I have an issue with a QFX5120-48YM 3 member VC where 9 times out of 10 my commit will fail. The commit check works fine, but there's some sort of communication error between the VC members. Below is one of the syslog messages I get when the commit fails. Just to be clear, we type commit as many times as we need to until the configuration changes actually commit. We do not modify anything in between attempts. Why would rcp fail while trying to replication files between the members? This fpc member and file isn't consistent either. Sometimes fpc 0 or 1. Sometimes it's a different //var/etc/... file.
<14>1 TIMESTAMP [HOST_REDACTED] ffp - libjnx_replicate_rcp_error [junos@26[ipv4_redacted].4.82.30 operation="rcp -l -ji //var/etc/stp_all_port.id+ fpc2://var/etc/stp_all_port.id+ " operational-status=""] rcp -l -ji //var/etc/stp_all_port.id+ fpc2://var/etc/stp_all_port.id+ :
------------------------------
JOHN GARRETT
------------------------------