- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
AWS Fortigate instance incommunicado...
I have a Fortigate instance in AWS that I cannot get to with ssh or https. It was accessible since creation (months), then all of a sudden it was not. I can ping it, and (by using nmap) can see that ports 22, 443, 500 and 4500 are accessible and listening. Security is wide open to this instance. Any ideas what could be wrong?
The console image shows the following:
System is starting...
Serial number is FGTAWS000133210D
FGTAWS000133210D login:
AWS instance-id: i-0133210d2e0c26da
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Have you check
trusthost ?
security-group?
Ken Felix
PCNSE
NSE
StrongSwan
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Wouldn't the fact that I can scan the relevant ports mean there's nothing blocking the traffic?
I looked at the flow logs for the WAN interface, and the traffic is being accepted ok.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I did more comprehensive scanning, and it's only finding tcp port 541 open.
Port 541 is used by FortiManager. Could my FortiGate have been somehow put into remote management mode?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
What changes did you make in AWS and fortiOS? If it was working and now not working, undo or revert any changes. I would also not over look any local-in policies.
Ken Felix
PCNSE
NSE
StrongSwan
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I made no changes prior to this issue cropping up.
Can you elaborate on "local-in policies"? Are these in AWS or on the FortiGate?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
You have 3 or 4 things to check
1: route
2: trusthost for any allowaccess
https://kb.fortinet.com/kb/documentLink.do?externalID=10868
3: local-in policy
4: proper SG in AWS
If your not getting a login prompt, I would research all of the above. if you did not change the fortios-cfg than the logical step is to look at the SG in AWS. Also I had a client that change the elastic ip and so he was trying to access the Fortigateinstance with the wrong EIP address.
Ken Felix
Ken Felix
PCNSE
NSE
StrongSwan
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks for those links.
Routing is not an issue, since, like I previously said, I can ping the instance and perform a port scan on it.
I did not enable "Restrict login to trusted hosts" nor configure local-in policies to restrict access.
The AWS Security Group for the instance is wide open, and I have confirmed that the traffic is getting to the instance by looking at the VPC flow logs of the WAN interface.
I believe the problem is that it's in some weird state, because the port scan shows nothing is listening on the usual access ports 22 and 443. The only port open is 541, the FortiManager access port.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Yes, I have stopped and restarted the instance multiple times.
