I've done some googling, the best I could find was this: http://stackoverflow.com/questions/4654480/android-httpclient-perfomance - according to this, it's an Android limitation. It would be nice to have some official confirmation, though - perhaps Jon could chime in?
I'm also not sure how it is decided what is the same host... if it's the same fqdn, or if it's IP based (if it's the former, in my case I could work around it by using the ip address in the long poll and have two more requests for user initiated stuff). Then I found this: http://developer.android.com/reference/org/apache/http/conn/params/ConnPerRouteBean.html which seems to suggest it should be possible to change the value - though I'm not at all familiar with java interop at this point. If you do go down that route and find it working, I'd be much obliged if you could share the solution. For my part, I've started (already had some in place) to be extremely strict in enforcing that only one request runs on top of my long poll, and I diligently abort my long poll requests if I see a network issue reported anywhere (from connection manager or my "is server reachable" process) in the hope that this could bring an end to my "happens every now and then but I don't have a method to reliably reproduce" issues where every new http request I make reports a timeout even though the server is perfectly reachable. Stephan I can confirm this... max 2 concurrent connections is allowed. If another one is done, it's blocked until one of the previous is released. I didn't notice this before because I didn't have (previously) any situation where more than 2 connections were made. Is this an Android limitation or a bug in MfA? On 30 September 2012 21:28, Stephan Steiner <stephan.steiner@>wrote: -- View this message in context: http://mono-for-android.1047100.n5.nabble.com/Push-notifications-for-mono-tp5711887p5712035.html Sent from the Mono for Android mailing list archive at Nabble.com. _______________________________________________ Monodroid mailing list [email protected] UNSUBSCRIBE INFORMATION: http://lists.ximian.com/mailman/listinfo/monodroid
