Skip to main content
fortinetuser2020
New Member
Nov 22, 2017 8:34 PM UTC
Question

exchange server traffic problem with utm enabled

topology :

2 x fortigate 100d with 5.6.2

load balanced vip leading to 2 exchange 2016 server for rpc traffic (port 443)

on the incoming vip policy there are av/ips/waf/proxy/ssl for single server, full inspection

when the utm engines are on, the traffic flows very poorly in speed and continuity. if i turn them off, everything fly perfect.

i've narrowed it down to just the ssl and proxy (one cannot live without the other of course)

if i have just those 2 on, the problem occurs. so that tells me that the problem is not at the av/ips/waf engines

 

the proxy is in default mode, no special settings, rpc over http disabled (enabling doesn't help), no large file limitations

the ssl profile is a single server protection, with the same certificate installed on the servers themselfs

 

any ideas what to look for?

6 replies

packetpusher
New Member
Nov 25, 2017 10:53 AM UTC

What volume of ingress & egress traffic traverses the rule during normal operations? How many existing & new created sessions do you see?

Translate
fortinetuser2020
New Member
Nov 26, 2017 6:21 AM UTC

it's hard to keep track because all of the traffic

why is that important? please educate me :)

Translate
packetpusher
New Member
Nov 26, 2017 3:49 PM UTC

It's called "The ‘Black Art’ of FortiGate Sizing". Depending on what your UTM setup look like, amount of traffic traversing the Exchange server policy, the overall number of policies, whether you utilize flow or proxy-based method, and etc. you will get a certain performance behavior.

 

"the proxy is in default mode, no special settings, rpc over http disabled (enabling doesn't help), no large file limitations the ssl profile is a single server protection, with the same certificate installed on the servers themselfs"

 

From the above, it appears that you covered some of the topics. Try to disable, one UTM profile at a time while recording the output of the following diagnostic commands:

get system performance statusget system performance firewall statistics 

di sys top-summary

di sys top-summary "-s mem"

 As a result, in order to conserve firewall resources, you may need to fine-tune your security profiles, session timeouts, DNS cache, FortiGuard cache TTL, and etc.

Translate
Risk and Resilience in the Age of AI. An OT Thought Leadership Summit for Houston's Critical Infrastructure Leaders. Date: Thursday, December 3, 2026. Time: 8:30 am - 1:30 pm (CT). Location: The Westin Houston, Memorial City | 945 Gessner Road, Houston, TX 77024