Community discussions

MikroTik App
 
hashbang
Member Candidate
Member Candidate
Topic Author
Posts: 156
Joined: Sat Jul 26, 2014 6:38 pm

bonding on ethernet with pppoe traffic

Tue Mar 24, 2020 9:20 am

Hi,
What hashing method to be used on bonding 802.3ad for pppoe encapsulated traffic. I'd tried l2 l2+3 l3+4 but the traffic doesnt balance equally or so. Its something like 90%+10%.
thanks
 
Zacharias
Forum Guru
Forum Guru
Posts: 1603
Joined: Tue Dec 12, 2017 12:58 am
Location: Greece

Re: bonding on ethernet with pppoe traffic

Tue Mar 24, 2020 12:29 pm

802.3ad supported hashing methods are only Layer 2 or ( Layer2 and Layer3 )... Also MII link monitoring is recommended ....
Also LACP does not break/balance connections, so you will not see traffic 50-50...
https://wiki.mikrotik.com/wiki/Manual:I ... ng#802.3ad
 
hashbang
Member Candidate
Member Candidate
Topic Author
Posts: 156
Joined: Sat Jul 26, 2014 6:38 pm

Re: bonding on ethernet with pppoe traffic

Tue Mar 24, 2020 12:52 pm

ty,
but when i create lag between huawei(switch) to huawei (router). It works fine but between huawei switch to mikrotik it doenst do load balancing. Any other method to make it working. Balance RR dont work.
 
Zacharias
Forum Guru
Forum Guru
Posts: 1603
Joined: Tue Dec 12, 2017 12:58 am
Location: Greece

Re: bonding on ethernet with pppoe traffic

Tue Mar 24, 2020 1:48 pm

802.3ad does not do load balancing anyways.. it divides traffic accross links..
As said earlier it will not break a connection and balance it across the 2 links...
I do not know how Huawei works... But if it supports 802.3ad you can use this mode... it will do what it has to do just fine...
 
hashbang
Member Candidate
Member Candidate
Topic Author
Posts: 156
Joined: Sat Jul 26, 2014 6:38 pm

Re: bonding on ethernet with pppoe traffic

Wed Mar 25, 2020 9:43 am

TY!
there are 1000 pppoe connections that means 1k mac addr. I suspect missing modes of bonding in MT. So this problem only comes in encapped traffic like pppoe
encap2+3

This policy uses the same formula as layer2+3 but it
relies on skb_flow_dissect to obtain the header fields
which might result in the use of inner headers if an
encapsulation protocol is used. For example this will
improve the performance for tunnel users because the
packets will be distributed according to the encapsulated
flows.

encap3+4

This policy uses the same formula as layer3+4 but it
relies on skb_flow_dissect to obtain the header fields
which might result in the use of inner headers if an
encapsulation protocol is used. For example this will
improve the performance for tunnel users because the
packets will be distributed according to the encapsulated
flows.

Who is online

Users browsing this forum: faxxe, matiss, MicpaH, MSN [Bot], scanner and 202 guests