Upload fails, webapp and job runner running on different machines
Hi everyone, I'm currently trying a new galaxy install with the webapp and the job runner running on different machines, sharing a nfs volume where the galaxy files are, and another one where the data is supposed to be stored, as specified in the file_path and new_file_path values in the universe_wsgi files. (one for the webapp and one for the runner, as stated in the doc) Both instances run properly with no error message but when I want to upload a file from wherever in a library, here's what happens: The file's row appears in the library with the "uploading" note. The job is registered in the database. And that's all. Nothing changes afterwards. Nothing is saved anywhere. This can't be a file-too-big issue since the test file I'm working with weighs 3.5kB. I probably stupidly missed something but I really don't see where the problem might be. Any idea? Cheers L-A
Hello Louise, Sorry about the delay, but significant time has passed since you sent the message below, so I wanted to check with you to confirm this is still an issue for you. Can you let me know? Thanks, Greg Von Kuster On Mar 23, 2011, at 12:09 PM, Louise-Amélie Schmitt wrote:
Hi everyone,
I'm currently trying a new galaxy install with the webapp and the job runner running on different machines, sharing a nfs volume where the galaxy files are, and another one where the data is supposed to be stored, as specified in the file_path and new_file_path values in the universe_wsgi files. (one for the webapp and one for the runner, as stated in the doc)
Both instances run properly with no error message but when I want to upload a file from wherever in a library, here's what happens: The file's row appears in the library with the "uploading" note. The job is registered in the database. And that's all. Nothing changes afterwards. Nothing is saved anywhere.
This can't be a file-too-big issue since the test file I'm working with weighs 3.5kB.
I probably stupidly missed something but I really don't see where the problem might be.
Any idea?
Cheers L-A
___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at:
Greg Von Kuster Galaxy Development Team greg@bx.psu.edu
Hello, and sorry for not updating This is not an issue anymore. If I remember correcly, it was due to the temporary files not respecting the new_file_path config values. We cheated a bit hardcoding the temp file path in the code but that was corrected recently, and we're now using a recent checkout so it works fine. Thanks for asking :) L-A Greg Von Kuster wrote:
Hello Louise,
Sorry about the delay, but significant time has passed since you sent the message below, so I wanted to check with you to confirm this is still an issue for you. Can you let me know?
Thanks,
Greg Von Kuster
On Mar 23, 2011, at 12:09 PM, Louise-Amélie Schmitt wrote:
Hi everyone,
I'm currently trying a new galaxy install with the webapp and the job runner running on different machines, sharing a nfs volume where the galaxy files are, and another one where the data is supposed to be stored, as specified in the file_path and new_file_path values in the universe_wsgi files. (one for the webapp and one for the runner, as stated in the doc)
Both instances run properly with no error message but when I want to upload a file from wherever in a library, here's what happens: The file's row appears in the library with the "uploading" note. The job is registered in the database. And that's all. Nothing changes afterwards. Nothing is saved anywhere.
This can't be a file-too-big issue since the test file I'm working with weighs 3.5kB.
I probably stupidly missed something but I really don't see where the problem might be.
Any idea?
Cheers L-A
___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at:
Greg Von Kuster Galaxy Development Team greg@bx.psu.edu
Hello, When adding a new tool, we used a cheetah'ed shell script defined in the <configfiles> tags of the tool's xml file. To get the absolute path of an external script located in the tools directory, I used the following code in the <configfile> script: ${ os.getcwd() }/${ $__app__.config.tool_path }/toolsubdir/scriptname.sh When I tried the new tool, I got an error message. Here are the last lines: File "/g/funcgen/galaxy/lib/galaxy/tools/parameters/basic.py", line 681, in value_from_basic if isinstance( value, dict ) and value["__class__"] == "UnvalidatedValue": KeyError: '__class__' Therefore I slightly modified the file basic.py. I replaced the line 681: if isinstance( value, dict ) and value["__class__"] == "UnvalidatedValue": by: if isinstance( value, dict ) and getattr(value, "__class__") == "UnvalidatedValue": and it works fine now. I fail to see why the original code didn't work though. Hope that helps. Regards, L-A
participants (3)
-
Greg Von Kuster
-
Louise-Amelie Schmitt
-
Louise-Amélie Schmitt