Traffic spike fix from 384.14_2 was reverted in 386.1

[ech]

Nothing new here? Should we just give up on the built-in traffic statistics?

I'm still hoping to test a fixed/test rstats on an RT-AC68U.

 

[Morac]

I stopped getting spikes about 5 days ago. I don’t know why.

 

[Wistuplu]

Amazing, Morac. Earlier tonight, I could get a ghost spike within 30 seconds or so, as usual. Your feedback made me want to test again.

I have just finished installing cake-qos (which I used to run before 386) on AC86U with 386.1 and I'm not getting a spike as fast as usual. Note that, as a side effect of cake-qos, I currently have:

​

HW acceleration	Runner: Disabled - Flow Cache: Disabled

It is probably a "false positive" because it is a low traffic time now, and a very short test period. But who knows...
BTW: it might impact my ability to reproduce the issue and test a patch...
Take care

 

[Morac]

I went and checked again and I did get another spike today. I upgraded to 386.1_2 today (in a very round about way since I couldn’t get the upgrade to work through the GUI). Prior to that I had an uptime of about 8 days. The first 2 days I got spikes, and then they stopped. Maybe they stabilize after a certain amount of uptime.

 

[ech]

Still getting spikes daily here - multiple. I doubt that since the problem appears to be a wraparound issue that they'd simply go away over time.

And even if they did, simply having them every reboot would effectively kill/swamp out the normal data represented on the graph for any period that the spikes are collated into, so without a fix, every time you reboot, you'd have worthless data until the spikes fully aged out.

 

[Wistuplu]

FYI, on my AC86U, with 386.1 (not 386.1_2), I currently get

* one or more spikes within approx. 5 minutes without cake-qos, so with hardware acceleration ON (Runner and Flow Cache enabled)

* no spikes within 10 minutes with cake-qos, so with hardware acceleration OFF (Runner and Flow Cache disabled)

I have uninstalled and reinstalled cake-qos, and it confirmed the difference in behaviour.
Regards

 

[ech]

Well, since I can't use cake-qos (ac68u not supported) but I am using FlexQOS, shouldn't my hw accel also be disabled? I'm still seeing the spikes with this config/hardware.

 

[Wistuplu]

The case of my AC86U shows that different settings could impact ghost spikes. The hardware acceleration could be a good candidate for a first test indeed.

 

[ech]

Don't think there's anything else to test, given this setup, is there?:

 

[Phil Outram]

Perhaps there is a better solution for you, like an isp that doesn't limit data at all? There is really no reason for an isp to be limiting downloads or uploads these days.

 

[ech]

That's not why I'm using the traffic data - I just like to have the data as a nice to have... if it's broken, it doesn't really matter to me, but it would be nice if it worked.

Guess I'll probably switch to the alternate (not part of merlin) traffic data stats solution mentioned earlier in this thread, since the built-in feature is broken again(and probably will break in the future as well).

 

[Wistuplu]

Hello,

Asus offered me to test a new Beta (or is it rather alpha? ;-) ) version for AC86U, supposedly adressing, among others, the spikes issue:
RT-AC86U_9.0.0.4_386_42210-g5edd47f_cferom_ubi.w.

As I'm happy with my current clean Merlin 386.1 setup with Cake, spdMerlin and connmon, I'm not sure I'll switch back to "Official" for the test. If you do test, please let me know.

https://www.asuswebstorage.com/navigate/a/#/s/010C684C55124C23927EF086BA4D693A4

Best regards

 

[ech]

Hello,

Asus offered me to test a new Beta (or is it rather alpha? ;-) ) version for AC86U, supposedly adressing, among others, the spikes issue:
RT-AC86U_9.0.0.4_386_42210-g5edd47f_cferom_ubi.w.

As I'm happy with my current clean Merlin 386.1 setup with Cake, spdMerlin and connmon, I'm not sure I'll switch back to "Official" for the test. If you do test, please let me know.

https://www.asuswebstorage.com/navigate/a/#/s/010C684C55124C23927EF086BA4D693A4

Best regards

I would test an AC68U firmware if they had one available to test.

 

[Crazy]

I’m on RT-AC5300 and I also see the spike on 386.1

 

[Wistuplu]

Hello,

FYI, RT-AC86U_9.0.0.4_386_42210-g5edd47f_cferom_ubi.w seems to solve the issue with my AC86U. No ghost spikes after approx. 20 hours, and plausible totals.

Kudos to the development and service desk at Asus

Take care
W.

PS: Of course 9.0.0.4_386_42210 is not Merlin but I guess it will trickle down some future Merlin release.

 

[palirobe]

SOLUTION that worked for me: I'm on RT-AC88U. Traffic Analyzer on version 384.19 worked well. After upgrading to 386 & enabling QOS I saw enormous traffic spikes (about 17 billion gig every 3 hours). After blocking internet to my devices (one at a time), I discovered the offending device was my AirTV. AirTV is a device that broadcasts OTA antenna channels via the Sling App. The AirTV device errored/spiked whether it was a WIFI connection or a wired connection. My work-around was simple. I changed the AirTV to have the highest priority under QOS & the traffic spikes stopped. I'm not sure why they stopped, but this change worked in adaptive, traditional & bandwidth limiter options. Traffic Analyzer is working normally for me now. If others on this forum can find the offending device(s) on their routers, I'm hoping that changing the priority on the offending device will fix the traffic spikes for them as well.

 

[Morac]

I’ve found that for some reason for about 3 or 4 days after the router is rebooted, there will be these spikes. After that point, the spikes go away until the router is rebooted again. My router has been up for 15 days and I haven’t had a spike in 11 days.

I did restart the WAN service and that didn’t trigger the spikes, so there’s something about rebooting the router that causes it to measure traffic incorrectly for the first few days.

 

[palirobe]

Unfortunately for me, the problem didn't go away after a few days. It continued every day, every three hours triggered by my AirTv which gave some astronomical spikes in data.

 

[Wistuplu]

After blocking internet to my devices (one at a time), I discovered the offending device was my AirTV. AirTV is a device that broadcasts OTA antenna channels via the Sling App. The AirTV device errored/spiked whether it was a WIFI connection or a wired connection.

Hello,
Thank you for sharing. It is very unexpected to me that this could be related to one or more 'special' devices. I might test this in my context. How did you block the Internet during your tests? I can think of many ways and I'd like to use the same as yours.
Thank you
Take care
W.

 

[Wistuplu]

I can consistently create a ghost spike when switching my Android 9 phone to "Airplane mode" (Huawei Mate 9).
I also got a spike when restarting my Microsoft Surface Pro 7 with WIN10 (and Cisco AnyConnect).
Both were connected in 5GHz. (There were no ethernet or 2.4GHz connections).
Really amazing. So, seemingly unrelated to reaching some maximum volume of traffic.