Re: [JSch-users] Infinite loop/hang on connect

[Adam F. <afa...@in...>]

Thank you for the reply Ned. That isn't the problem in this case. I
already use an Interrupter thread just as you described. It interrupts
one Thread.sleep(10) but is caught within jsch and continues the loop.

--
Adam Fanello

Ned Miles babbled on 02/27/2006 07:51 PM thusly:

> I had a similar problem with the sftp connection that would occur very
> occassionally.  The workaround I implemented was certainly not an
> eloquent one but it has worked.  Basically, I created a very simple
> class that would start a timer when the connection call was made.  If
> the call had not returned within some liberal amount of time (say 1
> minute) I interupted the thread and retried the operation.   Like I
> said, not real pretty but it has done the trick.
>
> -----Original Message-----
> From: Adam Fanello
> To: jsc...@li...
> Sent: 2/27/2006 10:00 PM
> Subject: [JSch-users] Infinite loop/hang on connect
>
> Hello fellow jsch-users (and author).
>
> I've been using jsch for some time, now at version 0.1.24, and am now
> having an occasional hang problem. This software is running on remote
> servers... I can't have it hanging! Here's the thread's stack trace
> where it gets stuck:
>
> "CommAgent" prio=1 tid=0x4de2b3f8 nid=0x8fa waiting on condition
> [0xbddff000..0xbddff65c]
>         at java.lang.Thread.sleep(Native Method)
>         at com.jcraft.jsch.Session.write(Unknown Source)
>         at com.jcraft.jsch.Session.connect(Unknown Source)
>         at
> com.intelio.common.io.SSHConnection.connect(SSHConnection.java:84)
>         at
> com.intelio.common.comm.PayloadTransmitter.sendPackages(PayloadTransmitt
> er.java:253)
>         at
> com.intelio.common.comm.PayloadTransmitter.run(PayloadTransmitter.java:1
> 13)
>         at
> com.intelio.edge.agent.comm.CommAgent.execute(CommAgent.java:164)
>         at com.intelio.edge.component.Component.run(Component.java:92)
>
> Apparently the variable "in_kex" never becomes true. I'm not sure about
> the interaction here, but apparently its waiting for another thread to
> change the value. Without a synchronization or declaring the variable as
> volatile, could it be that the value is in fact changing and it just
> isn't noticing? It seems to me that the CPU cache would flush
> eventually, while this is hanging for hours (until I receive a notice
> that files are piling up because the transfer is stuck, and I kill it).
>
> Anyway, I do think that the variable needs to be declared volatile, but
> it probably would only solve a short hang, not an infinite one. Any
> other ideas?
>
> -- 
> Adam Fanello
> Sr. Engineer/Architect
> Intelio Technologies, Inc
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting
> language
> that extends applications into web and mobile media. Attend the live
> webcast
> and join the prime developer group breaking into this new coding
> territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642>
>
> _______________________________________________
> JSch-users mailing list
> JSc...@li...
> https://lists.sourceforge.net/lists/listinfo/jsch-users
>