3com S7906E Manuel De Montage

 

1-18 

are received, the finwait timer is reset upon receipt of the last non-FIN packet and the connection is 

terminated after the finwait timer expires. 

z 

Size of the IPv6 TCP sending/receiving buffer. 

Follow these steps to configure IPv6 TCP properties: 

Enter system view 

— 

Set the finwait timer 

tcp ipv6 timer fin-timeout 
wait-time 

Optional 

675 seconds by default. 

Set the synwait timer 

tcp ipv6 timer syn-timeout 
wait-time 

Optional 

75 seconds by default. 

Set the size of the IPv6 TCP 
sending/receiving buffer 

tcp ipv6 window size 

Optional 

8 KB by default. 

 

If too many ICMPv6 error packets are sent within a short time in a network, network congestion may 

occur. To avoid network congestion, you can control the maximum number of ICMPv6 error packets 

sent within a specified time, currently by adopting the token bucket algorithm. 

You can set the capacity of a token bucket, namely, the number of tokens in the bucket. In addition, you 

can set the update interval of the token bucket, namely, the interval for restoring the configured capacity. 

One token allows one ICMPv6 error packet to be sent. Each time an ICMPv6 error packet is sent, the 

number of tokens in a token bucket decreases by one. If the number of ICMPv6 error packets 

successively sent exceeds the capacity of the token bucket, the additional ICMPv6 error packets cannot 

be sent out until the capacity of the token bucket is restored. 

Follow these steps to configure the capacity and update interval of the token bucket: 

Enter system view 

— 

Configure the 
capacity and update 
interval of the token 
bucket 

Ipv6 icmp-error { bucket 
bucket-size | ratelimit 
interval } * 

Optional 

By default, the capacity of a token bucket is 10 
and the update interval is 100 milliseconds. 
That is, at most 10 IPv6 ICMP error packets 
can be sent within 100 milliseconds. 

The update interval “0” indicates that the 
number of ICMPv6 error packets sent is not 
restricted. 

 

If hosts are configured to answer multicast echo requests, an attacker may use this mechanism to 

attack a host. For example, if Host A sends an echo request with the source being Host B to a multicast 

address, then all the hosts in the multicast group will send echo replies to Host B. Therefore, to prevent 

such an attack, a device is disabled from replying multicast echo requests by default.