Hello,
We are pushing our RTMP streams to the MirrorImage CDN using push publishing profiles. When we test a situation where a user is publishing a stream and the connection drops and then they reconnect we find that the stream republishes to Wowza just fine. However if the reconnect happens in less than 20 seconds it does not properly re-push to the CDN. In which case our end users consuming from the CDN do not get the stream but the publisher thinks they are broadcasting just fine. Generally speaking, if we force the reconnect to take 30 seconds it works just fine.
Does anyone know why this may be happening and/or have suggestions on how to mitigate this issue and bring the reconnect time down < 10 seconds or so?
Thanks!
(we’re on Wowza 4.1.2)
Hi,
It sounds like the stream could still be active on the CDN side which means when you publish it again, it fails, possibly due to the stream already existing. If this is the case, I’m not sure if there is anything that can be done one the Wowza side as this would require the stream to be destroyed on the CDN side. To troubleshoot further I recommend creating a ticket so we can take a closer look as I can only guess without the configuration and log files.
To create a ticket, send an email to support@wowza.com. Ensure that you include a ZIP file containing the [Wowza-Install]/conf/ and [Wowza-Install]/logs/ directories for analysis.
Please make sure you provide logs which show Wowza server starting.
If you are not sure how to get this information please see the following tutorial.
How to create a compressed zip file in Windows, OS X, and Linux
Regards,
Jason
Just an idea in case you haven’t solved the problem. It may be the case that the CDN caches the error response while the content is not available and serves that to customer even though the stream is active again. If the negative cache is 30 seconds, this could explain the delay you are seeing.