I recently upgraded jemplode to version 43. I can find my Rio over the network, but when i try to connect to it, it freezes after the connection process. It goes through all the steps of the process and stops at the end, with playlists being highlighted red. Any ideas?? Is there anyway I can reload my rio so its fresh? TIA

Does the problem also happen with Emplode?

If Emplode doesn't have a problem, then you don't need to "reload your Rio so it's fresh".

Emplode connects just fine. Is there a corrupt file maybe?

Probably just a bug in Jemplode. Mike will see this thread soon and maybe ask you for a trace log to locate the problem. Not sure how to get a trace log in recent versions of Jemplode, so you'll have to let him guide you through it.

Ok, great. Thanks. Also whenever it freezes on the retreiving playlists part the CPU goes to 100% for Jemplode.

The 100% CPU usage part is very interesting ... What OS are you using? Is this a totally stock jEmplode install? Did you setup soup playlists or anything with the 43 prereleaes?

If you can, go into .jempegrc in your home directory and add a line:

jempeg.debugLevel=31

somewhere in it. Then go to a command shell and run "java -jar jemplode20.jar > out.log", get it to lockup, then send me the traces.

Actually if you can get it to lock then also hit "Ctrl-Shift-Break" and/or "Ctrl-\" in the command shell and send me that output too (if it doesn't go into the log). That will give me a set of stack dumps.

ms

I'm running Windows XP Professional. I also did set up mirror soup ploaylists. Is that what is messing it up? Is there anyway to completely delete them?

Try killing your .jempegrc file and just see if it makes any difference.

Mike

Where is that located?

Nevermind, I found it. And it still freezes after i deleted it.

Here is the output from the file.

8; 31; 8
java.lang.UnsatisfiedLinkError: no jEmplode in java.library.path
at java.lang.ClassLoader.loadLibrary(Unknown Source)
at java.lang.Runtime.loadLibrary0(Unknown Source)
at java.lang.System.loadLibrary(Unknown Source)
at org.jempeg.empeg.protocol.Win32USBEmpegDiscoverer.<clinit>(Win32USBEmpegDiscoverer.java:19)
at org.jempeg.empeg.protocol.USBEmpegDiscovererFactory.createUSBDiscoverer(USBEmpegDiscovererFactory.java:18)
at org.jempeg.empeg.emplode.EmpegDiscoveryManager.refresh(EmpegDiscoveryManager.java:135)
at org.jempeg.empeg.emplode.dialog.ConnectionSelectionDialog.refresh(ConnectionSelectionDialog.java:233)
at org.jempeg.empeg.emplode.dialog.ConnectionSelectionDialog$RefreshRunnable.run(ConnectionSelectionDialog.java:303)
at java.lang.Thread.run(Unknown Source)
1036378914005: Broadcasting on 255.255.255.255...
1036378914005: Broadcasting on 192.255.255.255...
1036378914005: Broadcasting on 192.168.255.255...
1036378914015: Broadcasting on 192.168.1.255...
1036378914015: Found Empeg = Nempeg
1036378914045: SocketConnection (17689439: actually open
1036378914395: SocketConnection (17689439): close
1036378920374: SocketConnection (17689439: actually open
1036378920764: SocketConnection (17689439): close
1036378920844: SocketConnection (17689439: actually open
1036378924610: FIDPlaylistDatabase contains 2075 entries
1036378924630: FIDPlaylistDatabase maxFID = 39008
1036378924720: Ended up with 2075 of 2075
1036378925701: SocketConnection (17689439): close
1036378925711: Checking root playlist...
1036378925711: Starting tree->Repair() recursion...

I also couldn't get it to output the stack dump. I tried doign the commands but nothing happened.

The same thing happened to me yesterday.. JE 43 and 42 hung, 41 and Emplode (2.0b13) worked fine. After a whole bunch of trying to figure out what the problem was, I found I could close the progress window to get to the main JE window.. the All Music item was red because it was trying to fix something. Somehow I got it open (maybe by trying at the right time, I couldn't reproduce it again) and found that it had moved a folder there (presumably because it had no parent, or maybe I created a circular ref). It was named "Dire Straits".

So I went back to Emplode and searched for "Dire Straits".. it showed two results (the real folder, and an evil broken one), then hung before drawing the value for the length column. So I started again, turned off all the columns except Name and FID, searched again, and then I was able to delete that bad node and synch. All was well from there.

Sooo.. I think a bug in JEmplode 43 caused the problem somehow, but I unfortunately can't specifically tell you what.

44 will probably go up tonight .. I've added a little more debug info to the database download method (if it is in fact locking there). Grab that and try again with the log output. So for the stack traces, you tried Ctrl-Shift-Break in the console window and it didn't do anything? I can't remember what version they added Ctrl-\ in... Oh well. Thanks for trying.

I got it to produce the crash dumps. I attached the output. Hope it helps.

I'm guessing there's an infinite loop in here (because it doesn't appear to be locked up), which would be consistent with the 100% CPU utilization. I might have you grab a new jemplode20.jar that has more debug statements enabled in it to try and track down the infinite loop.... If you happen to have Emplode around, I'd be curious to have you open it up and see if it complains about a loop in one of your playlists...

Whats weird is that I deleted all of my playlists and all of my songs and transfered them back and i still have the problem. Emplode doesn't seem to find the problem although some times it does. Wouldn't removing all of my playlist and songs get rid of the infinite loop? Actually I just tried again and here is wht happens. I synched on Emplode and it told me there was an error and to synchronize to fix it. I did that. Then I ran Jemplode-36 and it found an error again and said to synchronize to fix. I did that and then I ran Emplode again. It found an error again. Is there somethig on the drive that is messing up?

One of those jEmplode versions in the 30's had a problem where Emplode and jEmplode had disagreements about when a particular problem was "fixed". I would try the latest version. Typically if Emplode can't fix something, there's something sinister going on (since it tends to be pretty awesome about resolving weird problems).

RIght. So if the my database is pretty screwed is there anyway I can format the drive to fix it? Or would that even fix it? Does Prerelease 44 have the debugging info you need? Ill download that and do the same thing.

44 probably doesn't have that debug data (primarily because the part where i need debug data would produce a ton of output and would slow down the production release)... definitely give 44 a try and see how things go... if you get an error and it says it fixed it, make sure you sync once to sync the fixes to your Empeg. the second time you run it after a sync, you aren't supposed to have any errors (same with Emplode, actually).

as far as formatting, you can check the faq -- i'm sure it talks about it.

primarily because the part where i need debug data would produce a ton of output and would slow down the production release

Well, why don't you just use Java's preprocessor? Oh, that's right. Because Sun, in their infinite wisdom, has decreed that you don't need one. I forgot. (Idiots.)