I am using the root protocol to access data files on a grid-based storage system.
The file access string I am using is in the form
root://atlasse.lnf.infn.it:1094//vo.padme.org/daq/2019/rawdata/run_0000000_20190228_131527/run_0000000_20190228_131527_lvl1_00_040.root
All data files from a single daq run are splitted in groups of 100 files each and submitted as independent jobs to a grid CE for reconstruction. The 100 files assigned to each job are read by the reconstruction program as a single TChain.
One of the jobs exited after a couple of minutes with error message
Error in TNetXNGFile::Open: [FATAL] Invalid address
repeated 100 times (once per file read).
Other jobs from the same run and running on the same worker node did not have the problem and are still running.
I repeated the reconstruction command with the same file list on a different node and it worked fine.
I could not find any documentation which could help me understand what is the meaning of the error and what can be done to avoid it.
Does anybody in this forum know what the problem could be?
Thank you
Please read tips for efficient and successful posting and posting code
ROOT Version:/cvmfs/sft.cern.ch/lcg/app/releases/ROOT/5.34.36/x86_64-centos7-gcc49-opt
_Platform:_Linux CentOS7
Compiler:/cvmfs/sft.cern.ch/lcg/contrib/gcc/4.9/x86_64-centos7-gcc49-opt