We use cookies on our website to ensure we provide you with the best experience on our website. By using our website, you agree to the use of cookies for analytics and personalized content.This website uses cookies. More Information
It seems like your browser didn't download the required fonts. Please revise your security settings and try again.
Barracuda CloudGen Firewall

Virtual Server Monitoring

  • Last updated on

To ensure and maintain the connectivity of a virtual server, you can define pools of IP addresses and/or network interfaces that are continuously monitored by the Barracuda CloudGen Firewall. If the health check of a monitored IP address or the link state of a network interface fails, the virtual server is automatically shut down. As soon as the health check target is successful, the virtual server is started again. Monitoring policies define which requirements must be met for the virtual server to remain active, or to be shut down. If you are using an HA cluster, you can use monitoring policies to define the behavior of the secondary HA unit. If necessary, you can use custom scripts which are executed when the virtual server is started or stopped.

Layer 3 Monitoring

The Layer 3 monitoring policy defines the settings for IP address monitoring. The policy configuration provides two address pool tables. Add the target addresses to the tables. These IP addresses must be reachable for the virtual server to stay up. The following Layer 3 monitoring policies are available:  

  • all-OR-all-present – All of the IP addresses from at least one IP address pool, e.g., from the Monitored IPs I table, must be reachable. If you enter IP addresses in both the Monitored IPs I and II tables, the IP addresses from at least one of these tables must be available. Otherwise, the virtual server is deactivated.
  • one-AND-one-present – At least one IP address from each monitoring pool must be reachable. If you only enter IP addresses in the Monitored IPs I table, at least one IP address from this table must be available. If you enter IP addresses in both tables, at least one IP address in each table must be available.

The control service runs an ICMP check on all IP addresses in 10-second intervals. If no answer is received, the IP addresses are probed every second for a 10-second period. If no response is received from a valid health check target during the 10-second period, the virtual server shuts down. The server is reactivated as soon as an answer is received for the subsequent probes. If the monitor target is part of a directly attached subnet, ARP responses are monitored too. If the ICMP response fails with presence of a valid ARP entry, the server is still marked as healthy.

Example Setup:

Layer 3 monitoring is configured for the virtual server S2, using both address pools with the following IP addresses and statuses:

Monitored IPs I Status Monitored IPs II Status

10.0.10.110

up 10.0.10.88 up
10.0.10.68 down 10.0.10.99 down

The status of the virtual server is displayed on Control > Server page:

If the monitoring policy one-AND-one-present is used, the server stays up because one IP address of each address pool is available.

pol1.png

If the all-OR-all-present policy is used, the server shuts down because at least no IP pool is fully available.

pol2.png

Layer 2 Monitoring

The Layer 2 monitoring policy defines the settings for interface monitoring. Add the interfaces that should be checked according to the policy in the Monitored Interfaces I and II tables. Layer 2 monitoring is available in the following modes:

  • all-OR-all-present – All of the interfaces from at least one interface pool, e.g. from the Monitored Interfaces I table, must be available.
  • one-AND-one-present – At least one interface from each interface pool table must be available. If you have added interfaces in one table, at least one IP address from this table must be available. If you have added interfaces in both tables, at least one interface from each table must be available.

The control service checks the link status of each interface on a regular basis. Depending on the selected policy, the server is shut down if the links on the monitored interfaces are unavailable. The server is restarted when the links of the monitored interfaces are up again.

Server Monitoring in High Availability Clusters

If your Barracuda CloudGen Firewall is part of an HA cluster, you can extend the monitoring policy to both units. For HA monitoring, you can select the following options: 

  • Monitoring on Backup Box – If set to No (default), server monitoring on box and HA box is processed only by the primary unit. In case of failover, the non-availability of health check targets is ignored by the HA box and the server stays up on the secondary unit. If set to Yes, the monitoring policy will also be enforced by the backup box. In case of a failover, the virtual server is then also deactivated on the second unit if the monitoring also fails on the secondary unit.
  • Shared-HA-Probing – Shared HA probing combines the IP address and interface information of both units. Both sets of IP addresses or interfaces must be available on both units. An IP address or interface that is not operational on both HA peers will be excluded from the HA logic decision. If a server is active on a unit and blocked on the peer unit, any probing results will be ignored. The probing decision will only be made if a situation persists over two probing cycles. This gives the system time to account for the delay between detection and synchronization and avoids aliasing effects.
  • Local-HA-Probing – (default) Only local health check target resources are probed. This means every HA partner performs its own monitoring procedure.

Step 1. Configure the Operation Mode

Configure the monitoring policies for IP addresses and interfaces that must be reachable in order for the virtual server to stay up. When your Barracuda CloudGen Firewall unit resides in an HA cluster, specify the monitoring policy for the case of HA failover:

  1. Go to CONFIGURATION > Configuration Tree > Box > Virtual Servers > your virtual server > Server Properties.
  2. In the left menu, select Monitoring Policy.
  3. Click Lock.
  4. From the Monitoring on Backup Box list, select whether monitoring should be performed and, in case of failover, adapted by a secondary HA unit.
  5. Select the Probing Policy. For more information, see Server Monitoring in HA Clusters.

Step 2. Configure the Monitoring Policy

Specify the monitoring policy for IP addresses and interfaces.

  1. Go to CONFIGURATION > Configuration Tree > Box > Virtual Servers > your virtual server > Server Properties.
  2. In the left menu, select Monitoring Policy.
  3. Click Lock.
  4. In the Layer 3 Monitoring section, specify the IP address monitoring policy. For more information, see Layer 3 Monitoring.
  5. In the Monitored IPs I / II tables, add the IP addresses that must be reachable via the ICMP protocol by the system that is hosting the server.
  6. In the Layer 2 Monitoring section, specify the interface monitoring policy. For more information, see Layer 2 Monitoring.
  7. In the Monitored Interfaces I / II tables, add the physical interfaces that must have a link in order for the server to stay up.
  8. Click Send Changes and Activate.

Configure Custom Scripts

Configure custom scripts for use with your monitoring policies. These scripts are run after the server starts or before the server shuts down due to unreachable IP addresses or interfaces.

Do not use phionctrl in your custom scripts; this might cause a deadlock.

  1. Go to CONFIGURATION > Configuration Tree > Box > Virtual Servers > your virtual server > Server Properties.
  2. In the left menu, select Custom Scripts.
  3. Click Lock.
  4. In the Start and Stop Script fields, enter the commands that should be executed when the server is started up or shut down (7-bit ASCII characters and standard Bash version 2-compliant).
  5. Click Send Changes and Activate.
Last updated on