rb532a v2.9.51 to v3.22 problem!

I know this is going to sound off the wall, and I’m sure some dumb ass is going to say “Search for the answer” but I hope someone out there has seen this issue before, so here it go’s. Around three days ago I updated all my rb532a’s from v2.9.51 to v3.13 and then to v3.22. Now I’m getting calls from customers saying that they can’t get a webpage up. After walking the first customer threw several things I finely decided to reboot the AP. After rebooting the AP their internet started working.

The strange thing is that while their internet wasn’t working I checked the following:

They had a good signal strength -62
I could get into their cpe without a problem
client could ping yahoo.com without a problem

My only guess is that my old firewall rules are messing something up!

But then again it isnt happening to all the clients at the same time!

Seems like a DHCP issue

But they are getting there ip’s.

Anyone else have any ideas?

my guess is that the client wasnt associated with the ap properly, ie routes, or arp entries. I notice this too when switching the same machine to different ether ports on my 493, that are different networks, takes a bit for it to work.

do people still get this issue again? or is it a one time issue for a few people?

Maybey at the time of testing, the site you tried was not working? I guess the problem has been resolved? not sure if you are still asking “what could of gone wrong” or “this is what happened” stament.

wierd things happen with technology.

BobcatGuy,

Thank you so much for your reply,

I’m not sure how the routes or arp entries could have done it unless the way they are handled is different between n2.9.51 and v3.22.

Two of the customers that had the problem stated that the internet just got slower and slower until it quit working. They could ping the website from a command prompt but the website wouldn’t come up in their browser. Since pings go threw a different port my best guess is that its something that is port related. They also said that their outlook express didn’t work either.

After writing the post I did some more reading and saw a couple of posts that suggested upgrading firmware before ROS. I didn’t realize this should be done, and hope that it doesn’t matter if it’s done before or after ROS upgrade. Anyway it took me all night to update the firmware on all my routerboard cpe’s and AP’s.

I’m hoping that will do the trick but am not holding my breath!

Thanks again for your input!

It seems strange to me that I find no mention of firmware on the ROS download page; it would seem to me that it should be mentioned. There should at least be a link from the ROS download page with some relevant information pertaining to upgrading firmware!

You know what they say about assuming!

Well it’s been three days since I had this issue.

I updated the firmware on all routerboards & removed all traces of old DNS info from all routerboards, and haven’t seen the issue popup again.

Knock on wood!!!

I may have to downgrade my rb133c though! V3.22 is causing the CPU load to hit 100% most the time, and it’s locked up on a couple of occasions. I had to cycle the power to get it to snap out of it!

Well it’s done it again!

I disabled my firewall rules and the customer was able to bring up a website where as they weren’t previously.

I haven’t changed any firewall settings for over a year and never had this problem.

Can anyone tell me if v3.22 handles firewall rules different than v2.9.51?

I figured out what the problem was, in case anyone was wondering!

I had another customer call having the same exact problem. So I had him try after I disabled one of my firewall rules. Then his internet started working!

add action=drop chain=forward comment=“allow only 20 simultaneous connections from each of the clients” connection-limit=21,32 disabled=no protocol=tcp tcp-flags=syn

My best guess is that these customers are infected with spy-ware, ect.