Problems on PPPoE over different Beta/Stable versions

natanielklug · Mon Apr 22, 2013 3:44 pm

Hello all,

I am running Mikrotik as a PPPoE Concentrator at my network border and I get all customers here using EoIP. I have not noticed any problems regarding EoIP tunnels but I will explain all the problems I got (this all where over RB1100AHx2 hardware):

v5.24 (stable)
PPPoE got a very strange latency going over 200ms for all PPPoE sessions. Could not understand why and I only saw something like this on Mikrotik running over x86 machines. I can't use this stable on RB1100AHx2.

v6.0rc11
RB reboots but could not determine why. If I run the same version over a RB1100AH not problem is found. Over RB1100AHx2 when some event drops PPPoE Sessions RB reboots or stop responding. I am attaching a supout.rif from the last reboot and I could not determine what caused the problem but last problem in this RB1100AHx2 was from saturday to sunday at 01h40 am. I had not problem over my network by that time and I only noticed the problem on sunday at 02h30 pm because another router in my network (RB750GL running v5.24) suddenly reboots and got a lot of OSPF neighbour 2-Way and Down from RB1100AHx2.

v6.0rc12
RB reboots from time to time. My time span was something between 6 to 8 hours running.

v6.0rc13
After some PPPoE Sessions got connected some of them became "Dynamic" but not "Running" and the interface was not stable.

If is there another v6.0rcXX to test send it to me and I will gladly test it. Hope to solve the problem and my network start to flow well.

honzam · Mon Apr 22, 2013 4:47 pm

Try pre-release 6.0rc14 http://www.mikrotik.com/download/share/ ... .0rc14.zip

*) improved PPPoE interface encapsulation performance;

natanielklug · Mon Apr 22, 2013 4:58 pm

Try pre-release 6.0rc14 http://www.mikrotik.com/download/share/ ... .0rc14.zip

*) improved PPPoE interface encapsulation performance;

honzam,

I will try and return in the forum. Thank you.

gallysoft · Tue Apr 23, 2013 10:44 am

Hi natanielklug.
Did you try the new version? It works?

Thank you.

natanielklug · Tue Apr 23, 2013 2:02 pm

Hello all,

Almost 24 hours running v6.0rc14 for tests and, untill now, no reboot from my RB1100AHx2. I am running a PPPoE Concentrator Cluster and for now I have 3 machines (2 RB1100AHx2 and 1 x86 Dell hardware). Just for understanding:

CSC-auth-08 = RoS v6.0rc14
CSC-auth-09 = RoS v6.0rc11
CSC-auth-12 = Ros v5.24

As you can see I have about 600 PPPoE sessions right now and they are all working. About 30 minutes ago I put CSC-auth-09 and CSC-auth-12 to run (as in my first scenario where RB1100AHx2 stop responding OSPF for some unknow reason). So I will be waiting for the scenes of next episodes. For now it's all working and I hope it keep this way.

gallysoft · Tue Apr 23, 2013 2:38 pm

Thank you.
Please, update if you have news, because I have to migrate from a Cisco 7200 to a RB1100AHx2, but I will do it only when it seems stable.

natanielklug · Wed Apr 24, 2013 1:58 pm

Quick Update:

1d17h running RB1100AHx2 v6.0rc14 no problem found.

PS.:
I had a problem over OSPF where many interfaces got "discarding locally originated packet" and running version 6.0rc14 this message is not showing anymore.

natanielklug · Thu Apr 25, 2013 2:33 pm

And the problems continue...

I said my dead line was something between 2nd and 3rd day. The problem is that I got to the company and looking into one of my PPPoE Concentrator I saw that just some customers were there (about 20 customers) and it should have +300 PPPoE Sessions.

