Hi all,
I have an issue when a gateway is moved to a configured Large Server Cluster. Visible alert/alarm/warning is something like "The redundant Gateway XPTO for Cluster XYZ is down or in use due to the failure of another Gateway in the Cluster. The Cluster currently has no redundancy". Behavior similar as reported in
https://community.imperva.com/discussion/gateway-disconnected-from-cluster-manager.This environment was deployed recently, v14.8, with 12 GW (4 GW clusters with 3 GW each in "Dual - Separate agents Network" mode) managed by a MX-HA, with absolutely same HW/VM settings (excepting IP address, all settings are the same). MX and GW are not in the same VLAN (as recommended on Imperva docs), but they have fully communication each other, once 11 of 12 GW's was clustered successfully with this settings.
On MX -> Setup -> Gateways, the gateway shows running. On MX -> Cluster Management -> MX -> tab Gateway Groups, the cluster envolved shows 2 running and 1 disconnected. On MX -> Cluster Management -> MX -> tab Gateways, the Running Status column shows "Initializing" since it was moved to Cluster, and when I move arrow on this status, it shows me "MX reports: Gateway Initializing. Cluster reports: Gateway Failed. Gateway reports: Running".
I opened a support ticket, but the support team is claiming that the gateways and mx must be on the same VLAN, even though I have 11 gateways running correctly on clusters with exactly the same configuration, claiming that this could be the problem even without MX specific logs (or other location) related to the join failure of the GW involved in the cluster.
How can I find specific cluster logs to see what can happening ?
Regards
#DatabaseActivityMonitoring------------------------------
Uiliam de Mello
Information Security Analyst
Brazil
------------------------------