Repeated Lockups in Favorites/Unresponsive Placeshifter

[uggie]

My Placeshifter (6.3.6 server and placeshifter, latest Mac Placeshifter version) connection to the server repeatedly locks up about every 24 hours when trying to manipulate favorites, occasionally to the point that no placeshifter client can connect. Server appears to still be running. The system is being used more heavily now (or at least I'd like to use it more heavily...), but 6.2.10 appeared to be more stable. The log shows traces like:

Thu 1/17 22:51:45.353 java.lang.ArrayIndexOutOfBoundsException: Array index out of range: 0
Thu 1/17 22:51:45.353   at java.util.Vector.remove(Vector.java:777)
Thu 1/17 22:51:45.353   at sage.bk$a.new(Unknown Source)
Thu 1/17 22:51:45.353   at sage.bk$a.a(Unknown Source)
Thu 1/17 22:51:45.353   at sage.bk.O(Unknown Source)
Thu 1/17 22:51:45.354   at sage.bk.if(Unknown Source)
Thu 1/17 22:51:45.354   at sage.bn.try(Unknown Source)
Thu 1/17 22:51:45.354   at sage.bn$1.run(Unknown Source)
Thu 1/17 22:51:56.744 Error in MediaServerConnection of :java.io.EOFException
Thu 1/17 22:51:56.745 java.io.EOFException
Thu 1/17 22:51:56.745   at sage.MediaServer$a.a(Unknown Source)
Thu 1/17 22:51:56.745   at sage.MediaServer$a.run(Unknown Source)
Thu 1/17 22:51:56.745   at java.lang.Thread.run(Thread.java:619)

Also, dmesg had been showing this lately:

[364261.200000] ivtv1: All encoder MPEG stream buffers are full. Dropping data.
[364261.200000] ivtv1: Cause: the application is not reading fast enough.
[364261.216000] ivtv0: All encoder MPEG stream buffers are full. Dropping data.
[364261.216000] ivtv0: Cause: the application is not reading fast enough.

This is on a completely fresh install of Sage using the debian package as of 24 hours ago. Prior to that the lockups could be so severe that the stopsage script would not work and would require a hard kill with the debug log complaining about a broken pipe:

Wed 1/16 19:02:08.671 java.net.SocketException: Broken pipe
Wed 1/16 19:02:08.671   at java.net.SocketOutputStream.socketWrite0(Native Method)
Wed 1/16 19:02:08.671   at java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:92)
Wed 1/16 19:02:08.671   at java.net.SocketOutputStream.write(SocketOutputStream.java:136)
Wed 1/16 19:02:08.672   at java.io.BufferedOutputStream.flushBuffer(BufferedOutputStream.java:65)
Wed 1/16 19:02:08.672   at java.io.BufferedOutputStream.write(BufferedOutputStream.java:78)
Wed 1/16 19:02:08.672   at java.io.DataOutputStream.write(DataOutputStream.java:71)
Wed 1/16 19:02:08.672   at sage.bk.N(Unknown Source)
Wed 1/16 19:02:08.672   at sage.bk.dQ(Unknown Source)
Wed 1/16 19:02:08.672   at sage.bk.do(Unknown Source)
Wed 1/16 19:02:08.672   at sage.bn.try(Unknown Source)
Wed 1/16 19:02:08.672   at sage.bn$1.run(Unknown Source)

Restarts of the server appear to clean things up which makes me suspect a memory leak. Anyone have any thoughts or similar experiences?

thanks,
-jason

[gplasky]

Install jave 1.0.6_4. The latest version.


Gerry