Upgrade to edge using
fwconsole ma upgrade endpoint --edge
if problem persists, open a support ticket.
Upgrade to edge using
fwconsole ma upgrade endpoint --edge
if problem persists, open a support ticket.
I would think you’d want to get a product stable before releasing it. The quantity, age, and impact of all these edge updates that aren’t considered stable yet is not a glowing recommendation.
backup was a huge change for 15. you might want to just set it all back up again from scratch.
You have a fundamental misunderstanding of what edge means and how it is employed. It’s existence INCREASES reliability of a system. I was sure we had an old blog post describing it in detail, but I can’t find it.
Again, module updates are published continuously throughout the entire supported life of the product. Before any new code is released to all users in the stable repo, it spends some time in the edge repo so it can be tested and so we can do a proper QA.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.
@lgaetz I noticed there are two dates on the Commercial Modules: Expiration Date and Free Updates until (which is sooner than the Expiration Date).
I seem to recall (and that’s never very reliable) that the rough time in Edge for a module was about two weeks. That no longer seems to be the case.
Can you talk more about the progression from edge to stable? What is the QA schedule like, and how long should we expect? Do some modules need more time in edge than others?
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.
I’m not aware of any detailed documentation, but it is all open source. You could start by examining the code for a very simple Application, such as Misc Destination.
For links, I recommend putting them on a line by themselves, replacing the last for with %2E e.g.
www.example.com
becomes
www.example%2Ecom
On a mobile, this works as a link. On desktop, triple-click to select it, then right-click and choose Go to …
This is probably elementary for some of you guys…
I have a user who has a DID to his phone that prepends the CID with “sales.” He knows if someone dials his DID because of this, opposed to queue calls, etc.
When this user forwards his phone and someone calls his DID. I don’t want to pass the originating caller ID. I want him to know someone dialed his DID.
I tried find me follow me hoping to produce a different outcome, but no dice.
Thoughts, suggestions, solutions, appreciated.
Jason
Figured it out… duhh…
Fixed CID value in FM/FM.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.
The license for use of the module is 25 years. The first year includes free updates/support. You have to pay an annual support fee per module to have the ability to get current updates / support on the module.
Thank you @BlazeStudios. There is good support here but I was concern about the updates, if I am a year out and my module is 14.0.12 will it stop there? or will I get 14.0.27?
Can you be more specific in what you are looking for?
I spent over 5 years down cumenting development in FreePBX. It is all in the wiki.
Are you looking for how to recreate it?
Generate appropriate dialplan including AGI scripts. Use the asterisk manager interface for more direct interactions.
https://wiki.freepbx.org/plugins/servlet/mobile#content/view/62359662
Sorry for the mobile link
But how do we upgrade?
The ‘beta’ upgrade tool still has ‘Beta Warnings’. I tried uninstalling the upgrade tool in case there was an updated version lurking in the repos but no luck.
This is too much of a tease to announce that the stable version is finally available… …when it isn’t!
I wanted to add a bit more on top of the blog post above from @jfinstrom back in 2016.
Edge module releases were originally two weeks.
That was back in 2016 before FreePBX had a proper QA team in place.
With the introduction of the full Sangoma QA team the time frame changes drastically for each module and depends on the complexity of the issue therein. Some issues can take a week. Some can take two weeks and others can go on for a while. QA does its testing from edge because that is the closest release to the user base they can get to (encoded and packaged and signed) so it’s the best way to test.
This means that modules can appear to be forever in edge while older releases are pushed out while new features or issues are worked on. In the past modules were just published when the tickets were closed.
This caused users to complain that modules releases were happening too quickly and sometimes with unstable bugs. Since the development cycle of freepbx is ongoing and doesn’t follow the same development cycle as asterisk we decided to have a staging system. Sort of like how chrome has a beta. Canary and nightly (and actually the backend mirror server supports all of these tracks but we only ever implemented edge!)
Since modules are continually worked on I doubt you’d notice very many modules that only say stable with no edge releases. Especially with over six people working on freepbx development full time!
If you have a pbxact system you notice this even less because modules are pinned to specific versions of releases. Even edge modules are pinned.
I believe when 14 went stable we had quite a few modules in Edge as well and even if we didn’t after 5 days we definitely had quite a few.
With a full QA team in place release cycles of Edge modules should probably be re-clarified because the blog post from 2016 and user expectations are a little outdated.
Thats not bad. It’s just progress.
You can do these same things without BMO. The difference is that BMO introduced object orientated code into freepbx. Before you’d have to write procedural code to do install upgrade remove backup agi through functions.inc.php. The bad thing about functions.inc.php is that they are loaded for every module on every freepbx run (page laid. Scripts. Etc). Since BMO is object oriented that means it can take advantage of auto loading. So module functionality is only loaded when requested instead of each request. This means lower memory footprint and faster performance