Knowledge article KB0000491: Answer If (an accessible copy of) the data is on a diskserver, and the issue is on a private machine (non-LXPLUS/BATCH), the user should be invited to retry on LXPLUS. If data access on LXPLUS is working, most likely the private machine has a local firewall that prevents CASTOR callbacks from working. On SLC machines, the correct configuration would be the following lines in /etc/sysconfig/iptables (these need to appear before the first REJECT statement):-A RH-Firewall-1-INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT # controlled by CASTOR-client RPM # allow callbacks for RFIO, port range corresponds to default callback range -A RH-Firewall-1-INPUT -m state --state NEW -m tcp -p tcp --dport 30000:30100 -j ACCEPT # end of CASTOR-client RPM modifications This configuration should be installed automatically with thecastor-lib RPM, but later modifications might have overwritten it. If you modify the file by hand, you'll need to run /sbin/service iptables restart