So after all, the problem was so slow/blocking communication with the Radius server. For the future, to debug such issue you can use the exec_msg_threshold to see what are the slow parts of your script:
http://www.opensips.org/Documentation/Script-CoreParameters-2-1#toc57

Regards,

Bogdan-Andrei Iancu
OpenSIPS Founder and Developer
http://www.opensips-solutions.com

On 26.01.2016 00:18, Aqs Younas wrote:
Resolved the issue by down grading to radius 2.2. But stuck in another problem going to post another thread.

Thank you All.

On 25 January 2016 at 19:55, Aqs Younas <[email protected] <mailto:[email protected]>> wrote:

    We are still looking where things went wrong. Actually code is not
    changed a bit , same radius configuration in production server
    works perfect. We just copied the setup to new server and facing
    the problem. I see a lot of people posted this issue before but
    none have shared the solutions.

    Could you help where we need to look for radius connections.

    Much thanks for your pointers

    On 25 January 2016 at 19:18, Bogdan-Andrei Iancu
    <[email protected] <mailto:[email protected]>> wrote:

        Hi Aqs,

        I assume after fixing your RADIUS issue the timer warnings
        disappeared   ?

        Regards,

        Bogdan-Andrei Iancu
        OpenSIPS Founder and Developer
        http://www.opensips-solutions.com

        On 22.01.2016 23:09, Aqs Younas wrote:
        Sorry, I must search archive before query here.

        The problem is related to radius accounting. I am too using
        radius for AAA and this problem appears after calls began to
        terminate.
        Below threads are purely related.

        http://comments.gmane.org/gmane.comp.voip.opensips.user/31415
        https://www.mail-archive.com/[email protected]/msg30675.html

        On 22 January 2016 at 22:37, Aqs Younas <[email protected]
        <mailto:[email protected]>> wrote:

            On the start of test. Using top

        [...]



            I see opensips stuck(flood with warnings) durings calls
            termination process which leads me to manually kill
            opensisp to prevent opensips from eating all my server
            resources.

            Test is performed using sipp with 30 cps and 2000
            concurrent calls of during 6 minutes.

            :lscpu

            Architecture:          x86_64
            CPU op-mode(s):        32-bit, 64-bit
            Byte Order:            Little Endian
            CPU(s):                32
            On-line CPU(s) list:   0-31
            Thread(s) per core:    2
            Core(s) per socket:    8
            Socket(s):             2
            NUMA node(s):          2
            Vendor ID: GenuineIntel
            CPU family:            6
            Model:                 45
            Stepping:              7
            CPU MHz: 1199.531
            BogoMIPS:              4001.49
            Virtualization:        VT-x
            L1d cache:             32K
            L1i cache:             32K
            L2 cache:              256K
            L3 cache:              20480K
            NUMA node0 CPU(s): 0-7,16-23
            NUMA node1 CPU(s): 8-15,24-31

            egrep --color 'Mem|Cache|Swap' /proc/meminfo

            MemTotal:       49530560 kB
            MemFree:        29677688 kB
            MemAvailable:   45436220 kB



            Let me know if you need anything else.

            Thanks.

            On 22 January 2016 at 21:18, Bogdan-Andrei Iancu
            <[email protected] <mailto:[email protected]>> wrote:

                Yes, the "children" option - 10 should be more than ok.

                What is the CPU usage from opensips during the test ?

                Regards,

                Bogdan-Andrei Iancu
                OpenSIPS Founder and Developer
                http://www.opensips-solutions.com

                On 22.01.2016 18:08, Aqs Younas wrote:
                Hi, Bogdan

                You mean children? First I thought it is due to children(10 
default) which I increased to 500 but no avail.
                This I have in my configuration file.

                debug=3
                log_stderror=no
                log_facility=LOG_LOCAL3

                fork=yes
                children=500
                open_files_limit=99999

                Thanks for replying.

                On 22 January 2016 at 20:49, Bogdan-Andrei Iancu
                <[email protected] <mailto:[email protected]>>
                wrote:

                    How many workers have you configured into opensips ?

                    Regards,

                    Bogdan-Andrei Iancu
                    OpenSIPS Founder and Developer
                    http://www.opensips-solutions.com

                    On 22.01.2016 17:43, Aqs Younas wrote:
                    I see this warning when i am sending calls with
                    more than 10 cps. On 5 to 10 cps everything
                    seems ok.

                    On 22 January 2016 at 19:38, Aqs Younas
                    <[email protected]
                    <mailto:[email protected]>> wrote:

                        Hi,

                        I am using sipp to load test my opensips
                        (version: opensips 2.1.2 (x86_64/linux))
                        after calls get terminated i see my
                        opensips being flood with below warnings.

                        Jan 22 14:31:38 66-226-76-150
                        /usr/local/origination/opensips/sbin/opensips[22844]:
                        WARNING:core:utimer_ticker: utimer task
                        <tm-utimer> already schedualed for 1575170 ms

        [...]




_______________________________________________
Users mailing list
[email protected]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Reply via email to