>
> Friends,
>
> This is my first toll.
>
> We're leaving mikrotik as a pppoe server. We are testing debian 8 64bits with the configuration that I will inform below.
>
> We are running quagga with IP summarization, accel-ppp and ipv6.
>
> We have 3,000 pppoe sessions. The problem is that the cpu gets from 6% to 9% and suddenly it goes on for a few seconds to 90% to 100% dropping the accel-ppp dropping all users and then back to normal.
>
> What can I do to improve? These peaks are with large intervals, sometimes 3 hours pass to happen.
>
> Server Configuration:
>
>
> Xeon v3 3.4ghz 8MB quad core
> 32GB ddr3 1600MHZ
> 2x 3TB SAS 6GB / s raid 1 via off-board controller Intel 1GB cache.
> 8 gigabit intel ports but we are only using 4 making 2GB / s LACP bonding for WAN and 2GB for LAN.
>
> Spending on average 1.1GB and 1.5GB / s of traffic at peak times. Traffic is not the problem because today happened with 500MB of traffic.
>
>
>
> Fernando Galvão
> Wantel Telecom
> +55 87 3866-5200
>
>
>
> Friends,
>
> This is my first toll.
>
> We're leaving mikrotik as a pppoe server. We are testing debian 8 64bits with the configuration that I will inform below.
>
> We are running quagga with IP summarization, accel-ppp and ipv6.
>
> We have 3,000 pppoe sessions. The problem is that the cpu gets from 6% to 9% and suddenly it goes on for a few seconds to 90% to 100% dropping the accel-ppp dropping all users and then back to normal.
>
> What can I do to improve? These peaks are with large intervals, sometimes 3 hours pass to happen.
>
> Server Configuration:
>
>
> Xeon v3 3.4ghz 8MB quad core
> 32GB ddr3 1600MHZ
> 2x 3TB SAS 6GB / s raid 1 via off-board controller Intel 1GB cache.
> 8 gigabit intel ports but we are only using 4 making 2GB / s LACP bonding for WAN and 2GB for LAN.
>
> Spending on average 1.1GB and 1.5GB / s of traffic at peak times. Traffic is not the problem because today happened with 500MB of traffic.
>
>
>
> Fernando Galvão
> Wantel Telecom
> +55 87 3866-5200
>
>
>