Facing abnormality in Find command

UNIX for Beginners Q & A
1

Whenever the below command is being executed by a scheduler at UNIX environment, we are getting below error

cd /tmp/log
find . -ignore_readdir_race ! -name . -prune -iname 'XYZ*' -type f -mtime +4 -printf "%f\n"
./tmpfile.script_name.2.31885.201906071336.tmp': No such file or directory
./tmpfile.script_name.1.31885.201906071336.tmp': No such file or directory
./tmpfile.script_name.1.31869.201906071336.tmp': No such file or directory
./tmpfile.script_name.1.31879.201906071336.tmp': No such file or directory
./tmpfile.script_name.2.31869.201906071336.tmp': No such file or directory
./tmpfile.script_name.2.31879.201906071336.tmp': No such file or directory
Aborted after 00:00:14: ERROR listing files for XYZ* in /tmp/log

But interestingly XYZ* files are present in the current directory /tmp/log
what could be the reason behind the find command failing ?
This command was working pretty fine until our UNIX storage server was changed from one vendor to other

2

The files being found are deleted before the find finishes i guess.
Are you sure you are not running multiple scripts against those folders, altering content with each ?

As for the storage part, you would need metrics from operating system before and after the change.

Regards
Peasant.

3

I am surprised that despite the -ignore_readdir_race option the find complains like this. Perhaps a bug?
It might do the job nevertheless.

Try to forge a good exit status with

find ... || exit 0

This is shell code that works if the scheduler invokes system().