Probably filed under pf: I have firewall systems using pf, with ftp_proxy as well. Under certain circumstances, relayd will crash when updating a table if something else (such as ftp_proxy, tftp_proxy, or a local rules edit) has /dev/pf locked. I've seen other cases, where an individual ftp_proxy will also fail, if something else grabbed /dev/pf as well -- and seeing "pfctl: DIOCADDRULE: Device busy " failures when doing manual updates is quite common. relayd logs attached: Oct 28 17:26:15 fw-1/fw-1 relayd[31847]: table devproxy-http: 1 added, 0 deleted, 0 changed, 2 killed Oct 28 17:26:15 fw-1/fw-1 relayd[31847]: fatal: sync_ruleset: cannot add address to pool: Device busy Oct 28 17:26:15 fw-1/fw-1 kernel: Oct 28 17:26:15 pubgw2 relayd[31847]: fatal: sync_ruleset: cannot add address to pool: Device busy Oct 28 17:26:15 fw-1/fw-1 relayd[31848]: hce exiting, pid 31848 Oct 28 17:26:15 fw-1/fw-1 relayd[31849]: relay exiting, pid 31849 Oct 28 17:26:15 fw-1/fw-1 relayd[31851]: relay exiting, pid 31851 Oct 28 17:26:15 fw-1/fw-1 relayd[31852]: relay exiting, pid 31852 Oct 28 17:26:15 fw-1/fw-1 relayd[31854]: ca exiting, pid 31854 Oct 28 17:26:15 fw-1/fw-1 relayd[31853]: ca exiting, pid 31853 Oct 28 17:26:15 fw-1/fw-1 relayd[31850]: ca exiting, pid 31850 Oct 28 17:26:25 fw-1/fw-1 relayd[31846]: lost child: pfe exited abnormally Oct 28 17:26:25 fw-1/fw-1 kernel: Oct 28 17:26:25 pubgw2 relayd[31846]: lost child: pfe exited abnormally Oct 28 17:26:25 fw-1/fw-1 relayd[31846]: parent terminating, pid 31846