Recent content by ja133

We service >3000 endpoints in production and moved all those PBXs to FS PBX

nothing wrong with going in that direction either. Maybe just have a primary and failover server. All traffic routes through the primary. Failover is only when there’s a problem or scheduled maintenance and you make dns change so that no one will ever need to know about the second login

How about two central DB servers that are replicated against each other in two separate geographic regions, and then spin up as many PBX servers as you desire, that will either connect to centralized DB server A or centralized DB server B?

That’s exactly what I did. May not be the most perfect solution but it does the job. The alternative that comes to mind is to create a lua script for each outbound call. The lua script runs a Postgres query against your configured DIDs. If there’s a match, then no diversion header is set (and...

What I do is I set a tag on all inbound calls. When placing an outbound call, if the tag condition is met, I automatically add diversion header, as this assumes that the call is being forwarded. Any call without a tag means that the outbound CID was set from the system and can safely be passed...

Something else to consider. recent FCC mandate requires VoIP resellers to do their own stir shaken. You cant legally rely on your carrier to do it for you if the phone bill is coming from your own company.

If you are experiencing memory leaks then you should look into your programming. I manage multiple servers that handle 1000 endpoints each and I have not rebooted in several months

While this patch helped resolve the issue at 00:00 on 1/1, there is still another issue that someone else brought up. Long story short, this issue can still be reproduced after the new year when freeswitch gets restarted. So I am recommending to revert back to original code while the developers...

No, it wouldnt make sense to have to pause replication every time you wanted to back up. That wouldnt be ideal practice. Proxmox images are for disaster recoveries or scheduled maintenances That said, if you are performing scheduled maintenance and need to move things around, that’s one...

Yes. Disable PG replication on both nodes. Take your snapshot. Restore. Re enable replication. The command, off my head, is: alter subscription <subscription name> disable;

In theory it should although I have not tested that. If I take a snapshot of my instance right now without suspending PG replication, and in an hour from now I restore that snapshot, in the last 60 minutes there would have been, at minimum, CDR transactions that took place. If the database...

Restores are straightforward If you use DB replication you just need to get that re-established with the new IPs and/or with the most up to date database information

You basically just validated what I wrote. I can't recall having to coordinate with my vendors every time I need to upgrade their software. Imagine having to call Apple every time you need to update your iPhone? Do I have to take my Tesla to the service center every time I need to upgrade that...

We are 90 minutes past 00:00 UTC which is the time set on all our servers. No issues this year! Tested BLF lights as it seemed too good to be true! Thanks for the Christmas miracle!! HNY Quick AM update: everything is still stable, no abnormal behavior is observed

@UCtech I have to jump in and side with @pbxgeek on this. Mark and the Fusion team really are great people, and they’ve always been solid to work with. This isn’t a knock on them personally at all. But from an operational standpoint, we hit a point where we had to try something else because...