New to WAAS

Unanswered Question
Jul 19th, 2008

Hi, I am new to WAAS and testing this setup: datacenter (site A) and branch (site B). CM and core WAE 512 at site A and WAE 511 in site B. I am using WCCP negociated return because the WAE's are on server subnet in site A and client subnet in site B.


"show ip wccp" on the routers shows increasing CEF count for both 61 and 62.


However, in the CM monitoring section, it shoes zero savings.


There are no errors in the logs and no other apparent misconfiguration. What am I possibly doing wrong? There is no apparent optimization taking place.


Many thanks!


P.S. WAN link is a DMVPN and eveything else works fine.

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Zach Seils Sun, 07/20/2008 - 15:10

Pierre,


What type of traffic are you trying to optimize?


Do you see any optimized connections (sh tfo conn summ) on either WAE?


Zach



plemieux72 Mon, 07/21/2008 - 03:28

Hi Zach, actually, yes, I do see connections on both WAE's and they happen to be when I access intranet sites at either site remotely. I also see some TCP 3389 connections, which was expected. Some connections show as "PT App Cfg" and others as "PT In Progress".


The main goal is to optimize traffic for remote users at site B accessing Sharepoint sites at site A. I assume I will need to preposition/pre-load specific Sharepoint content?


Eventually, I'd also like to optimize Windows Meeting Space and SMB traffic. Or is this automatically included thru TFO, DRE and LZ?


Thank you!

Craig Norborg Mon, 07/21/2008 - 05:28

One question I would have is what OS are you using? The newer OS's tend to be easier to work with, less chance of having problems. I hear 4.1 which is due out in August is even better.


You shouldn't have to preposition traffic at the one site unless you know there are a ton of files that need quick access and won't be accessed that often. It should load your cache automatically for you.


Also, posting at least part of your configs on your Layer-3 devices and giving us IOS versions might help a bit.

Zach Seils Mon, 07/21/2008 - 06:05

The connections listed as "PT ..." mean that they are being handled as pass-through, or without optimization.


Do you see any connections showing as optimized (F,F,F,F or T,T,T,T)?


Zach



plemieux72 Mon, 07/21/2008 - 07:17

Aha! All connections show as "PT...". And, I see tons of pass-through traffic in the CM's monitoring tab, and zero savings.


All 3 WAAS WAEs run 4.0.19.14.


Core routers doing the redirection on the GRE/IPSec tunnel interfaces to the branch (61 out/62 in) are running 12.4(13f) and remote branch is 12.4(11)T3 (61 out/62 in on the tunnel interfaces too).

plemieux72 Mon, 07/21/2008 - 12:14

It seems your post has pointed me to the right direction.


Using CM, in each Device, under Acceleration / Policy Definition, I edited the "N/A" Application Policy Classifier Action from "Passthrough" to "Full Optimization" and now I see F,F,F,F.


I thought I had read that the AllDevicesGroup contained this "Other" policy definition with Full Optimization by default, however, I was wrong since this group shows no policy definitions created at all.


So, if I understand this correctly, suppose I am satisfied with the default application policies as listed in appendix table A-1, there is nothing further for me to do?

pmccubbin Mon, 07/21/2008 - 13:32

Hi Pierre,


You have been given some great advice by Zach and I hope you go out and buy his book from Amazon:


Deploying Cisco Wide Area Application Services (Networking Technology)


Now from my experience, it is always best to begin by making sure that the default application policies are in place. If you are in doubt that someone might have configured something different, then use the GUI to set them to default. This will give your troubleshooting and your tuning a well-known base to begin with.


Hope this helps. Please let the forum know how you are progressing so others might benefit from your experiences.


Best,


Paul

Actions

This Discussion