I dont think this fully solved.
Yeah, turning on persistent NAT on the Sonicwall dramatically fixes them popping on and offline - they bounce up and down quite quickly watching the management web interface - but there are still emails coming in, sure a lot less, but still coming in with that change of device state.
This is one thing i posted about them (this was well after i enabled that NAT setting:
But today, i got an email asking if there was a wifi issue in one of the rooms, and as i was checking the management web interface, i got an email about the AP in that room changing state to “connection lost”
Remember i was in the management interface, and I didn't see that AP go down, the status lights remained green.
So, there is something still going on, and i don't buy that “must be latency in your network, sorry i just don't. We are a school and have 160 AP’s (mostly AP650(AH) and some outdoor AP1130’s) and ill get those emails when no one is at school, and thats why i dont believe that latency is the cause. It might be for some people, but not here especially when no one is on campus.
But have a read of the link in this post, because its very odd that those emails went away several updates ago. I’m talking when you login to the management interface and have the banner at the top that states “been updated click to see whats new” message (and not firmware updates), and emails came back after an update or two ago. Very, very odd…… Are whatever you are doing in the background updates affecting the AP’s at customer locations. What are you updating.
Post back here what you think.
Only me then! Hmmmm…
Because ive received 11 emails since 1:35 am about AP’s losing their connection, here are 10 of the emails:
But your wifi shouldnt be impacted by the aps losing their connection to the XIQ ? Even if all your aps are offline to the XIQ your wifi should still work or not?
Thanks for your reply.
Well and that's the thing. Last week id have said you are absolutely correct, even with those emails coming in, the AP’s are up, working and passing traffic.
However, this week I received two of those emails regarding three AP’s on a floor of one of our buildings. As i was going through the XIQ interface to make sure they were indeed up, I received an email from a person in a room with one of the AP’s in the email that I had received, asking me if there was an issue with wifi. The XIQ interface showed them as green, being up.
I find it hard to call that a coincidence, I get two emails about three AP’s on that floor, one of them in that persons room, and not 2 minutes from the time I got that email, and was in XIQ, I get an email from that person asking if there is a wifi problem.
She asked, “is there a problem with wifi, as my students cant connect to it”.
So you see, there does appear to be some kind of issue, even if you see the AP’s as green within XIQ interface. I don't know if it lasts just a few seconds, whats causing those connection lost emails, and why they would show up in the interface as online when people clearly have an issue. Its possible it only lasts a few seconds or at least under 30 seconds and maybe thats not enough time for XIQ to consider them down, but the AP’s are doing something, maybe internal to themselves. Regardless, I cant call that a coincidence.
Are you saying when I get that email, its only telling me it cant “call home”? Because I don’t think that's the case; when I have had a PoE switch go down, ill get an email with all the AP’s listed that are connected to that switch, saying they have lost connection. I don't think its about communicating with home base and vice-versa. If it was, then it would have to be communicating a heck of a lot with home base, although i suppose it might be, as we tend to know an AP is down after a minute or so. Having the internet between the AP’s and its controller (whether thats an appliance, VM, or a service) isnt the way id have gone. I used to have Ruckus for my wifi, and having the hardware in my server room made it more efficient, insomuch as letting me know in under 30 seconds that an AP was down.
Are you saying when I get that email, its only telling me it cant “call home”?
Only your cloud user groups should be effected. All local configured users should still connect
There could be several other reasons for users having problems.
power consumption of the switch…. to less internet speed for the user capacity...Channel Interference...wrong radio Profil...
Thank you dpanev, I appreciate your reply.
It was just odd to get the email, and within 2 minutes a user asking if there was a wifi issue PoE was good, Internet speeds we about 500mbps up and same for down speeds, radio profile is good (Ive had to modify it way back when we first got it, so i know its fine), so that leaves channel interference. Ill have to look into that, although it will have to be on a case-by-case basis of course.
Thanks again, appreciate the input. J.
I realized, i didn't clarify the bandwidth up and down. We have 1gb up, and the same for down, but it was around 500mbps up and down when i got that email from the user (i did a speed test at that time), so there was plenty of bandwidth.
Here may be something to add to this stale conversation. We had an AC issue, and it got how enough to make our two DC’s power down (middle of the night). They were down for quite some time, as the AC issue was addressed, around 5 hours of downtime.
Starting about an hour after the DC’s were back up and working, i received tons of emails about AP’s going offline (loss of connection to CAPWAP. I thought it was probably just catching up after the outage. Well event he next day, im getting dozens of individual email notifications about an AP here, and AP there, all over the place.
I dont know if it means anything, but i find it odd, the volume of emails are back to the worst they have been, even though the AP is actually up and working.
Contact Us:Sam PirokCommunity@extremenetworks.com