Hi,
I just discovered that the problem is caused by the Firewall. But the both
container logs (server and client) don't say anything about it. I think
this is because firewall allows only one connection to be opened (transfer
control channel, but not the actual transfer channel). The message
exchange goes fine but the actual transfer doesn't happen. And that's why
the logs don't say anything.
When the firewall is disabled everything is working fine.
Thank you very much for your efforts.
Regards.
--
Fatih TURKMEN
dit.unitn.it/~turkmen
> That is a really weird error. It makes me think that your gridftp
servers are not working well.. Can you do the following changes and
restart the globus container and send me the new container logs :
>
> 8.2. More verbose error messages
>
> Problem: Make RFT print more verbose error messages
>
> Solution: Edit $GLOBUS_LOCATION/container-log4j.properties and add the
following line to it: log4j.category.org.globus.transfer=DEBUG. For more
verbosity addlog4j.category.org.globus.ftp=DEBUG, which will print out
Gridftp messages too.
>
>
>
> Regards
>
> On Feb 26, 2008, at 9:39 AM, Fatih Turkmen wrote:
>
>> Hi,
>> I attached the container logs (both server and client). If you don't mind,
>> I didn't CC this email to the list because of security concerns (the
content of the logs). I guess the server is working properly. As we are
>> doing our tests in wide area, I suspect from authorization issues. If
necessary, I can send the content of grid-mapfile too.
>> Thanks in advance.
>> --
>> Fatih TURKMEN
>> dit.unitn.it/~turkmen
>>> Hi
>>> Can you get a hold of the container log files and send it out ? I
wonder if you have a gridftp server setup and listening on port 2811
on HOST..
>>> On Feb 25, 2008, at 6:35 PM, Fatih Turkmen wrote:
>>>> Hi all,
>>>> I have a problem globus ws job submission. There is a
>>>> really short RSL file with some staging statements
>>>> (below).
>>>> <job>
>>>> <executable>my_echo</executable>
>>>> <directory>${GLOBUS_USER_HOME}</directory>
>>>> <argument>Hello</argument>
>>>> <argument>World!</argument>
>>>> <stdout>${GLOBUS_USER_HOME}/stdout</stdout>
>>>> <stderr>${GLOBUS_USER_HOME}/stderr</stderr>
>>>> <fileStageIn>
>>>> <transfer>
>>>> <sourceUrl>gsiftp://HOST1:2811/home/fatih/aaa.txt
>>>> <destinationUrl>file:///${GLOBUS_USER_HOME}/aaa_copied</
>>>> destinationUrl>
>>>> </transfer>
>>>> <transfer>
>>>> <sourceUrl>gsiftp://HOST1:2811/home/fatih/echo
>>>> <destinationUrl>file:///${GLOBUS_USER_HOME}/my_echo</destinationUrl>
>>>> </transfer>
>>>> </fileStageIn>
>>>> </job>
>>>> Do you have any ideas what goes wrong from the debug
>>>> output? I keep getting :
>>>> globusrun-ws: Job failed: Staging error for RSL
>>>> element fileStageIn.
>>>> Transient transfer error
>>>> Reply wait timeout. (error code 4) [Caused by: Reply
>>>> wait timeout. (error code 4)]
>>>> Transient transfer error
>>>> Reply wait timeout. (error code 4) [Caused by: Reply
>>>> wait timeout. (error code 4)]
>>>> Thanks in advance.
>>>> Ps : I also attached the debug output of the
>>>> globusrun-ws. I modified the host name in RSL but not
>>>> in the debug file.
>>>> Fatih
>>>> <debug.txt>
>>> --
>>> Ravi K Madduri
>>> The Globus Alliance | Argonne National Laboratory | University of Chicago
>>> http://www-unix.mcs.anl.gov/~madduri
>> <server_container_log.txt><client_container_log.txt>
>
> --
> Ravi K Madduri
> The Globus Alliance | Argonne National Laboratory | University of Chicago
> http://www-unix.mcs.anl.gov/~madduri
>
>
