You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@maven.apache.org by "?rjan Austvold (JIRA)" <ji...@codehaus.org> on 2006/01/06 21:37:01 UTC

[jira] Created: (MNG-1938) site:deploy hangs forever on remote upack command

site:deploy hangs forever on remote upack command
-------------------------------------------------

         Key: MNG-1938
         URL: http://jira.codehaus.org/browse/MNG-1938
     Project: Maven 2
        Type: Bug

    Versions: 2.0.1    
 Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
    Reporter: Ørjan Austvold
 Attachments: thread-dump.txt

site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:

Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip

Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.

I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).

Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "Brett Porter (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_56836 ] 

Brett Porter commented on MNG-1938:
-----------------------------------

Does anyone have a reproducible case with Maven 2.0.2? As I said, it's fine for me.

> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "?rjan Austvold (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_55120 ] 

Ørjan Austvold commented on MNG-1938:
-------------------------------------

Thank you for the snapshot build, but the behaviour didn't change. AbstractSshWagon still hangs attempting to get input from the error stream of the ChannelExec.

To me the code in executeCommand seems to only return successfully if the error stream of the ChannelExec returns null on readLine. Why so?

I'll check out head on maven and attach a debugger to the build I got from you. Maybe I can understand some more that way.

> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "?rjan Austvold (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_55132 ] 

Ørjan Austvold commented on MNG-1938:
-------------------------------------

I don't know if this would help, but we've seen this problem on large sites since early stages of maven 2.0 The problem has consistently been appering and has always been reproducible.

> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "?rjan Austvold (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_56868 ] 

Ørjan Austvold commented on MNG-1938:
-------------------------------------

It is totally reproducible on a project we have have at work.

The failure is occuring each time when I  upload the site to a CentOS machine. I even tried to upload to localhost by setting the url in distributionManagement to scp://localhost/tmp/foo -- and the same behaviour appears.

Do you have any comment on why the AbstractSshWagon tests for null on the error stream of the ChannelExec to detect succesful completion? As I stated earlier the author of Jsch seems to test for channel.isClosed() to detect completion.

> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "Brett Porter (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_55136 ] 

Brett Porter commented on MNG-1938:
-----------------------------------

are you sure you have the latest code?

I was just able to upload a 23mb site without issues.

The fixes done on scp were predominantly in the executeCommand method.

> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "Brett Porter (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_55127 ] 

Brett Porter commented on MNG-1938:
-----------------------------------

We've seen this before: MNG-476. It was when there was output on the server. However, I was pretty sure this new code dealt with that. Might require some more investigation.



> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Closed: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "Mike Perham (JIRA)" <ji...@codehaus.org>.
     [ http://jira.codehaus.org/browse/MNG-1938?page=all ]
     
Mike Perham closed MNG-1938:
----------------------------

    Resolution: Duplicate

WAGONSSH-36 has the fix.  It just needs to be applied as I don't have commit rights to wagon.

> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "?rjan Austvold (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_55126 ] 

Ørjan Austvold commented on MNG-1938:
-------------------------------------

I've been debugging the AbstractSshWagon and jsch but I cannot find out why the inputstream doesn't return null in the executeCommand method.

Other commands (like mkdir (actually executed twice I found out... :-) )) returns successfully.

Could it be that jsch doesn't handle multiple commands on the shell (cd + unzip + rm)? This fact and that lots of bytes have been transferred during copy is the only difference I can see that could make the command fail.

I'll check with the jsch author about how to go about checking for termination (in error or successful).

> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "Adrian (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_56748 ] 

Adrian commented on MNG-1938:
-----------------------------

This bug is currently blocking our continuous integration cycle. We run nightly builds that update our website documentation for developers. With this bug this fails and developers are flooded with emails from continuum telling them the build failed. In actual fact the code compiled and was tested and the site deploy is what failed.

This is very important to us.

Thanks

> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "?rjan Austvold (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_55121 ] 

Ørjan Austvold commented on MNG-1938:
-------------------------------------

Continuing on the termination condition thing.

The author of jsch seems to test for channel.isClosed() after a channel.connect() on a ChannelExec. See http://www.jcraft.com/jsch/examples/Exec.java

I'll try to patch the AbstractSshWagon and see if this makes a difference.


> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org


[jira] Commented: (MNG-1938) site:deploy hangs forever on remote upack command

Posted by "Dan Tran (JIRA)" <ji...@codehaus.org>.
    [ http://jira.codehaus.org/browse/MNG-1938?page=comments#action_55119 ] 

Dan Tran commented on MNG-1938:
-------------------------------

would problem go away if you use this recommened 2.0.2 snapshot build?

http://maven.zones.apache.org/~maven/builds/branches/maven-2.0.x/m2-20060105.071501.tar.gz 



> site:deploy hangs forever on remote upack command
> -------------------------------------------------
>
>          Key: MNG-1938
>          URL: http://jira.codehaus.org/browse/MNG-1938
>      Project: Maven 2
>         Type: Bug

>     Versions: 2.0.1
>  Environment: maven 2.0.1 with wagon and jsch fixes as given by Brett Porter in http://www.nabble.com/call-for-testers:-scp-wagon-t854948.html.
> Site runs on Linux (CentOS 4.2). Both scp and sftp protocol produces the same error
>     Reporter: Ørjan Austvold
>  Attachments: thread-dump.txt
>
>
> site:deploy work fine for small artifact sites (<1MB?) but hangs forever on the unpack command:
> Executing command: cd /var/foo/.; unzip -o wagon17515.zip; rm -f wagon17515.zip
> Looking on the /var/foo directory on the server I can see that the site has been sucessfully unpacked and the zip file has been removed.
> I did a kill -3 on the vm to get a thread dump of what was going on and found that there where 3 threads waiting on a jsch inputstreams (see the attached thread dump). The main thread seems to be waiting for input on the executeCommand method of the AbstractSshWagon (line 292).
> Could it be that some of the fixes made on the wagon for repository deployment/downloading should be done for executeCommand to?

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org