INSSLR - Test Plan

Regression Tests

When running with failover mode disabled, INSSLR behaves as INSSL with added http -> https redirection. Verify INSSL test plan passes successfully.

New Test Cases

Common tests

• Verify new properties match requirements.
• Verify description is properly updated.
• Verify the appliance is properly cleaned up.

http -> https redirection

• Set l7_accept to https and try to request http://appliance_ip/, http://appliance_domain/, http://appliance_domain/some/valid/path - they should all get redirected to the appropriate https URL
• Repeat the test but with l3_accept_proto set to all, check that all tcp traffic goes to aux except for http and https
• Repeat the test but with l3_accept_proto set to tcp, check that all tcp traffic goes to aux except for http and https
• Repeat the test but with l3_accept_proto set to tcp and l3_accept_port 80,443. Check that no traffic goes to aux
• Repeat the test but with l3_accept_proto set to raw and l3_accept_port set to 6. Check that all tcp traffic goes to aux except for http and https

ctl terminal

• Create an application with two INSSLR appliances and two WEB5 appliances.
• Start both INSSLR appliances with fover_mode set to none and verify that the ctl terminal does not accept connections ( try to connect from the WEB5 appliances)
• Change fover_mode to symmetric on both appliances. Check that the ctl terminal accepts http connections when l7_accept is set to none, http, https and both
• Check that the following calls work (http request ion the ctl terminal for /?action=):
  • status - should return the current state of each INSSLR appliance. At any time only of the appliances should be reported as active, the other should be passive. For the appliance that is reported as active - check that it actually is active (do `ifconfig` and verify that the external IP address is configured).
  • active - run this on the passive node, in a few seconds it should become active (issue a status command to check). Run the call again and verify the appliance is still active
  • passive - run this on the active node, in a few seconds it should become passive (issue a status command to check). Run the call again and verify the appliance is still passive. Stop the other INSSLR appliance and wait for the current appliance to become active. Issue a passive request again and verify that the call DOES NOT succeed (since the failover cant be complete). Check the status (it should return "active") and verify that the service is available and is not interrupted by a "passive" call when the other node is not working (run a ping from an external server to the external IP and check there are no packets lost).
  • kill - start both INSSLR appliances and issue a 'kill' on the active one. The appliance should shutdown and the other node should takeover.

Failover Test Cases

in1.fover_nfy_prefix = /dir1/dir2/script.php?
in2.fover_nfy_prefix = /dir1/script.php?username=user&password=pass&
web1.logs_enabled=on
web2.logs_enabled=on
web1.access_log_filename=web1.log
web2.access_log_filename=web2.log

No need to fover_nfy_prefix dirs/scripts on the web servers, we`ll just check the logs that the valid URL is requested.

• Start the application above with both INSSLR appliances using fover_mode set to symmetric. Verify that one of the appliances becomes primary and the other is backup. Stop the active INSSLR appliance and check that the other appliance takes over. After the takeover is complete, start the first appliance and check that it does not automatically take over. Check that each INSSLR appliance sends a notification to the grid dashboard and requests the URL configured in fover_nfy_prefix (check the logs on the web server, they should look like /fover_nfy_prefixfover_mode=fover_mode&state=<start|stop>&ip_addr=ip_addr&fover_local_ip=fover_local_ip&fover_remote_ip=fover_remote_ip&fover_netmask=fover_netmask
  .).

• Start the application above with both INSSLR appliances using fover_mode set to asymmetric. Verify that the appliance with the lower fover_local_ip becomes primary and the other is backup. Stop the active INSSLR appliance and check that the other appliance takes over. After the takeover is complete, start the first appliance and check that it automatically becomes primary. Check that each INSSLR appliance sends a notification to the grid dashboard and requests the URL configured in fover_nfy_prefix (check the logs on the web server).