• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 98 subscribers
  • Views 793 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DM365 IPNC system_server busy and system crash

Chung-Chi Tsai
Prodigy 65 points

 

Dear All:

We encounter a big problem about system_server rise its cpu loading up unexpectedly. 

  Our IPNC version is V2.0 and settings  : 

    

Stream

audio

Alarm (aplay)

Alarm duration

Alarm trigger

Storage

3

ON

ON

Non STOP

Motion or external Input

Not necessary

And after about 14 hours, the aplay fucntion will stop and disappear, even though  we re-enable the alarm.

                We use "top" cmd to monitor the IPCAM status and surprisingly find out system server rise its cpu loading up to about 30%.

But we still can type #aplay -f U8 alarm_1_8K.wav -r 8000 and it could work.

And we list below results for system_server and the other services.

>> system_server around 10%        => normal status

>> system_server up to about 20% => aplay and storage functions don't work.

                >> system_server up to about 50%  => boa server and Upnp failed to use. but  VLC could still receive the streaming.

                >> system_server large than 90%   => system crash....

                However,  we change our IP environment to Static IP mode, the system and services are health during 3-4 day long-term testing.

And the result is same as the Appro IPCAM in version 2.0.

 

Hence,  had anyone met the same problem? Or any idea to solve the strange phenomenon? 

I really appreciate and thanks for your kindly help.

 

 

 


  • 0
    TI__Genius 11000 points

    Hi,

     

    We had fixed similar issue in latest version 2.6 of the IPNC camera, can you please check on this.

     

    We had Motion detect ON load going high issue on 2.0 and also, similar crash.

     

    PLease refer to v2.6 from Appro site for updates and let us know the feedback on if it has solved the issue.

     

    Also, there was memory leak which we had fixed for long duration tests

     

    Regards,

     

    Raghu

     

  • 0 in reply to Raghu Kudva
    Prodigy 65 points

    Hi, Raghu

                    Thanks for your kindly reply. 

    We also test the same setup in V2.6, but the result isn't fully pass.

    Although the loading of the system server seems to be normal, the aplay function doesn't work

                    anymore.  

    (if you type # aplay -f U8 XXXXXXX, then the system shows that "device busy" ....)

    By the way, we have changed the structure of motion detection (videoMotionDetection.c)

                    reference to V2.6 so as to avoid the busy messages passing and mounts of threads

                    generated.

     

    All I want to know is how to solve the issue in V2.0, since we will enter the next stage for

                    our IPCAM.

    Could you provide some suggestions of the issue, and we will appreciate what you do. 

    And I will also do memory leak checking for verify the system stability.

     

    Betters

  • 0 in reply to Chung-Chi Tsai
    TI__Genius 11000 points

    HI,

    In v2.6 you need to pass message, by defualt audio playback is ON to support two way audio.

     

    In ALALRM page, there is "alarm on audio" button, can you please raise that to check with ALARM if playback works fine?

     

    Regards,

     

    Raghu

     

  • 0 in reply to Raghu Kudva
    Prodigy 65 points

    Hi, Raghu

    About the suggestion, it works fine with alarm.

    And we will still track the problem.

    I list some  phenomenons :

    (1) aplay function doesn't work.

    (2) storage function doesn't work.

    (3) with increasing message , # of ProcAlarmThread increase and do not terminate.

    It is easy copy the phenomenon in Appro V2.0.

     

    Sincerely

    Betters

     

  • 0 in reply to Chung-Chi Tsai
    Prodigy 65 points

    Additionally, 

    it seems that halt in the system cmd "# aplay -f U8 XXXXXX "of  AlarmAudio_Play().

    But the same setup in static ip mode, it is normal.

    However, I cannot find the relationship for why DHCP does.

    In our design, we lighten the busy messages passing and lock until almost finish

    or terminate.

    Hence, we just focus on the ProcAlarmThread and AlarmAudio_Play() to find the

    root cause.

    Does anyone met the same problem ? Or any idea to solve.

    Sincerely 

    Betters

     

  • 0 in reply to Chung-Chi Tsai
    Prodigy 65 points

    Hi,

    I found the new problem in V2.6 yesterday. When I setup the general setting with

    motion alarm and alarm aplay, I also do shell #aplay -f U8 XXXXXXX.

    And then the AV server will be crashed unexpectedly. But the system server is still stable.

     

    About the type IP (static IP or DHCP), DHCP  caused system server busy easier than

    Static IP mode. After we long-term test duration a week, the Static IP mode also cause

    tha same problem in V2.0.  (In V2.6 is health.)

     

      Hence, the stability of system_server in IPNC 2.0 doesn't be solved. 

    And the stability of system_Server in IPNC 2.6 is ok, but there are some ways to

    make the system unstable.

     

                    Sincerely

    Betters