A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hello, I assume that you have resolved your issue since I have not heard back from you. If not, feel free to post a response to this thread, or open a new thread regarding this issue. Regards, Shlomi
I have not fully resolved my issue. I trying now to generate a nwp log file to see if my issue is related to this or not.
What I am seeing is that after some connect/disconnect's I get a BSSID reported back of 0:0:.... This is from a CC3220Sf as sta. with the AP being a CC3220SF also. Once the BSSID is reported as 0:0:0... the AP seems to be in a state that can not recover.. I am working on getting a good set of data to fully demonstrate the problem.
I have two CC3220SF's one on Unit A and one on Unit B. Unit A is acting as an AP for a laptop and Unit B to connect to as a Station. The laptop connects using DHCP to get an IP address. Unit B connects using a static IP address.
What I am finding is after some time in the field customers are having trouble with Unit B connecting back to the Laptop via Unit A's AP. Doing some testing on a returned unit and a factory fresh unit, I can reproduce the problem.
The problem shows up after some connect/disconnect cycles of Unit B. By forcing the connection with a turn on/off of the Unit B, I can get the AP to "lock up". (Also the laptop is connected, and seems to be a needed connection for the problem to show up.) From the Host processors logging over Uart, I see that the AP BSSID seems to be getting set to 0:0:0:0:0:0. Reseting Unit B does not fix the issue. Disconnecting the Laptop does seem to fix the problem. I can not connect to the AP when it is in this state with another device. I have captured a Network Processor log file of the AP (Unit A) using the pin 62 output. I have attached this log file.
Not sure exactly where in the log it happens (only towards the end?).
Anyway, from the log I can see that unit B gets disconnected since the NWP got a MGMT frame but I cannot tell what frame without looking at a sniffer capture. Do you have the capability to capture one?
Regarding the laptop that use DHCP, I can see that it gets rejected during DHCP procedure since it is using a DHCPINFORM option with all 0 IP in its ciaddr although it gots a valid address 10.0.1.2. I am not sure whether it is related to unit B issue or not.
I can try to create three logs, time stamped. One with the NWP log file, one with my codes output. and one with Wireshark(in promiscuous mode on a mac).
The current log file you have, has a sent of Unit B connecting, then getting the rejected connection. then a PC disconnect, followed by a PC connect.
I have created the logs. In capturing the Wifi sniffer data, I went to a location without much wifi interference. There are two extra Wireshark files that provide the "noise" wifi signals that I was getting during the capture.
Also, in looking at the NWP log file(as it was print on the terminal), I noticed that when I was getting the error of Unit B not connecting, I would get a long set of 2" as seen by this:
&('#BÝ& 72€% ( 2“
2“
2“
2“
2“
2“
2“
2“
2“
2“
2“
2“
2“
This would repeat while the unit A AP was not accepting the Unit B connection.
I have placed the log data from the host processor here that shows the sequence of events:
Unit A boots:
Unit B first added to AP on unit A: [WLAN EVENT] STA was added to AP: BSSID: c:b2:b7:7:15:74 Laptop connected to AP on unit A: [ WLAN EVENT] STA was added to AP: BSSID: 28:b2:bd:7c:32:56 Laptop gets IP: [NETAPP EVENT] IP Leased to Client: IP=10.0.1.2 Unit B removed from AP on unit A: [WLAN EVENT] STA was removed from AP: BSSID: c:b2:b7:7:15:74 Unit B will not reconnect..... shows ERROR BSSID 0:0:0.... Reseting Unit B does not fix problem. Laptop removed from AP on unit A: [WLAN EVENT] STA was removed from AP: BSSID: 28:b2:bd:7c:32:56 Uint B reconnects: [WLAN EVENT] STA was added to AP: BSSID: c:b2:b7:7:15:74 Laptop reconnects: [WLAN EVENT] STA was added to AP: BSSID: 28:b2:bd:7c:32:56
Disconnecting laptop allows Unit B to reconnect. Laptop will reconnect without issue.
1. The reason we have one device on static ip(unit B) is so we know what IP address it is... and will always be at.
2. I have now tested with two different computers acting in the role of laptop. One is my development laptop, which is windows 7, Lenovo. The second computer was a MS surface running windows 10.
On dev laptop, I get the problem. It is easy enough to reproduce, sometimes it takes a few hours, but sometimes it is almost immediately.
On the Surface, it does not yet seem to do it. I tested for about 2 hours yesterday without any issue.
Also, when no laptop/computer is connected to Unit A, it have not seen the issue.
You first spoke of about the dhcp of the laptop, DHCPINFORM option with all 0 IP , Would this be some the dig in on?
Also, I was wondering if adding the NWP log from Unit B would help?
Do you have any suggestions on reading for better understanding the wifi connection/association of two units? I have never spent a deep amount of time learning the lower levels of wifi...
Is there anyway to get a converted log of the nwp that would have some details as to what the output means?
What does the repeating 2" mean from the log... this is when I see Unit B not connecting.
After tracing through the data in cc3220nwp_log_3.zip above, I have found the set a packets in which the network processor does not allow a connection from Unit B.
From 410.2782 to 410.2818 a deauthentication, authentication, association request and association response with success returned from AP.
From 416.8840xx on to 426.814xxx show a pattern of no authentication response from the AP. There are 4 de-auth. packets sent from Unit B. 6 auth. request packets sent without respones.
From 460.xxx to 464.xxx a set of null functions are sent from the laptop to the AP(Unit A). At 464.3776xx the laptop sends a deauthenticate packet.
At 470.9639 Unit B starts the "connection" process with a deauthentication, authentication request/response and association request/response.
I am not sure how to sync up the nwp logs with the packet times. The only item on the log that matches the block of no authentication responses is the set of 2" without other readable data...
Can you still test with DHCP on unit B instead of static? Even when using DHCP, you know what the IP address you are assigned so I do not see a good reason of using a static IP. What if another station connects while unit B is disconnected and gets the same IP address as the static IP you think is dedicated to you?
It would be good to get unit B NWP log in parallel to unit A. Please do.
Unfortunately, the log is binary and you cannot read it. Only in TI we can convert back and decrypt it.
I have not had time to get back on this. We have a few other firmware features we are trying to get added so we can resolve a few problems customers are having with units(not related to this problem).
I can close this out for now and in a few weeks create a new post. Would that help close off "open" threads?