Looking into logs I found the problem, for some reason, it's now loosing OSPF neighbours and it's dropping sessions. The strange part is that I am now running 3 concurrent PPPoE concentrators and they are 2 RB1100AHx2 (v6.0rc14) and 1 x86 (v5.24). They are all in the same LAN segment so they share OSPF. In RB1100AHx2 I have this logs:

For CSC-auth-08

Apr/25/2013 03:58:23 pppoe,info PPPoE connection established from 00:27:22:76:3A:FF
Apr/25/2013 03:58:23 pppoe,ppp,info <pppoe-0>: waiting for call...
Apr/25/2013 03:58:46 pppoe,ppp,error user ssilva@certto.com.br authentication failed
Apr/25/2013 03:58:46 pppoe,ppp,info <pppoe-0>: terminating... - user ssilva@certto.com.br authentication failed
Apr/25/2013 03:58:46 pppoe,ppp,info <pppoe-0>: disconnected
Apr/25/2013 03:58:54 pppoe,info PPPoE connection established from 00:27:22:0C:63:A4
Apr/25/2013 03:58:54 pppoe,ppp,info <pppoe-0>: waiting for call...
Apr/25/2013 03:59:00 pppoe,ppp,info <pppoe-0>: terminating... - authentication failed
Apr/25/2013 03:59:00 pppoe,ppp,info <pppoe-0>: disconnected
[b]Apr/25/2013 03:59:37 route,ospf,info OSPFv2 neighbor 172.16.240.25: state change from 2-Way to Down
Apr/25/2013 04:00:27 route,ospf,info OSPFv2 neighbor 172.16.240.25: state change from 2-Way to Down
Apr/25/2013 04:00:46 route,ospf,info OSPFv2 neighbor 172.16.240.25: state change from 2-Way to Down
Apr/25/2013 04:00:50 route,ospf,info OSPFv2 neighbor 172.16.240.25: state change from Full to Down
Apr/25/2013 04:00:51 route,ospf,info OSPFv2 neighbor 172.16.255.14: state change from Full to Down
Apr/25/2013 04:00:51 route,ospf,info OSPFv2 neighbor 172.16.255.23: state change from Full to Down
Apr/25/2013 04:00:53 route,ospf,info OSPFv2 neighbor 172.16.255.22: state change from Full to Down
Apr/25/2013 04:00:54 route,ospf,info OSPFv2 neighbor 172.16.255.11: state change from Full to Down
Apr/25/2013 04:00:55 route,ospf,info OSPFv2 neighbor 172.16.240.28: state change from Full to Down
Apr/25/2013 04:00:55 route,ospf,info OSPFv2 neighbor 172.16.240.28: state change from Full to Down
Apr/25/2013 04:00:55 route,ospf,info OSPFv2 neighbor 172.16.240.28: state change from Full to Down[/b]
Apr/25/2013 04:01:56 pppoe,info PPPoE connection established from 00:27:22:0C:63:A4
Apr/25/2013 04:01:56 pppoe,ppp,info <pppoe-0>: waiting for call...
Apr/25/2013 04:01:56 pppoe,ppp,error user dharmel@certto.com.br authentication failed
Apr/25/2013 04:01:56 pppoe,ppp,info <pppoe-0>: terminating... - user dharmel@certto.com.br authentication failed
Apr/25/2013 04:01:56 pppoe,ppp,info <pppoe-0>: disconnected
Apr/25/2013 04:02:46 pppoe,info PPPoE connection established from 00:27:22:76:3A:FF
Apr/25/2013 04:02:46 pppoe,ppp,info <pppoe-0>: waiting for call...
Apr/25/2013 04:02:46 pppoe,ppp,error user ssilva@certto.com.br authentication failed
Apr/25/2013 04:02:46 pppoe,ppp,info <pppoe-0>: terminating... - user ssilva@certto.com.br authentication failed
Apr/25/2013 04:02:46 pppoe,ppp,info <pppoe-0>: disconnected

For CSC-auth-09

Apr/25/2013 03:43:48 pppoe,ppp,error user ssilva@certto.com.br authentication failed
Apr/25/2013 03:43:48 pppoe,ppp,info <pppoe-0>: terminating... - user ssilva@certto.com.br authentication failed
Apr/25/2013 03:43:48 pppoe,ppp,info <pppoe-0>: disconnected
[b]Apr/25/2013 03:44:44 route,ospf,info OSPFv2 neighbor 172.16.240.24: state change from 2-Way to Init
Apr/25/2013 03:45:26 route,ospf,info OSPFv2 neighbor 172.16.240.24: state change from 2-Way to Init[/b]
Apr/25/2013 03:45:27 pppoe,ppp,info,account leidmar@certto.com.br logged out, 373 2491 4152 26 25
Apr/25/2013 03:45:27 pppoe,ppp,info <pppoe-leidmar@certto.com.br>: terminating...
Apr/25/2013 03:45:27 pppoe,ppp,info <pppoe-leidmar@certto.com.br>: disconnected
Apr/25/2013 03:45:28 pppoe,info PPPoE connection established from 00:1B:11:E5:2D:2E
Apr/25/2013 03:45:28 pppoe,ppp,info <pppoe-0>: waiting for call...
Apr/25/2013 03:45:28 pppoe,ppp,info <pppoe-0>: authenticated
Apr/25/2013 03:45:28 pppoe,ppp,info <pppoe-0>: connected

