Any SRX may face the NSD failure on both nodes along with core-dumps and causes a traffic outage during cluster reboot, failover or policy addition

SRX cluster with the logical systems (LSYS), the network security daemon (NSD) process crashes due to cluster failover or reboot of node in a standalone firewall setup or while adding security policies. This crash generates core dump and prevents the system from configuring security policies.

As a result, all policy configurations are removed, leading to a traffic outage.

1.NSD keeps reporting configuration reading error from messages logs

  nsd[80616]: NSD_POLICY_CFG_SKIPPED: In LSYS [lsys-name], 1 policy is skipped due to some configuration reading errors. Please commit the configuration again. 

  nsd[80616]: NSD_RESTART_COMP_CFG_READ_FAILED: Subcomponent could not read configuration database

  alarmd[80605]: Alarm set: NSD id=-1728053247, color=RED, class=CHASSIS, reason=NSD fails to restart because subcomponents fail

  craftd[17901]: Major alarm set, NSD fails to restart because subcomponents fail

2. NSD core will be generated on both nodes if failover happened during problematic state 

Node0:

root@SRX-Node0> show system core-dumps no-forwarding

-rw-rw---- 1 root wheel  10307135 [time-stamps] /var/tmp/nsd.core-tarball.0.tgz >>>>>>>

Node1:

root@dz01094hi01> show system core-dumps no-forwarding

-rw-rw---- 1 root wheel  10596824 [time-stamps] /var/tmp/nsd.core-tarball.1.tgz >>>>>>

Solution:

1. Restoration-Steps

• Restart both the nodes at the same-time if device is configured in a cluster 
• Restart the device if configured as a standalone node.

2. Upgrade the platform to a fix version stated on PR1857379