We are seeing excessive Rx Drops and Rx Errors on a RB750 when using multilink pppoe using two ADSL modems. The counters increment slowly when a small amount of traffic is being passed and much faster when the link is saturated. The Rx Drop and Rx Error values always match.
Although we are seeing these errors, we do not see any degradation in performance or any packet loss. This happens on both 4.17 and 5.4 versions of the RouterOS. I’m wondering what these errors actually are. We have tried several MTU/MRU/MRRU configurations and the same thing happens. Any insight in this matter would be appreciated. Thanks!
Edit:
Could these increments be a result of re-ordering in multilink ?
i am assuming that ADSL connection is to ISP, does this ISP supports PPPoE bundling by MLPPP?
Post output of /ip address, /interface pppoe-client monitor [find] and other related configs.
Does anyone have the answer to this as i have exactly the same issue and the ppoe becomes unstable and disconnects. If i use a single connection the issue goes away.
It’s a little late to reply to the OP, but the MTU is always 1524, as that value is forced upon the Routerboard from the LNS.
I guess you are using Cisco equipment to terminate the MLPPP connection, as 1524 is the Cisco default
value for MLPPP. You should be able to change it on your LNS; but not the fact that this value is always
going to override the settings on the Routerboard.
Regarding your problem - can you please give some more detail on configuration and devices used?
We provide RB750’s to our customers who want to use multi-link with us. They are usually set up with 2-4 modems. Depending on the customer needs they will get a dynamic IP, static IP or a static IP with a /29 subnet routed to it. We have started using a MTU/MRU of 1454 as some of the sessions may pass through a L2TP tunnel at some point.
I usually setup port 1 as the “LAN” port which connects to the customers network. Ports 2-5 are for DSL modems. Each of these ports are set to not have a master port. NAT is used on dynamic and static IP’s, but turned off when the customer is given a subnet and routes are installed as needed.
In all cases performance is good and the Rx Drops and Rx Errors keeping going up and I’m not sure why. I suspect re-ordering (which is fine) but want to be sure.
Our RBs experience the same effect - a lot of RX errors on the pppoe-out interface (depending on load),
without any detectable impact on service quality (except maybe small latency jumps).
Therefore i tend to agree with your opinion, it’s probably related to packet reordering.
we use MLPPP two and on Mikrotik to Cisco LNS you have to Set the Link Speed
In the PPPoE Profile for UP -and Download, and uses TCP-MSS Rule to Set right MTU,
Then it will work for us!
we use MLPPP two and on Mikrotik to Cisco LNS you have to Set the Link Speed
In the PPPoE Profile for UP -and Download, and uses TCP-MSS Rule to Set right MTU,
Then it will work for us!
Hi,
I also work for an ISP and do the majority of the config for the mikrotiks we ship out. Usually rb750gl’s. We set the mtu to 1480 but find that this is no good if the client wants to add VPN to it as well as this can jack the packet size up over the 1500 limit when combined with the encryption added via mlppp. Can you please further explain what it is you are saying here as it isnt all that clear. I have a few clients that are getting random drops via their Mikrotiks and am looking for anything I can to isolate the issue.
Hi can I get you to clarify this a bit for me? I work for an ISP and we use the 750GL for bonding mlppp accounts. I’m getting multiple reports from clients that the pppoe session is dropping on the mikrotik but the modems never lose sync. Think this may be the solution.