Here are the RF counters from one of my 5 GHz radios with an average of 20 stationary PC clients:
Tx Fragment Count 278129134
Multicast Tx Frame Count 1056845
Tx Failed Count 3514621
Retry Count 296335531
Multiple Retry Count 5022954
Frame Duplicate Count 0
RTS Success Count 0
RTS Failure Count 0
ACK Failure Count 24882460
Rx Fragment Count 0
Multicast Rx Frame Count 0
FCS Error Count 1980165
Tx Frame Count 278129134
WEP Undecryptable Count 4418
Do any of these numbers jump out at you? Or is this "normal"? I see that about 2% of transmissions are "multiple retries" which I would think is acceptable. But there are more retries than frames transmitted. Should I be concerned here? How about the ACK failures (about 9%).
Nobody is complaining so far. :)
What type of AP is this from?
Marcus, its a lightweight Cisco Aironet 3502e.
How long have these counters been running? Are the users complaining ?
It is difficult to make a judgement without all the details. The ACK failures look high.
The multiple re-tries refer to more that one re-transmission. A re-transmission is the first re-try after initial effort fails. A re-transmission of max 2% is recommended . I have never seen a figure for multiple re-transmissions but multiple re-transmissions are bad.
Do the users complain of poor service - like downloads of web pages that freeze for a while ?
The counters have been running for 67 days and no the users are not complaining. The PCs are doing RDP only.
Here's some more info about the AP - local mode, 3 omni dipole antennas per radio. the other radio (2.4 GHz) is serving 10-15 wireless printers. I don't like the current installation/antenna positions, so I'm going to remount the AP and see if the numbers get better. Its blocked completely on one side by a steel i-beam, except for an inch or so of one 2.4 GHz antenna and one 5 GHz antenna.
ACK failures looks a bit high.but I gather that the apps are not time sensitive so small delays are ok.
I gather that you could try to lower the AP to clear the antennas abit more to see what happens.
The users are not complaining though. Take some RSSI readings at various points before moving the AP. Move during a low useage period and re-check the RSSI. That way you could tell if coverage was improved without affecting the users.