ISP Redundancy link nothing probes
At a glance
Base severity
Area
Platform
Evidence level
Benchmark reference
Reads the export, not the device
ConfigProof reads a configuration export. This control is evaluated against the objects in that file, and it reports what the file says. It does not connect to the device and sees no traffic, so a finding here describes the configuration, not what the firewall is doing right now.
Why it matters
When that ISP fails beyond its first hop, the gateway still sees the link as up and keeps sending traffic, or failing back, to it, so ISP Redundancy does not move traffic away from a dead ISP.
How to fix it
Web interface
The steps below are written for the web interface. A finding raised on your own export carries the same steps with your object names filled in.
- In SmartConsole, open the gateway or cluster object > Other > ISP Redundancy, edit the link and either clear the option that stops pinging the next hop or add hosts beyond the ISP under Tracking > Monitored hosts.
- Publish, then install the policy on the gateway.
Command line
Check Point mgmt_cli and Gaia clish
mgmt_cli login > id.txtmgmt_cli set-generic-object uid "<isp-link-uid>" mispDontPingNextHop false -s id.txtmgmt_cli publish -s id.txtmgmt_cli logout -s id.txt
An example. Values in angle brackets are yours to supply; a finding on your own export fills in the names it read. Nothing here installs a policy. Publish, review, then install the policy on your gateways yourself; Gaia changes need save config.
Read it back
mgmt_cli show-generic-object uid "<gateway-uid>"Values to fill in2
- <isp-link-uid>
- The uid (objId) of the ISP link, from show-generic-object on the gateway
- <gateway-uid>
- The uid of the gateway or cluster object
Before you type it4
- Written from the Check Point Management API reference and Gaia clish syntax, not yet run on a device. Read each line before typing it.
- Run on the management server, or add -m with the server and -u with an administrator to each call; publish makes the change visible to other sessions.
- Install the policy on the affected gateways after reviewing the published change.
- ISP Redundancy has no command of its own: these change the gateway's generic object. show-generic-object on the gateway or cluster lists each ISP link with its uid (objId). SmartConsole does the same under the object's Other > ISP Redundancy.
How to confirm the fix
- On the gateway, run cpstat fw -f multi_links and confirm the link shows its probe state; take the ISP down in a maintenance window and confirm traffic moves to the other link.
Limits
- A link whose next hop is the ISP's own router fails over only when that router stops answering, not when the ISP's network beyond it fails.
Compliance references
No benchmark recommendation is mapped to this control in this version.
More in Routing and Network Services
See it on your own firewall
One complete analysis of one firewall costs nothing. A card is required and nothing is charged.