Troubleshooting Tip: FGSP recv_err Counter Increasing in FortiGate Clusters

[kumarh]

Description	This article describes how to address an issue where the FGSP recv_err counter increases in FortiGate clusters, which may indicate session synchronization errors. Although the increase in this counter may not immediately affect cluster performance, prolonged errors could lead to stability concerns. This issue can impact session synchronization in clustered environments.
Scope	FortiGate.
Solution	To check the FGSP recv_err, can run the below command: diagnose sys session sync sync_ctx: sync_started=1, sync_tcp=1, sync_others=1, sync_expectation=1, sync_nat=1, stdalone_sesync=1. sync: create=2984287144:1822781351, update=766071495, delete=388955745:30323655, query=1573512550 recv: create=1490107562:3502980314, update=1488235433, delete=2618142462:28283218, query=4009723522 ses pkts: send=1900832066, alloc_fail=0, recv=3366892477, recv_err=598764316, sz_err=0, ses_convert_err=0 udp pkts: send=679, recv=231061 nCfg_sess_sync_num=5, mtu=1500, ipsec_tun_sync=1 To resolve the issue of increasing FGSP recv_err counters, configure identical session-sync-dev interfaces for all FortiGate members in the FGSP cluster. This ensures proper session synchronization, which will prevent the counter from increasing without causing an impact on traffic or session management. Configure identical session-sync-dev interfaces by running the following commands: config system standalone-cluster     set session-sync-dev "<interface 1>" " interface 2" end It is required to use the same interface for session-sync-dev: For example, If two FortiGates in the FGSP cluster then configure them like this: In FortiGate A: config system standalone-cluster     set session-sync-dev "port1" "port2" In FortiGate B: config system standalone-cluster     set session-sync-dev "port1" "port2" Related articles: Technical Tip: HA session-sync-dev configuration Technical Tip: Suggested Parameters to use for a FortiGate Session Life Support Protocol (FGSP) clus...