Miscellaneous

This FAQ page collects operational notes for Miscellaneous.

My clients are getting ghost calls from weird numbers not logged in MiRTA PBX

Those ghost calls are attempts made by “hackers” to place rogue calls, usually to premium rate numbers. They start by analyzing large part of Internet trying to connect to port 5060, the standard port used by PBX and phones. If they detect an answer, they try placing some calls using different formats. If one of these calls has success, then they start to send hundred of calls to premium rate numbers. They get some money rewards by phone companies. To avoid this issue, you can place the phone behind a firewall or NAT router, allowing only the PBX to connect or you can configure the phone to accept calls only from registered server. This option has several names depending by the phone brand.

When a call has no callerid, is received as “asterisk”. How can I change it?

You can change setting the “callerid” parameter in the sip.conf and then reload sip from asterisk. Remember it will disconnect all clients connected.

My Music on Hold is starting from the middle, but I want it to be started from the start each time

There are two ways to run Music on Hold... you can run a single process for all clients waiting in queue or you can start a new process for each of the clients. Obviously the first way is preferred if you have a big number of clients awaiting, but it has the drawback of having the MOH process to stream the music continuously, so a client joining the queue will start listening it not from the start. The second way can be activated by changing the musiconhold.conf and disabling cachertclasses

[general]
cachertclasses=no   ; use 1 instance of moh class for all users who are using it,                                                                                                                                 
                    ; decrease consumable cpu cycles and memory                                                                                                                                                  
                    ; disabled by default

This setting is PBX wide and cannot be turned on/off based on tenant

When I try to login into the web interface, an error message tell me "You are not allowed to connect right now, try later", but my username/password are correct

The problem is due to the fact someone from your IP has tried connecting too many times and the system has been configured to check for past failed attempts. You can disable the "Web fail2ban" by using SSH and editing the table se_settings. MySQL root password is "passw0rd"

echo "update se_settings set se_value='' where se_code='WEBFAIL2BAN'" | mysql -u root -p

When using Mail to Fax, I cannot acces my Gmail account, even if the password is correct

Gmail doesn't like you to access your mailbox over IMAP or POP3 too often. If you want to still access it every minute (the default Mail to Fax rate), you need to enable "Allow less secure apps". https://support.google.com/accounts/answer/6010255?hl=en

400px|thumb

What should I enter to select "Voicemail Same Number" in an extension field in the New Items Defaults?

Please enter SAMENUMBERVM

I enabled Google ReCaptcha with a wrong configuration and now I can't login anymore

Please connect to your database server and run:

echo "update se_settings set se_value='' where se_code='USERECAPTCHA'" | mysql -u root -ppassw0rd asterisk

What are the devstatesender and devstatereceiver processes for?

The the related application page and the related application page are two processes that need to be always running. They are run at startup by /etc/rc.local using two scripts, restartdevstatesender.sh and restartdevstatereceiver.sh. When one of the PHP process dies or it is killed, the sh script will run it again. You should never run them manually, but always let the sh script to run them. If the sh scripts are not running, you may think of restarting the server or run them manually by executing the script as shown in /etc/rc.local.

This PHP scripts read the AMI events from one of the servers and distribute them to all other servers. These pair is scripts is needed to run even if there is a single node in the cluster.

My provider wants to receive the PAI address in clear even for Anonymous calls

In this case you can use the additional header in the provider definition with something like:

using the default outbound callerid of the extension:

P-Asserted-Identity: "${HASH(PEER,ex_cidnum)}${HASH(peerdetails,ex_cidnum)}" <sip:${HASH(PEER,ex_cidnum)}${HASH(peerdetails,ex_cidnum)}@pbx.yourdomain.com>

or using the billingcode defined in the tenant

P-Asserted-Identity: "${BILLINGCODE}" <sip:${BILLINGCODE}@pbx.yourdomain.com>

or using the newly defined variable PAICID

P-Asserted-Identity: "${PAICID}" <sip:${PAICID}@pbx.yourdomain.com>