for 'production' use. My system is small and I don't need much in the way of resources but uptime and having my phones working is important to me.
Please correct me if I'm wrong about hiformance.

Setup takes less than one minute. Once you had a working platform there, you then could have run the GVSIP installer to bring the platform up to speed. That takes about 20 minutes. Then you could have ported your settings from any other platform using cut-and-paste in under an hour and been done.

This was what I did not know. In the end I did do this, just with another VPS at the same host (vultr). I just want now to migrate back to my original (smaller plan.) I guess the easiest way at this point would be to just fresh install that server, build iFPBX, add naf and then backup/restore my FPBX settings using the backup module in the admin.
Or cut and paste what files?

Just a final heads up, when you start trying to image a VPS platform with rsync and move it somewhere else, your odds of failure are astronomical. Every one of these cloud platforms tends to be proprietary and, when you interject code from another provider, you will be lucky to get it to run at all. And, if it runs, the chances of regular system failures because of incompatibilities are HUGE. Then you'll be writing dozens of posts that your server is unstable and wondering why. So... DON'T DO IT!

Yes, this I've come to realize. I have read dozens of posts on it at this point and much advice about what to exclude (/proc, system directories, and on and on) but the one thing that has stood out is it's not a simple cut/paste operation nor a one-size-fits-all.
I did also read a post by you at some point (pretty sure) how to rsync a cloud IPBX server back to a on-prem hyper-v for a cheap backup/failover type of solution (standby server I guess is more correct.)
I have to find and re-read that I think because I will understand it more thoroughly now.
It may *also* be a solution for this situation though I did try it back then and it did not work for me.
(yes, the above rebuild seems easier. But this might teach me more and so long as my 'new' server is working I can perhaps play a bit without much risk. I can always revert to the above solution if/when this falls flat)

In retrospect anyway, perhaps the question is not "how to rsync my whole server from A to B" but rather "What all directories are used by FPBX/Asterisk/Incredible that are needed to move/sync from one box to another to have a mirrored FPBX/Asterisk/Incredible?"

I'm assuming this is practical in some way since there is your article on it and your article on the rsync to local hypervisor might answer just that (though iirc it too was a full server type of imaging.)
I'll have to go find it again, read it and see. Or maybe something new can come from this. Since 'everything in linux is a file' I have to imagine it's possible to just rsync some subset of the file system from a to b and have fpbx work. Or am I delusional and/or misinformed?


Thanks Ward. Thanks all.

 

[Dieter Preiser]

After a fresh install I have four issues with my system:
1. GVSIP trunks give me intermittent no receive audio on outgoing calls. The call connects and the other party can hear me, but I cannot hear them.
2. Outgoing calls using GVSIP trunks do not provide ringback tone. There is silence until the other party answers.
3. HylaFax works fine for incoming faxes, but will not work for outgoing faxes. Always get messages like: "requeued Busy signal detected". The dialed number or numbers are not busy.
4. I've set one of my GVSIP trunks with an inbound route to extension 53669 (Lenny) but the inbound call with not complete. There is a long silence, the a beep, then disconnect. Extension 53669 works with internal calls.
Update on 8-1-2018: Did a complete new install again. Only problems #2 and #4 remains. Will try to troubleshoot #4 via the log to figure out what is going one.

 

[HermanMiller]

After a fresh install I have four issues with my system:
1. GVSIP trunks give me intermittent no receive audio on outgoing calls. The call connects and the other party can hear me, but I cannot hear them.
2. Outgoing calls using GVSIP trunks do not provide ringback tone. There is silence until the other party answers.
3. HylaFax works fine for incoming faxes, but will not work for outgoing faxes. Always get messages like: "requeued Busy signal detected". The dialed number or numbers are not busy.
4. I've set one of my GVSIP trunks with an inbound route to extension 53669 (Lenny) but the inbound call with not complete. There is a long silence, the a beep, then disconnect. Extension 53669 works with internal calls.

#1 - Have not had the same experience - all my calls so far are working both ways
#2 - Confirm, same here. Haven't had time to play. I thought, when following the dslr thread, that gvsip-naf was enabling early media and we should get ringback from there. Haven't had any time to dig into it at all though.

#'s 3 and 4 I can't speak to.