Anyone seen this before? I spent a week trying to track down the source of packet loss. I have a chain of 4 towers all linked together with Airfiber 5x’s and an RB 3011 at each site connected via OSPF. I was seeing 1% packet loss on each link, so 4% by the time we got to the end of the chain. I spent most of my time troubleshooting airfibers thinking they were the problem, but I got desperate and started trying random things. Turning off the LCD displays on each router solved the problem! Once I disabled them the packet loss instantly went away, and I mean instantly! I’m wondering if the display doesnt emit some type of interference into the ethernet ports. I had mine set to always on with no backlight timeout and they were set to change every 1 min to informative slide show.
I can turn the lcd back on and within a few pings causes loss again.
There must be more to it. I have LCD enabled on my RB2011 and I can easily get 80 Mbps (the speed of my line), much more than the 32 Mbps that topic indicates.
In my case I could push up to 300 megs on that link during testing and never had a bandwidth problem, just a packet drop problem. If turning the LCD off is the trick that’s good enough for me, but man I should have posted this a week ago lol!
Im going to to a big, new thread on this tomorrow, but after 5 days of trying to figure out why only a Rb3011 was showing weird BW speed issues, the fix was to DISABLE THE LCD SCREEN.
Im talking BW issues, during which CPU load is maxing at 30-40% tops, so not high load or high BW… but fix was to DISABLE THE LCD SCREEN.
I have been tracking down a few reports within our network due to 1-2% packet loss.
Isolated the issue to only RB3011UiAS devices. After finding this thread, I disabled the LCD on one unit, and it immediately resolved the packet loss issue. Disabled LCD on 15 or so other units, also resolved the issue.
Running 6.42.3 on all of the affected devices. We will be deploying a mass update to our network disabling the LCD on all RB3011s.
In our network, it is now considered a KNOWN bug. I hope this helps others!
Same problem here. Had to go through our entire 30km network to come to the 3011 to find that the issue was only occurring with the 3011.
Once I knew the problem was with the 3011 google led me here. Thanks a ton for this, whoever found it. I was suspecting it but couldn’t exactly place the finger on the LCD.
Hope they release a fix on the bugfix branch. running 6.40.8 (bugfix).
as i posted above (about a year or 2 ago) i fully confirm this issue on seveal different RB2011 and rb3011s (and one other RB with an LCD screen that im forgetting). fix is always to disable/turn off LCD screen (which i do on all devices now, unfortunately)
I can say that on a few CCR1009-7G-1C-1S+ this issue does NOT occur (granted CCR is entirely different hardware set and much more powerful). at many sites we do latency and service monitoring (graphing) of external locations, so its pretty easy / quick to see when something is off (ie its a red flag when your 1st hop / default GW is showing latency and small packet loss)
(