fortios_system_automationtrigger in mixed version environment

[zippanto]

Hi,

We are facing the same issue described in #212. The fix provided works well when working with a single 7.0 firewall or multiple 7.0 firewalls. However our environment is a mixed 6.4 / 7.0.

Having the logid_block is causing an error for 6.4. Please see below.

resource "fortios_system_automationtrigger" "ssl_vpn_login_fail" {
  name       = "SSL VPN login fail"
  event_type = "event-log"
  logid      = 39426

  # FortiOS 7.0
  logid_block {
    id = 39426
  }
}

Please see the error below.

fortios_system_automationstitch.ssl_vpn_login_fail: Modifications complete after 1s [id=SSL VPN login fail]
╷
│ Error: Error updating SystemAutomationTrigger resource: Internal Server Error - Internal error when processing the request (500)
│ Cli response: 
│ unrecognized logid:0
│ node_check_object fail! for logid 0
│ 
│ value parse error before '0'
│ Command fail. Return code -651
│ 
│ 
│   with fortios_system_automationtrigger.admin_login_fail,
│   on alerts.tf line 32, in resource "fortios_system_automationtrigger" "admin_login_fail":
│   32: resource "fortios_system_automationtrigger" "admin_login_fail" 

Please advise.

Thanks.

[zippanto]

It's worth noting that there was a similar issue with fortios_system_automationstitch where previously on 6.4 we had to use action{} block, but on 7.0 we now have to use actions{} block. Having both action and actions blocks defined in fortios_system_automationstitch resource causes no issues for 6.4.

Thanks.

[MaxxLiu22]

@zippanto ,

Thank you for raising this issue, the reason causing this error is the argument type changed from string to list in different version FOS, and they share the same name, this is why there are logid and logid_block in Terraform, but they are both corresponding to the logid in FOS, I will inform the development team about this issue along with your last case

Thanks, Maxx