That explains a little... RRM changed significantly from 4.x and 7.x, so it's not surprising that you'd see more of your AP's running at lower power. That said, RRM is set pretty aggressively by default, so I'd take a look at WCS RRM page, and pay particular attention to the Total channel changes in the last 24 hours. If the number is significant, you may need to adjust DCA channel assignment method interval and DCA Channel sensitivity. If I remember correctly, the settings are not exposed on WCS, but need to be set on each controller, and for each radio.
By default - DCA is set to an interval of every 10 minutes (way to often in my book), and the DCA Channel Sensitivity is set to High. Depending on your environment, the default could result in a lot of channel changes. I use DCA interval of 6 hours, and a DCA channel sensitivity of Low. TCP (Tx Power Control) can also be a factor here, especially if your coverage is light. That is, TPC will increase Tx power in response to coverage holes and other factors. If TPC is forced to make a lot of changes to Tx power levels, and DCA setting are aggressive - then TPC increasing the power on one AP could then force DCA to change channels on other AP's in response to it (mitigate interference). Jeff >>> Christina Klam <[email protected]> 9/2/2011 6:53 AM >>> Upgrade Path: We upgraded from 4.X to 7.0.116. TX power: As you all mentioned, I went from having 98% of all APs running at power level 1 before the upgrade to just 55.5% of our b/g and 90% of our a after the upgrade. (We are running a mix of b/g, n, and a.) I also looked at the APs in the areas of the reported connectivity and performance issues and they are all using 1131s with TX power of 1. But, I did notice that the channels changing over the course of a week. This is an anomaly in our environment. As planned this summer, I will continue to rollout new APs in these areas. As a tangent, I was hoping to contain a b-only clients in a separate SSID or VLAN. How do others deal with B clients? I did a report and found that we have at least five b-connections a week. In our environment, that is small but not statistically insignificant. Thank you everyone. Christina Klam Network Administrator Institute for Advanced Study Email: [email protected] Einstein Drive Telephone: 609-734-8154 Princeton, NJ 08540 Fax: 609-951-4418 ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/.
