Voicemail drops after 20 seconds always

[swiftechinc]

Hopefully I'm overlooking something obvious. v18 latest release, professional edition hosted on Amazon EC2.

A few days ago I was running out of space and trying to find out how to increase storage space. I shut down the server on EC2, increased the disc size, restarted, and learned that more needs to be done. Likely I need to simply setup a new drive, backup, reinstall, and restore.

As I was planning to do this I saw that for some reason, Amazon just totally lost my static IP while the instance was down. I pay more for that than the /22 I have from ARIN so you would think it would stick around a few minutes. lol So I got a new IP because support at EC2 is too slow for me to wait for them to try to fix it. I updated it everywhere it needed to be updated and in 10 minutes everything was working fine. I found that I had about 6 gig of backups that were created before I moved to storing backups on a separate drive (would be nice if clicking on backups showed all backups from all sources with a column that showed the location) so I deleted them, found I had plenty of space now, and was content to leave it alone for now.

Then I started getting complaints. Voicemails drop at almost exactly 20 seconds. It doesn't matter if it's a direct call to a DID, routed through the system, external, internal (ext to ext), softphone enabled or not, etc. I used a land line to call my own extension directly, and started a timer at the point the recording portion began... 19 seconds. although fractions can add up to an exact 20 which makes more sense.

I cannot for the life of me figure it out. I found some old articles, such as with version 9 that there was a setting in "Parameters" that could be edited. But it's no longer there. I found that I can extend over the 2 minute mark with some command line options. But that doesn't explain WHY it would be doing this.

Is there any other setting anywhere that might cause this to happen as a side-effect? I'm almost to the point where I will restore the entire system but if I've backed up a bad config, all I'll get from that is a bad config.

 

[leejor]

Have you made use of the 3CX Activity Log, for any clues, as to why this is happening? Does this also happen when checking your own messages, or just when leaving a message?

 

[ChrisC_3CX]

To get a bit more information on this:

• How much HDD space is currently available as seen from the Management Console Dashboard?
• Does this happen for every single voicemail call?
• Can you create a new test extension and try to leave voicemail for that extension. do you get the same behavior?
• Does this by any chance happen for normal calls too (not to voicemail boxes)?
• Have you restarted the system and tried again at all?

 

[swiftechinc]

Well, I can't tell you how weird this is. I had rebooted the EC2 server twice and from the services from the console three times. I noticed that the 3cx instance seemed to reboot entirely too quickly so after posting this I went through and rebooted one service at a time so I could watch it. When I was done I made some test calls and it all worked properly again. So all good there.

Then I noticed something else.

Earlier in the week I had shut it down and increased the disc space on EC2 because I was running at about 97% with just under a gig of storage left. When I booted back up and logged into the system I was still at 97% despite having doubled the disc to 30GB. It took me a few seconds to remember reading a year or so ago about the containers and that I had to either use the command line to increase the container size or use the backup/expand/restore process. At the time I had also stumbled across a much simpler way to do it and remembered bookmarking it but I lost that bookmark for reasons I won't get into.

So I started hunting all over for that simpler method. I even put up a post here hoping someone else knew how to do it. The only responses I received were basically, the command line edit of the container size or the backup/restore method. I started thinking I had simply remembered it wrong. Rather than go through all that trouble I thought I would go through and look for possible ways to reduce the disc usage.

This resulted in me finding over 6GB of backups from last fall just sitting on the local drive. These were from before the connection to Google Drive so I deleted them. I didn't see them before because you have to actually change backups from cloud storage back to local storage to make them visible. (Something I think should be addressed by 3cx). I deleted all those old backups and felt that the additional 6GB would allow me to put off the drive storage for quite some time.

So I was done messing with it. Earlier in the week I had to change the IP throughout the system because it just disappeared when I shut the server down to upgrade the storage...Don't ask how or why. I have no clue.... It was indeed a static, I've shut down numerous times over the last few years and kept the same IP even when I did the export/install/import routine last fall to upgrade to v18 from v16).

This bizarre 20 second voicemail cap started about the same time that I changed the IP. I didn't have a clue how in the world that could cause the problem but it seemed like the only thing that made sense. I had already done all of the things mentioned above and was stumped. Out of ideas I thought "what the heck" I'll reboot 3cx one more time and reply to both of your comments. So I logged into Amazon and rebooted there, then came back to 3cx and rebooted the services in the console.

Sure enough, the problem remained. I was about to come back here and report the failure and hope that someone had another idea. But something seemed off. The reboot of the console seemed to come back up far too quickly compared to usual. It kicked me out and made me log back in as usual but normally I have to wait a few seconds before it lets me log back in. This was almost immediately after kicking me off.

So I decided to investigate this. I went into the console and rebooted each service one at a time and watched them reboot.

After this I made yet another call from my cell to another user and left a voicemail....and it recorded me for a full 2 minutes!
Why rebooting one service at a time worked where a normal reboot didn't, I'm not sure. Maybe someone else knows.
But what's I really want to know is how THIS ALSO EXPANDED THE PARTITION TO THE FULL DISC SIZE of 60GB?!?
I think I may have stumbled onto that alternate method I had found before.

I'm seriously considering taking the weekend to test this further by increasing my disc size, then rebooting the services one at a time. If anyone knows why or how this would make any difference at all, that would be great because I'm at a complete loss. It doesn't make any sense to me at all but if I can nail it down I think it would be a great thing for everyone to know.

So now I'm humming along quite happily. Voicemail is no longer cutting out, and I now have an additional