RB1200: High CPU Usage results in ping loss

RouterOS General
1

Hi All

As a main router we are using a RB1200. I’ve connected Port 6 to my ISP and Port 7 to our main switch. Setup a bridge and bridged port 6&7.

When one of the servers connected to the switch generates quite a bit of bandwith (in example when i’m downloading from my servers i’'m downloading with 120mbit/second). CPU usage on the RB1200 goes upto 100%.

When pinging the servers and the RB1200 directly i can see (while the big bandwith is happening) loads of ping timeouts, it could even result in some servers not being reachable at all.

Is there something which i can do to prevent this from happening (i.e. some script which loweres the amount of bandwidth if cpu usage hits 100%)?

2

It does not seem normal to me to get to 100% with 120Mbps bandwith, have you used the “profile” tool to check how those 100% cpu are used ?

3

In this screeny i’m pinging from a business dsl connection (its a bit sloppy, i know) to 2 different hosts
Where
… 70.132 is the IP of my RB1200
… 6.8 a random address on the internet to check latency (ping.xs4all.nl)

54.5% queuing
22.5% firewall
12.0% briding

I’ve checked the Queue list section, both Simple Queues and Queues tree’s are blank
On the interface queue section: all ports are set to ‘only-hardware-queue’

Thanks!

4

Try using only ports 1 through 5 instead of 6 through 10.

5

1 to 5 are switch ports, firewall policy’s won’t apply.

Ps. There are known issues with 9&10, i’m not using them at all