[OpenSIPS-Users] timer execution delay

[Bogdan-Andrei Iancu]

Hi Pasan,

Are you able to run (just for testing purposes) opensips directly on the 
host machine? I just trying to see if the difference will impact.

Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
   http://www.opensips-solutions.com
OpenSIPS Bootcamp 2018
   http://opensips.org/training/OpenSIPS_Bootcamp_2018/

On 09/20/2018 07:20 PM, Pasan Meemaduma wrote:
> Hi Bogdan,
>
> Nice to here from you :) Thank for clarifying about the warning. My 
> delay times are only vary between 40000 us - 50000 us. I got 6 vCPUs 
> allocated to my kvm guest , host machine has 16 cores in total. I 
> wondered if these are related to guest CPU stealing as I get 0.02% of 
> max cpu steal value according our monitoring graphs.
>
> Host CPUs
> lscpu
> Architecture:          x86_64
> CPU op-mode(s):        32-bit, 64-bit
> Byte Order:            Little Endian
> CPU(s):                64
> On-line CPU(s) list:   0-63
> Thread(s) per core:    2
> Core(s) per socket:    16
> Socket(s):             2
>
>
>
> On Thursday, 20 September 2018, 9:27:24 PM GMT+5:30, Bogdan-Andrei 
> Iancu <bogdan at opensips.org> wrote:
>
>
> Hi Pasan,
>
> That warning (harmless in 99% of the cases) reports the delays between 
> (a) triggering a timer job in timer processes and (b) executing that 
> timer job in a worker process.
>
> If there is no load (worker processes are idle, no busy with anything 
> else), the reported delay may be generated only by the interprocess 
> communication (passing the job from the triggering process to the 
> executing process via internal pipes).
>
> What are the values you typically observe ? maybe the warning is jst 
> about a slow context switching on your server (btw, how many cores ?).
>
> Regards,
> Bogdan-Andrei Iancu
>
> OpenSIPS Founder and Developer
>    http://www.opensips-solutions.com
> OpenSIPS Bootcamp 2018
>    http://opensips.org/training/OpenSIPS_Bootcamp_2018/
> On 09/14/2018 06:16 AM, Pasan Meemaduma via Users wrote:
>> Hi Guys,
>>
>> I'm seeing the following warnings on my opensips 2.3.5 boxes even 
>> when they are idle. (just pinging about 8 asterisk gateways in 
>> drouting module for health check)
>>
>> Sep 14 06:38:23  /usr/sbin/opensips[6373]: 
>> WARNING:core:handle_timer_job: timer job <blcore-expire> has a 40000 
>> us delay in execution
>> Sep 14 08:31:53  /usr/sbin/opensips[6362]: 
>> WARNING:core:handle_timer_job: utimer job <tm-utimer> has a 40000 us 
>> delay in execution
>> Sep 14 12:15:38  /usr/sbin/opensips[6365]: 
>> WARNING:core:handle_timer_job: timer job <tm-timer> has a 40000 us 
>> delay in execution
>> Sep 14 12:15:38  /usr/sbin/opensips[6363]: 
>> WARNING:core:handle_timer_job: timer job <tm-timer> has a 40000 us 
>> delay in execution
>>
>> The servers doesn't process anything most of the time. I'm wondering 
>> whats causing these warnings. I tired script_trace and enable debug 
>> level 5 and nothing explains why it complains like this. It just pops 
>> out in the log without any other related log line.
>>
>> /root>> opensips -V
>> version: opensips 2.3.5 (x86_64/linux)
>> flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, 
>> F_MALLOC, FAST_LOCK-ADAPTIVE_WAIT
>> ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, 
>> MAX_URI_SIZE 1024, BUF_SIZE 65535
>> poll method support: poll, epoll_lt, epoll_et, sigio_rt, select.
>> main.c compiled on  with gcc 6.3.0
>>
>> Its a KVM guest and has 6 CPUs allocated.
>> cat /proc/cpuinfo
>> processor       : 0
>> vendor_id       : AuthenticAMD
>> cpu family      : 23
>> model           : 1
>> model name      : AMD EPYC Processor (with IBPB)
>> stepping        : 2
>> microcode       : 0x1000065
>> cpu MHz         : 2095.956
>> cache size      : 512 KB
>> physical id     : 0
>> siblings        : 6
>> core id         : 64
>> cpu cores       : 1
>> apicid          : 0
>> initial apicid  : 0
>> fpu             : yes
>> fpu_exception   : yes
>> cpuid level     : 13
>> wp              : yes
>> bogomips        : 4191.91
>> TLB size        : 1024 4K pages
>> clflush size    : 64
>> cache_alignment : 64
>> address sizes   : 40 bits physical, 48 bits virtual
>> power management:
>>
>> Would be nice to know how I could hunt down this issue.
>>
>>
>> Thanks
>>
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.opensips.org <mailto:Users at lists.opensips.org>
>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20180921/901abe05/attachment-0001.html>