We still have the problem that some CPEs loose IP Connectivity once a day.
Looks like it’s not bound to wireless stuff. Looks like it’s a bridging problem
in Layer2/NAT mode. Rebooting the PMP320SM or CAP320 solves the problem
for a while. Just playing with the bridging parameters searching for a setting
without this problem. It just behaves like the SM vanishes from the bridge table
of the CAP320. We see this happening the same time for several SMs on the
same CAP320. Rebooting one of the SMs this SM is pingable again. Rebooting
the AP all SMs are pingable. Service Flows stay up.
I’m very bored with this equipment. It needs another year to mature.
C’mon Cambium dont talk, do your homework.
We’re experiencing the same issues, but we’re looking at the etherWAN switch as the culprit.
Is your network VOIP or data only. We do a lot of VOIP and while our Packets Per Second is not at the stated 148,000 limit for the etherWAN, I don’t think it is up to the task.
Have you thought of bypassing the etherWAN? I have successfully done it in the Lab, but Moto and Cisco have NEVER gotten along when it comes to Layer 1 port negotiation. My core-2-egde network is all Cisco and when I plug the CMM4 directly in to a 3550, almost HALF of the packets are lost. Foundry/TrendNet and other crappy little switches have all worked for me in the lab fixing the Layer 1 issues, but I can’t generate enough traffic (PPS) to force an issue.
Half the packets lost when hooked up to a 3550? Yikes! Are you forcing 100/FDX or letting it auto-negotiate? Cisco gear and FastEthernet don’t always do well in auto mode.
cscheper@pngmail.com wrote: We're experiencing the same issues, but we're looking at the etherWAN switch as the culprit.
Is your network VOIP or data only. We do a lot of VOIP and while our Packets Per Second is not at the stated 148,000 limit for the etherWAN, I don't think it is up to the task.
Have you thought of bypassing the etherWAN? I have successfully done it in the Lab, but Moto and Cisco have NEVER gotten along when it comes to Layer 1 port negotiation. My core-2-egde network is all Cisco and when I plug the CMM4 directly in to a 3550, almost HALF of the packets are lost. Foundry/TrendNet and other crappy little switches have all worked for me in the lab fixing the Layer 1 issues, but I can't generate enough traffic (PPS) to force an issue.
We use Mikrotik routers at the towers. So each PMP320 is connected to a
MT port which is tunneled with EOIP to a central router which bridges all
EOIP-Tunnels together. We've no layer1 problems.
When I see this problem I get Voip-Packets from behind the SMs which
I cant ping. So it looks like the PMP320 AP does not know where to forward the
packets to the SM but the SM knows and can send through the AP.
Looks like the CAP320 looses some bridgetable entries for the SMs.
salad:
My core guys are looking into this now. I don’t lose any packets to the CMM4 management interface, just to the three APs.
This is my lab CMM4 because the GPS timing module got blown up in a near strike. We are tracking this down as a possibility. I have the CMM4 set to sync from the Master (GPS Module) and the APs to pull timing from the CMM4. They all show No SYNC. I know the CMM4 pulses timing to the APs on the CAT5, but why would it effect data?
Do any Moto Engineers still read this?
ste:
I have mirrored the port at the tower that connects our edge 3560 and the etherWAN switch. When this happens to us, I see packets egress our 3560 bound for Mikrotik Routers downstream of the CPEs, but nothing coming back.
Rebooting either the CPE or the AP solves the problem for us too. NOT A GOOD WORKAROUND.
At all times, our CPEs are connected and we can log in to their http interface on the management VLAN. FWIW.
cscheper@pngmail.com wrote:
Do any Moto Engineers still read this?
No. We have an open case for this but after first level support from
Europe (Teamviewer session testing some settings) the case went to US and now
there is *silence*.
Did anyone ever get a response on this from Moto? We are experiencing the same problem. Sm’s aren’t responding but after reboot of either SM or AP they are fine again. All. Throne the SM doesn’t respond there is still an RF link to the AP.
Hi,
We have the same problem and we open case "Idle Cpe " . Please open case and Cambium run special skrypt for you and snifing your problem.
Regards
We will call Cambium today. From our monitor system It seems like they lose connectivity for exactly 9minutes then start replying again…
we noticed this as well and it actually ended up being a DHCP problem for us, when the lease would expire, the CPE would drop dead both cambium brand, and awb. after a while they would reboot and self fix, we solve the problem by offering a 7 day lease and for whatever reason they will release and renew just fine.
we use juniper routers (srx210 or 240 each 100 series AP and wimax base stations are individually routed on unique subnets) at our towers or VLAN back to one of them for routing, since extending our lease to 1 week, we have not had the problem once. also we noticed a dhcp reservation fixed the issue as well, but it still needed to be longer than 48 hrs.
and obviously static ip did too…
not sure it’s the exact same problem everyone was experiencing, but the symptoms matchup.