for some of the LIGO workflows we are seeing stages jobs repeat parts of deep LFN's, resulting in the jobs failing.

For example for LFN - 113523/L-L1_HOFT_C02-1135235072-4096.gwf

 dags/work/main_ID0000001.000/00/00/stage_in_remote_local_0_0.in
[
 { "type": "transfer",
   "id": 1,
   "src_urls": [
     { "site_label": "local", "url": "file:///cvmfs/oasis.opensciencegrid.org/ligo/frames/O1/L1_HOFT_C02/L/1135/L-L1_HOFT_C02-1135235072-4096.gwf", "priority": 100 }
   ],
   "dest_urls": [
     { "site_label": "local", "url": "symlink:///home/steven.reyes/projects/cbc/ER10/pycbc-v1.5.4-verification-workflows/pycbc-masterSettings/output/local-site-scratch/work/././main_ID0000001/00/00/113523/113523/L-L1_HOFT_C
02-1135235072-4096.gwf" }
   ] }
]