Did a speedtest yesterday for the first time in a while, and got 50Mb/s on my 750Mb/s service. I repeated it a few times with different servers same 50Mb/s. Rebooted my router and it came good.
So my question is what would cause that to happen.
Crap Speedtest but good after reboot
Moderator: Moderators
Re: Crap Speedtest but good after reboot
Can't suggest any likely cause as my Gargoyle devices have never exhibited this sort of behaviour even when left running continuously for as long as 6 months, but actions you could take which might help diagnose a possible cause if/when it happens include:
- check system state for exhausted memory and/or unusually high load averages (i.e. greater than about 80% of the total number of CPU cores)
- check system log for error messages
- run top to check for unusual process activity
- check system state for exhausted memory and/or unusually high load averages (i.e. greater than about 80% of the total number of CPU cores)
- check system log for error messages
- run top to check for unusual process activity
Re: Crap Speedtest but good after reboot
Most likely cause: conntrack table exhaustion. On a 750 Mb/s link you churn through TCP connections fast. The netfilter connection tracking table has a fixed size (nf_conntrack_max), and once it's full the kernel starts dropping or
refusing new connections. Speed tanks, reboot clears the table, everything's fine again. Classic symptom pattern.
To confirm: cat /proc/sys/net/netfilter/nf_conntrack_count vs cat /proc/sys/net/netfilter/nf_conntrack_max — if they're close, that's the answer.
refusing new connections. Speed tanks, reboot clears the table, everything's fine again. Classic symptom pattern.
To confirm: cat /proc/sys/net/netfilter/nf_conntrack_count vs cat /proc/sys/net/netfilter/nf_conntrack_max — if they're close, that's the answer.