Is anybody out there using CIFS on a recent nightly build?

[sacco]

For me it fails completely on every nightly build from the last week, just like this:

http://forums.alfresco.com/viewtopic.php?t=11185

I can't see any problem with my configuration that could cause this kind of
effect; more to the point, the same configuration (modulo the necessary
changes to the class names of Desktop actions)) works without problems with
2.9.0B Labs.

[mikeh]

This may help for now: http://forums.alfresco.com/viewtopic.php?t=10523

Thanks
Mike

[sacco]

Thanks Mike.

You're working very late if you're in the UK!

Unfortunately, though, I'm trying to get the nightly running on RHEL, not
Windows, so I don't think the .DLL can be the problem.
The  system comes up, and I can see some files through CIFS,
but Windows Explorer locks up (a blocking call I guess, as it can't even be
killed until the timeout has expired) and the log fills up with the error reported under
my link (above).

I'm also swapping the Desktop Action class-names, and my CIFS Server host
name is manually set in any case.

Does this want to go in the JIRA?  (I was assuming that it was just a glitch
due to moving the classes around.)

[sacco]

Incidentally, the error:

java.lang.IllegalArgumentException: Invalid NetworkFile class
        at org.alfresco.jlan.server.filesys.cache.FileStateLockManager.lockFile(FileStateLockManager.java:43)
        at org.alfresco.jlan.smb.server.NTProtocolHandler.procLockingAndX(NTProtocolHandler.java:1955)
        at org.alfresco.jlan.smb.server.NTProtocolHandler.runProtocol(NTProtocolHandler.java:284)
        at org.alfresco.jlan.smb.server.SMBSrvSession.runHandler(SMBSrvSession.java:1377)
        at org.alfresco.jlan.smb.server.SMBSrvSession.run(SMBSrvSession.java:1298)
        at java.lang.Thread.run(Thread.java:801)
‍‍‍‍‍‍‍‍

seems to correlate with this on shutdown:

INFO: Pausing Coyote HTTP/1.1 on http-8080
28-Feb-2008 22:40:03 org.apache.catalina.core.StandardService stop
INFO: Stopping service Catalina
28-Feb-2008 22:40:03 org.apache.catalina.core.StandardWrapper unload
INFO: Waiting for 1 instance(s) to be deallocated
28-Feb-2008 22:40:04 org.apache.catalina.core.StandardWrapper unload
INFO: Waiting for 1 instance(s) to be deallocated
28-Feb-2008 22:40:05 org.apache.catalina.core.StandardWrapper unload
INFO: Waiting for 1 instance(s) to be deallocated
28-Feb-2008 22:40:05 org.apache.catalina.core.StandardWrapper unload
INFO: Waiting for 13 instance(s) to be deallocated
28-Feb-2008 22:40:06 org.apache.catalina.core.StandardWrapper unload
INFO: Waiting for 13 instance(s) to be deallocated
28-Feb-2008 22:40:07 org.apache.catalina.core.StandardWrapper unload
INFO: Waiting for 13 instance(s) to be deallocated
‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

I'd investigate my configuration further, but none of this happens with the
2.9.0B Labs build…

Thanks.

[mikeh]

Does this want to go in the JIRA?  (I was assuming that it was just a glitch
due to moving the classes around.)

I think it should - in case there is a problem lurking in the recent JLAN updates. It will at least bring it to the attention of the developer, who is in a better position to say if it's config or code causing your problem.

Thanks,
Mike

[dinger]

Did you raise a JIRA for this?

Rob

[dinger]

I raised a JIRA for this: http://issues.alfresco.com/browse/JLAN-13

Rob

[dinger]

For anyone who is seeing this it's going to hopefully fixed soon:

http://issues.alfresco.com/browse/AR-2231

Rob

[sacco]

Did you raise a JIRA for this?

Rob

To my shame, I'm afraid I got side-tracked and forgot, and since then family stuff
has meant I've only been paddling just enough to stay afloat, work-wise… ops:

Sorry, folks!  Well done Rob for spotting and fixing the problem!

[sacco]

Do you know if the good news on the JIRA means that this will appear
in HEAD / nightly builds soon?