11-17-2013 10:09 PM - edited 03-04-2019 09:36 PM
Hi experts,
Recently we have been testing gre+vrf deployment for our guest segment. Attached is high level topology.
Below is how its setup
- Guest SSID terminated on Wism in our of our DCs 6509 switch
- WISM provides IP address to clients in specific VLAN
- Guest Network in vrf tunneled through the corporate network for segmentation and terminated to Cisco 29xx router in DMZ
- Tunnels in same vrf as guest vlan vrf, communication between tunnel etc works fine
- Guest can access and ping internet etc fine can see traffic both ways on external firewall
Issue: Guest network is having issues accessing entire internet, strange behaviour is seen. Google/gmail/youtube/Google news and bing works fine but none of the other sites load, 29xx routers show packets forwarded in cache flow where as packets logged on firewall as well. initiating a search in google is working but when the content provider link clicked does not present the page. Layer 1/2/3/4 show no issue. IPS is exempting gst network from being scanned, checked logs no drops at that level as well. GRE keepalive through Cisco ASA which is the internal facing firewall does not work either, NAT is used to create tunnel and is up/up but when keepalive implemented tunnel goes up/down
Other issue when debug initiated on 6509 switch getting lot of below log, whereas when debug ip packet detail initiated on 29xx router does not log packet forwarding from guest network ip source but ip cache flow has source and dest as gst network and internet
pak 500A4908 consumed in input feature , packet consum
ed, MCI Check(55), rtype 0, forus FALSE, sendself FALSE, mtu 0, fwdchk FALSE
Would appreciate if someone could confirm if its not an IOS issue?
Thanks
Solved! Go to Solution.
11-21-2013 05:33 AM
The symptoms that you describe suggest that it might be a fragmentation issue. I suggest that you try this command and see if the symptoms improve
ip tcp adjust-mss 1360
1360 is a sort of conservative/safe starting point. If this does improve the situation you might experiment and see if larger values work as well.
HTH
Rick
11-21-2013 05:33 AM
The symptoms that you describe suggest that it might be a fragmentation issue. I suggest that you try this command and see if the symptoms improve
ip tcp adjust-mss 1360
1360 is a sort of conservative/safe starting point. If this does improve the situation you might experiment and see if larger values work as well.
HTH
Rick
12-01-2013 03:14 PM
2 things, 29xx wasnt fragmenting and as i enabled content-scan on to forward traffic to ScanSafe it was crashing. Upgraded the IOS and no issues since then. Thank you for your prompt response Richard.
12-01-2013 05:44 PM
I am glad to know that you got the issues resolved. Thank you for posting back to the forum to update us about this issue and how you solved it. That could be beneficial to other readers of the forum. Also thank you for using the rating system to mark this question as answered.
HTH
Rick
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide