[HTCondor-users] condor_status blocks from an execute machine

[mohammed shambakey <shambakey1@xxxxxxxxx>]

Hi

I'm trying to install htcondor on two 24.04 Ubuntu machines. M1 has the 3 roles, M2 is only an execute machine. I had to install it in a strange way on each machine. First, I had to comment out the part in the get_htcondor.sh script that checks if there is a previous installation. Then, I had to install mini-htcondor on each machine. Then I used the script file to install the required roles on each machine. Finally, I removed the mini-htcondor configuration file from each machine.

I specified the CONDOR_HOST as M1 in the configurationÂfiles of each machine. Firewall is disabled completely. I added the machines to the hosts files of each one. I checked that each machine can ping the other. I don't have NFS installed, but I instructed the get_htcondor.sh to use one anyway (I don't know if this contributes to the problem). I added "ALLOW_READ=*" and "ALLOW_WRITE=192.168.10.*" to the /etc/condor/condor_config file on both machines.

After restarting the condor.service on both machines, condor_status blocks from M2 saying it cannot access M1. And M1 can only see itself. I wonder how to solve it.

Regards

--

Mohammed