Yes, I've followed that procedure exactly, and tried variations including hardware short. Totally unusual response - no beeps at all, all lights are on and none go off even after 2m of holding down either reset button. It won't boot in any fashion and does look like a warranty job, but I'm in the ha...

Once upon a time routers had RS232 ports. When they misbehaved we could connect a serial cable and access the console to see why.

Now I find myself with an RB912-UAG that refuses to boot or netinstall. Is there any method / alternative to connect to a console for troubleshooting?

Your links are down (domain not resolving in DNS) so we can't download.

I've got an RB433UAH connected to the NextG network using USB port + PPP client. Easy and works great. Until you upgrade past 5.14. Then you lose the ability to connect to the USB port with the modem and instead get a new LTE interface. I've played around but have not yet been able to figure out how...

Hmm, interesting results. CPU is same as what I see. Actual throughput is not. Using latest firmware.

Will try same test on a different board and see if itv is something uniquely flaky with that one.

Does it even matter which port you assign as master? assuming ports on same switch chip of course. I had: Ether 1 Master Port=None Ether 2 Master Port=None Ether 3-5 Master port = ether2 Ether 6 Master port=none ether 7-9 Master port =Ether6 Are you suggesting Ether 1 Master Port=None Ether 2 Master...

That of course is the config with bridge no switch which works great. Modifying ONLY the section to remove from brisge all LAN interfaces except 2 / 6 and adding other ethernets to use master port for switch causes hideous performance drop.

Indeed I do have DHCP enabled on the bridge. Heres the export: [admin@MikroTik] > export compact # feb/08/2012 11:37:30 by RouterOS 5.12 # software id = PIXE-94VZ # /interface bridge add l2mtu=1520 name=Local /interface wireless set 0 band=2ghz-b/g/n disabled=no frequency=2452 l2mtu=2290 mode=ap-bri...

I was pretty confident it was set up correctly, and from a pure ping test point of view it 'worked', its the performance I was concerned about. All else being equal, how can a software bridge perform *far far* better than a hardware switch chip? Ether1 has masquerade and a pretty light set of firewa...

I'm trying to figure out exactly *why* I'm getting a non-intuitive performance result. I have an RB493G (5.12) servicing a small office. Ether 1 is WAN. Trying to get all 8 other ports + wifi operating as a single subnet. So I switched 3,4,5 to 2 and 7,8,9 to 6. Then I created a bridge and added 2, ...

Good QOS is pretty important for business networks. And on good stable WAN connections this works very well. Unfortunately in regional Australia good stable WAN connections are in very short supply. Right now my 3G connection is fluctuating within a 20% range - so either I lose tight QOS when connec...

Bryron Bay preferred. I'd make it to Brisbane.

Okay, I managed to get it. Create a bridge interface, desired assign source IP to it and then route across VPN from that address. The steps are: Create IPSec peer /ip ipsec peer print address=YYY.YYY.YYY.YYY/32:500 auth-method=pre-shared-key secret="superstrongpwd" generate-policy=no excha...

I have a reasonably complex routing requirement. I've made it work on Cisco, I've made it work on Snapgear but I can't nut it out on Mikrotik. Side A (source) has public IP address XXX.XXX.XXX.XXX and private LAN AAA.AAA.AAA.AAA/24 . Side B (destination) has public IP YYY.YYY.YYY.YYY and private LAN...