Re: [HTCondor-users] Bug STARTER at <ip> failed to send file(s) to <<ip>:9618>; remaps resulted in a cycle:

[Zach Miller <zmiller@xxxxxxxxxxx>]

Hello Vikrant,

This is a circuit breaker in the code to prevent infinite recursion when doing filename remaps during transfer.

The default (as you found) is that 20 directories deep means something might be going wrong.  Luckily, there is a config setting you can change that determines how deep HTCondor will go before it considers it a problem.  In your configuration, set:
	MAX_REMAP_RECURSIONS = 100  # Or any number you like if you need to go deeper

Let me know if that works for you.


Cheers,
-zach



From: HTCondor-users <htcondor-users-bounces@xxxxxxxxxxx> on behalf of "ervikrant06@xxxxxxxxx" <ervikrant06@xxxxxxxxx>
Reply-To: HTCondor-Users List <htcondor-users@xxxxxxxxxxx>
Date: Monday, April 6, 2020 at 12:15 AM
To: HTCondor-Users List <htcondor-users@xxxxxxxxxxx>
Subject: [HTCondor-users] Bug STARTER at <ip> failed to send file(s) to <<ip>:9618>; remaps resulted in a cycle:

Hello Experts, 

Jobs went into held status with the messae shown in description. This happens when path mentioned forÂoutput/log attribute having more path with more than 19 nested directory.


Not working: "/tmp/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/out.txt"

working: ""/tmp/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/out.txt"


Thanks & Regards,
Vikrant Aggarwal