Flint3 frequent internet disconnect

ukuser4489 · August 7, 2026, 7:24am

Hi all,

Got a flint3 a few weeks ago to replace ISP router.

Firmware is 4.9.0

Every few days the internet connectivity dies. Happened again yesterday and i end up rebooting to get everything going again.

I noted this in the log:

Thu Aug 6 10:37:37 2026 kern.warn kernel: [196536.582709] nf_conntrack: nf_conntrack: table full, dropping packet Thu Aug 6 10:37:37 2026 kern.warn kernel: [196536.658491] nf_conntrack: nf_conntrack: table full, dropping packet Thu Aug 6 10:37:37 2026 kern.warn kernel: [196536.847914] nf_conntrack: nf_conntrack: table full, dropping packet Thu Aug 6 10:37:38 2026 kern.warn kernel: [196537.363930] nf_conntrack: nf_conntrack: table full, dropping packet Thu Aug 6 10:37:38 2026 kern.warn kernel: [196537.420479] nf_conntrack: nf_conntrack: table full, dropping packet Thu Aug 6 10:37:38 2026 kern.warn kernel: [196537.777698] nf_conntrack: nf_conntrack: table full, dropping packet

Not an expert on conntrack at all but from what i can gather, the “dying” (all seem to be TCP connections from a range of ips on the lan) seems to constantly accumulate and i suspect this is the root of the problem.

08:16:09 total=8723
dying=8120
08:16:15 total=8739
dying=8120
08:16:20 total=8748
dying=8120
08:16:26 total=8504
dying=8136
08:16:31 total=8515
dying=8136
08:16:37 total=8524
dying=8136
08:16:42 total=8528
dying=8136
08:16:48 total=8534
dying=8136

Not sure what causes this or how to diagnose and resolve - or even if i’m on the right path.

Any thoughts or pointers would be welcomed!

Thanks

EDIT: After a little tweaking i found that disabling HW acceleration stops these mounting up - so i’m coming to the conclusion this is some weird HW acceleration behaviour. Still not clear on root cause or fix/workaround as i require HW accel enabled in regular use.

charles2 · August 10, 2026, 3:21am

Hi,

Thank you for the additional testing and for sharing this observation.

Since Hardware Acceleration changes the packet forwarding path, this behavior change may provide a useful direction for further investigation.
Before confirming the root cause, we would like to further check the current conntrack usage and identify whether a specific LAN device is creating an unusually large number of connections.

When Hardware Acceleration is enabled and the issue happens again, could you please run the following commands via SSH and share the results with us?

cat /proc/sys/net/netfilter/nf_conntrack_count
cat /proc/sys/net/netfilter/nf_conntrack_max

In addition, could you please run the following command to help identify which LAN device is using the most conntrack entries?
cat /proc/net/nf_conntrack | grep "src=192.168." | awk -F'src=' '{print $2}' | awk '{print $1}' | sort | uniq -c | sort -nr | head -20

After collecting the command outputs, could you please send the results together with the system log to us via private message? This will help us further analyze the behavior when the issue occurs.

We will continue investigating based on the information provided.

Thank you for your understanding and cooperation.