# vim:tw=80:ft=txt

These are known issues with CallWeaver


APP_QUEUE CDR LOGGING

  Problem
    - If having members in a queue reachable via Zap like member =>
      Zap/g1/somenumber the member is called up correctly, but the call never gets
      logged to CDR.
  
  Solution
    - To have callweaver log CDRs correctly, change this to member =>
      Local/somenumber@somecontext and add [somecontext] to your extensions.conf
      for chan_local use.


CHAN_SIP CALL COUNTER ISSUES

  Problem
    - If using call-limit in chan_sip, counters may not be decreased after end
      of call. This is documented in trouble ticket #45, see
      http://callweaver.org/ticket/45.

  Solution
    - Use dialplan logic with the GROUP and GROUP_COUNT functions intead of the
      counters.

RTC LOST INTERRUPT

  Problem
    - My kernel logs are filled with:
      Sep  6 11:55:06 callweaver kernel: rtc: lost some interrupts at 1024Hz.

  Solution
    - check: cat /proc/sys/dev/rtc/max-user-freq
      if less then 1000 use: echo 1000 > /proc/sys/dev/rtc/max-user-freq
      put this in your startup script.
    - if the problem persist check the output of "dmesg", lost interrupt can 
      be also caused by problems with your hard disk(s) (some bad sectors).

CLOCK_MONOTONIC SUPPORT IN LIBC

  Problem
    - For better timing performance CallWeaver need that your system libc has
      support for CLOCK_MONOTONIC 

  Solution
    - on Linux, you have it with GNU libc6 >= 2.3.3
    - on BSD, you need OS version >= 7.0

CANNOT DIAL NUMBERS THAT START WITH '#'

  Problem
    - Some SIP client uses UTF-8 encoding to manage datas and give some 
      problem dialing numbers that start with '#'

  Solution
    - First of all check if your phone does not use '#' as a "send" key
    - Second just put "pedantic=yes" in your sip.conf, Snom phones are
      affected by this.

OTHER BUGS

  - For more info about unsolved and solved bugs, please see
    http://callweaver.org/tickets.