Project

General

Profile

Support #21295

output files not recovered

Added by Jose Soto 5 months ago. Updated 6 days ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Target version:
-
Start date:
11/02/2018
Due date:
% Done:

0%

Estimated time:
Duration:

Description

Hello,
I'm having problems when using the fife_launch feature. The script is run correctly, but at the end of the job the output files are not recovered.

This is the message that appears at the end of the error file:

129+1 records in
129+1 records out
67981168 bytes (68 MB) copied, 0.874525 s, 77.7 MB/s
Traceback (most recent call last):
  File "./fife_wrap", line 1176, in <module>
    w.copy_back()
  File "./fife_wrap", line 681, in copy_back
    self.copy_back_n(i)
  File "./fife_wrap", line 754, in copy_back_n
    rt = self.file_stats(f)
  File "./fife_wrap", line 629, in file_stats
    checksum = '["enstore:%s"]' % json.loads(checksum)["crc_value"]
TypeError: list indices must be integers, not str
Fri Nov 2 16:20:21 CET 2018 fife_wrap COMPLETED with exit status 1

The script I used to run it is in /dune/app/users/jsoto/fcl/SAMPLES/SN_scan_light_PL
And this is a test job showing the problem:

Any idea of what is happening?

Thanks in advance,

History

#1 Updated by Marc Mengel 5 months ago

Sounds like you're using the latest ifdhc (v2_3_8) but not the latest fife_utils (v3_2_4) to go with it -- the latest ifdhc reports checksums differently.

So setup fife_utils v3_2_4 wherever you're setting it up here, and I think you'll be fine.

#2 Updated by Jose Soto 5 months ago

You are right, I was using fife_utils v3_2_3. I set up v3_2_4, and now it works.
Thanks!

#3 Updated by Marc Mengel 6 days ago

  • Status changed from New to Closed


Also available in: Atom PDF