MJPEG transmission breaks every minute

[JohannesL]

Hi,

I have a problem with MJPEG transmission. The setup what I'm having is that I have a camera connected over WiFi to a Mac Mini, which is running VitaminD. For some reason, the transmission throughput goes to zero every minute. Like a clock. For a couple of seconds. I don't understand why. I don't know if the problem is in the uplink or in downlink. The camera generates about 7Mbps to the WiFi network. There is not any other devices within the same network, so there should be enough capacity within the WiFi. Also, in the Mac Mini, VitaminD is the only SW which is running, there is plenty of processing power available.

Any idea where the problem is? I cannot find anything from the log files, that would make sense to me. The WiFi router what I'm having is 5th generation Airport Extreme. Camera is Brickcom WOB100-Ap, which should be a high quality camera. When I connect the camera directly with my browser, using the same URL, everything works fine. The stream is not -frozen- every minute or so.

Thank's for your support
Regards
Johannes

[ryan]

Hi Johannes,

What camera model are you using?

[JohannesL]

Hi Johannes,

What camera model are you using?

Hi,

I'm using Brickcom WOB-100Ap camera. There are two cameras connected to the same WiFi network. Hard to believe that the failure is because of the cameras, because exactly at the same moment the throughput is getting down from both of those. At the same time, there are these error messages in the Mac OS X lion (10.7.3 version):

Thu Feb 2 02:07:08 <kernel> IO80211ScanManager::startScan: Initiating scan.
Thu Feb 2 02:07:08 <kernel> IO80211ScanManager::startScan: Broadcast scan request received from pid 154 ().
Thu Feb 2 02:07:01 <kernel> IO80211ScanManager::scanDone: Scheduling cache purge timer in 30 seconds.
Thu Feb 2 02:06:25 <kernel> IO80211ScanManager::cachePurge: NOT Rescheduling.
Thu Feb 2 02:06:25 <kernel> IO80211ScanManager::cachePurge: clearEntries[0] Firing!
Thu Feb 2 02:06:08 <airportd[331]> _doAutoJoin: Already associated to “Net 5 GHz”. Bailing on auto-join.


Regards
Johannes

[ryan]

Hi JohannesL,

That kernel error you posted is unrelated to Vitamin D but does seem to indicate your wifi connection is being dropped. Googling it turned up a lot of very lengthy threads such as https://discussions.apple.com/thread/31 ... 0&tstart=0 which are full of suggestions. Hopefully something there or in a similar thread can help you.

Your browser is likely more aggressive about refreshing/retrying. VDV has a somewhat heavyweight bringup/teardown process for dealing with cameras, as well as a generous buffer for cameras who temporarily freeze or are dropping packets. This is why VDV might appear to 'hang' a bit when the connection is lost and why it takes a bit to reconnect.