Vicidial displaying DEAD status instead of INCALL

[kamirie]

Good Day,
ViciBox 3.1.14
AstGui VERSION: 2.4-351a
Asterisk 1.4.38

The problem I'm facing with this server is that instead of showing INCALL status , in the realtime monitor all connected calls are tag as DEAD and in agent gui , all calls are marked as Call-Hangup after a few seconds of being Live Call, but the line remains connected and agents can talk with the other line. And also can't do calling with predictive , when the dialer is set to ratio , after a couple of minutes the dialer will stop dialing leads , it just totally stops. No any movement in CLI . I'm not the one who installed or build this server and just took over the managing of it. What can cause this problems ? it's the first time that i encounter something like this.

Thanks

[williamconley]

Is this a new problem in an older server or a new server that has never worked?

It can be related to many things such as bad networking, corrupt DB tables, wrong IP in vicidial's config file ...

Have you rebooted recently?

[kamirie]

I think this is a new problem on an older server that was used before. My suspect was corrupt DB tables because i just suddenly lost every record on the lists and was able to recover by using mysqlcheck --auto-repair but the original DEAD status problem didn't go away , and after 20 mins i think the db crash again because i lost every record in lists again and can't generate any reports.

[williamconley]

Your hard drive is probably full. LOL

Code: Select all

df -h

you can freely delete any files inside /var/log/asterisk and /var/log/astguiclient to free space up, but you'll need to find other space as well most likely (such as removing old recordings ...).

[bmorrison]

William's advice is solid. A full HDD/SSD will cause this kind of behavior. When I get a call from someone needing help with this issue, this is what I do:

First I check the space available on the drive:

Code: Select all

df -h

If the drive is anywhere near or over 90% full, I begin looking for files to delete. Using this command, I get a list of the 10 largest directories on the system in /var, where most of the disposable files reside:

Code: Select all

du -a /var | sort -n -r | head -n 10

Quite frequently, I'll see that among the largest directories is /var/spool/asterisk/monitorDONE. This is where your call recordings live. At that point, I get the owner back on the phone and tell them that either we need to delete, or better yet, move their call recordings off the server and onto another server via FTP. After figuring out what they want to do, I usually end up removing call recordings that are older than X number of days (in this example, a day):

Code: Select all

find /var/spool/asterisk/monitorDONE/ORIG -maxdepth 2 -type f -mtime +1 -print | xargs rm -f

Depending upon the number of recordings needed to be deleted, this operation could take some time. After it completes, I check the drives available space, and if there's (much) more room, I reboot the server and ensure that everything comes back okay.

This, of course, is just one example. There are many ways of going about freeing up space on a drive that's filled to capacity.

[williamconley]

Noting of course that ORIG contains duplicates of the files before compression and can generally just be deleted out of hand (to get a server back online quickly). Then the FTP process can be used to duplicate the data that the recently scared business owner now wants to back up (wake up call accomplished) and a good DB backup is a great idea at this point as well, also pushed off the server to an FTP location.