Jump to content

mc_

Administrators
  • Posts

    1,796
  • Joined

  • Days Won

    4

Posts posted by mc_

  1. Depends on your definition of easy :)

    For the most control from your perspective, a Pivot callflow would allow you to maintain dynamic access codes and return the necessary callflow JSON to either complete the call in an authorized scenario, or respond/play media in an unauthorized scenario.

    If the access codes are more static, you could look at the menu callflow action and have the valid access codes be child branches of the menu action. Just back of napkin here but I could see:

    1. Set the callflow's "patterns":[..] to match international number formats, this way you won't conflict with extension of local dialing rules

    2. First action of the "flow" is the menu with a prompt to enter access code to authorize the call

    3. Children for each access code and a catch-all for invalid entries

    4a. Valid child branches will then have the "resources" action to route the call accordingly.

    4b. Invalid will play a media file or respond with SIP error codes.

     

    Maybe this can give you a little direction? You could also look at the DISA callflow to see if that would work better (similar step 1 though to match international calls).

  2. I think there's some confusion on terminology here; let's clarify a bit :)

    Pivot is an app that, on a per-call basis, sends an HTTP request to your server and receives callflow instructions (versus configuring the callflow statically). It has nothing to do with metaflows directly.

    So when you say "testing the metaflow through PIVOT" I'm not sure what that means.

    How are you initiating calls? What are the steps to recreate the issue you see? Basically, what are the steps in your two scenarios, from start to finish, for how the initial phone call is placed up until the call ends?

    Another thought, device ID is typically sent as 'Authorizing-ID' when the call is started by a known device, perhaps that is present for you to use?

     

    Finally, if you're in a bad spot, you might consider contracting with us for support, officially, and get dedicated engineering time to help you move forward. Obviously that's a business decision on your end but figured I'd mention it.

  3. It appears that app (not a 2600Hz monster app) is not compatible (at least not directly) with KAZOO 4.3:

    root@four monster-ui]# sup crossbar_maintenance init_app /var/www/html/monster-ui/apps/whitelabel http://192.168.122.119:8000/v2
    trying to init app from /var/www/html/monster-ui/apps/whitelabel
      failed to incorporate app because there was no app.json in /var/www/html/monster-ui/apps/whitelabel/metadata

    Unless there are missing steps to getting that metadata directory populated, anyway. Compare to SmartPBX's: https://github.com/2600hz/monster-ui-voip/tree/master/metadata

    For the Spanish stuff, I can only point to Monster's docs: https://docs.2600hz.com/ui/docs/internationalization/

    In any case, I think it is important to distinguish between MonsterUI (the web frontend) and the KAZOO backend - they are two distinct entities. But I also think the internationalization capabilities of Monster have improved (we have more native Spanish speakers on staff so hopefully that's the case :) ).

    Thanks for sharing what bogged you down. Hopefully you'll give it another shot at some point!

  4. The control queue is "ecallmgr@sandbox.testcomp.com-ecallmgr_call_control-<0.363.4>-35669c98". This tells you that the control queue is on the "ecallmgr@sandbox.testcomp.com" Erlang VM, an "ecallmgr_call_control" process at PID "0.363.4". Your logs, afaict, don't include the logs for that PID. You may need to increase ecallmgr's log level. sup, by default, talks to kazoo_apps@{hostname}. You can include "-necallmgr" to run commands against the ecallmgr VM: "sup -necallmgr kazoo_maintenance syslog_level debug" for instance.

  5. @Guillermo Prado sorry to hear your frustration with the platform. What in particular was out of date or poorly documented for you? What was hard that you thought should be easier? Would be great to know where to direct some attention.

    For reference, https://docs.2600hz.com/dev is refreshed every time a commit is merged into the master branch of KAZOO so it should always reflect the latest and greatest. Admittedly there is still some work to be done to backfill more information but we need to know what people are using so we can focus there.

  6. Possibly permissions on the database files themselves (I think they're under /srv/dbs ?). Try accessing a single bigcouch node and see what its logging.

    Also, n=4 (or any even number) is problematic in a network split. It is generally good to have n be odd to make it more likely to have a majority on one side of the split.

  7. I mean the RPMs aren't hosted by 2600Hz anymore, as far as I know anyway. Perhaps there's an archive somewhere.

    But more importantly, why can't you upgrade production? There are a myriad of reasons why you should be upgrading, chief among them security improvements. What limitations are you up against to upgrade off a 3+ year old version of KAZOO?

  8. Depends what you mean by "features" :)

    There are a lot of internal changes, optimizations, and improvements around:

    • Leveraging more FreeSWITCH functionality
    • Overriding default Erlang distribution mechanism with kz_dist - will improve cluster comms
    • Preparing for the breakening - effectively separating kazoo applications out into their own repos
      • Moving community apps like ACDc, Frontier, etc to a new Github org - kazoo-community
      • Will allow the community to drive the apps forward without blocking on 2600Hz review/merge
    • Per-app dependencies
      • Each Kazoo application can now incorporate dependencies without requiring all installations to fetch them
      • Will allow us to clean up deps/ with "common" dependencies and per-app individualized deps
    • /v3 of the API will begin; /v1 official deprecation

    If there's anything in particular you're interested in, let us know!

×
×
  • Create New...