After upgrading to 1.4g, I’m having problems with my connection speed on Route 10. My connection is 900Mbps, and until recently, it was that high, but for some time now I haven’t been able to get more than 400Mbps, and today it’s even down to 200Mbps. I’ve been in contact with my internet provider about this since last week, but they claim everything is fine. Today, I restarted Route 10 for a test because it wasn’t showing the temperature; it was 0°C. After restarting, the speed returned, but not for long after about an hour, it started to drop again. I’m slowly getting fed up…
I’d suggest a speed check from the Route10 to narrow it down further. Also, perform a speed check from a WiFi device (with a connection that would saturate the nominal ISP speed).
Maybe you have already seen the post by @dalewhlrr, and if not check here for setting up and running speedtest on Route10 through SSH+CLI:
Also, as a further test, set up Route10 as an iperf3 server and check from the PC as client, just to make sure there isn’t something coincidentally fishy with cables or the PC network card.
Alta-Josh and ebuckland81 ok so,
tests 1. speedtest on router10 via CLI show 800-880 Mb/s
tests 2. speedtest on PC via wire shows 300-400Mb/s (same endpoint)
tests 3. iperf3 PC to Route10 shows 900Mb/s
Can you continue to test and monitor for fluctuations throughout the day? Your tests so far indicate that the issue is upstream with your service provider.
Can you reproduce the issue from other wired clients? What about from other Route10 interfaces? Are there firewall rules or VLAN configurations in place that may be impacting your network?
Please review the hardware acceleration options noted by @ebuckland81 above.
Our team also suggested that you may try disabling IDS/IPS to test. Make sure you do not change everything at the same time; instead, test each change independently, one at a time, to ensure there is no mixup of the potential impact.
Okay, so disabling hardware acceleration improves the results. I can achieve around 750 Mb/s, and that would be fine. Additionally, if I disable the limiting (900) on the WAN, the results improve to 830 MB/s.
With hardware acceleration enabled, it doesn’t matter whether WAN limiting is present or not; the results are the same, around 400 MB/s. It also doesn’t matter whether it’s port 2, 3, or 4, or whether it’s the main network or a VLAN.
The improvement in results only occurs when hardware acceleration is disabled. The only thing I’ve noticed is that when limiting is set, tests performed from CLI Route10 are more stable, but this has no impact on computers connected behind Route10.
small correction, several minutes after leaving hardware acceleration disabled, the results from the internal network are again around 400Mb/s, tests with CLI are ok, i.e. around 850Mb/s
Forgive me if this has been answered and I just missed it in the other posts, but how is everything connected to the Route10? I’m just curious which port you ISP is connected to and which port your PC has been connected to.
Otherwise, just in the interst of saving time, I would think it might be worthwhile to either reset the Route10 and re-adopt it to your site or create a new site as a test and move the Route10 over to it. Either could be worthwhile to make sure things are as close to default as possible and there isn’t some kind of strange setting or config that’s misbehaving deep down in the router.
ISP1 main → Route10(wan1)
ISP2 backup → Route10 (wan2)
Main LAN on LAN4
Vlans on lan2 and 3
AP6Pro
/
ISP1 ----------------- S8
\ / \
Route10 -> Main switch AP6
/ | \
ISP2 Other serwers POE SWITCH ---- USW Lite ------ Main PC
something like this
Don’t get me wrong, resetting the device isn’t a fix. Of course, it’s possible, but since configuring it with your own controller is, to put it mildly, tedious, I don’t want to do it.
That’s why I’m writing on the manufacturer’s forum, to find out how we can identify the problem.
Fair enough! Wasn’t sure if the goal was “just make it work” or “find the root cause and properly fix it”.
I’m just some guy on a forum, but here’s some other thoughts! Sorry if you tried any of these and I just didn’t catch it further up the thread.
Since this happened once the 1.4g update came through, it might be worthwhile to click off DPI for the router since the switch to do so is now available and that was the big feature added to that update. Probably set the hardware acceleration back to on as that didn’t have a lasting effect and would be on normally.
The 1.4g update also came with some multi-wan fixes, so another thought would be to only have your main ISP connected to make sure there isn’t something misbehaving with where the router is sending traffic. It kinda sounds like speed to the router is pretty consistent but forwarding through the WAN is where things fall apart.
If we want to get a little extra wild we could go to the terminal and run uci export which should spit out most of the routers internal config for review.