But at the same time, over x86, nothing happens so I am forced to think that the problem is only over RB1100AHx2 or version 6.0rcXX.

For CSC-auth-12 (x86)

Apr/25/2013 03:48:06 pppoe,info PPPoE connection established from 00:27:22:0C:63:A4
Apr/25/2013 03:48:06 pppoe,ppp,info <pppoe-0>: waiting for call...
Apr/25/2013 03:48:06 pppoe,ppp,error user dharmel@certto.com.br authentication failed
Apr/25/2013 03:48:06 pppoe,ppp,info <pppoe-0>: terminating... - user dharmel@certto.com.br authentication failed
Apr/25/2013 03:48:07 pppoe,ppp,info <pppoe-0>: disconnected
[b]Apr/25/2013 03:49:17 route,ospf,error Discarding packet: locally originated
Apr/25/2013 03:49:17 route,ospf,error     src address=172.16.17.140
Apr/25/2013 03:49:17 route,ospf,error Discarding packet: locally originated
Apr/25/2013 03:49:17 route,ospf,error     src address=172.16.17.76
Apr/25/2013 03:49:17 route,ospf,error Discarding packet: locally originated
Apr/25/2013 03:49:17 route,ospf,error     src address=172.16.31.253
Apr/25/2013 03:49:33 route,ospf,error Discarding packet: locally originated
Apr/25/2013 03:49:33 route,ospf,error     src address=172.16.17.140
Apr/25/2013 03:49:33 route,ospf,error Discarding packet: locally originated
Apr/25/2013 03:49:33 route,ospf,error     src address=172.16.17.76
Apr/25/2013 03:49:33 route,ospf,error Discarding packet: locally originated
Apr/25/2013 03:49:33 route,ospf,error     src address=172.16.31.253[/b]
Apr/25/2013 03:51:02 pppoe,ppp,info,account marcelofarias@certto.com.br logged out, 100437 480322693 6225831816 3561455 6366706
Apr/25/2013 03:51:02 pppoe,ppp,info <pppoe-marcelofarias@certto.com.br>: terminating...
Apr/25/2013 03:51:02 pppoe,ppp,info <pppoe-marcelofarias@certto.com.br>: disconnected
Apr/25/2013 03:51:10 pppoe,ppp,info,account carboni@certto.com.br logged out, 935 94301 298508 411 531
Apr/25/2013 03:51:10 pppoe,ppp,info <pppoe-carboni@certto.com.br>: terminating...
Apr/25/2013 03:51:10 pppoe,ppp,info <pppoe-carboni@certto.com.br>: disconnected

In this x86 I only see this strange "locally originated".

I am not sure where to go now. I need to run my network, I am seeing many problems, but no solution found. I hope someone from Mikrotik can see this and help me/us getting into the right tracks.

natanielklug · Thu Apr 25, 2013 2:48 pm

A quick update... Going further into the analysis of the problem I found that the 2nd to 3rd day dead line is something true (or not). Look at this screen:

In this screen you can see many VLAN's as neighbours but at CSC-auth-08 it's not connecting to the internal VLAN's like 1001, 1002 and 1005. This are our POP (which get to the customers in the last mile) VLAN's. Why is this router loosing OSPF neighbours for this side?

I am trying to reboot but it hangs in some part of reboot for some unknow reason and as I am using two ethernet ports (ether1 and ether6) it get ether1 led blinking (while trying to reboot) and ether6 led off.

Hope to get some more comments.

natanielklug · Mon Apr 29, 2013 4:34 pm

Hello all,

Quick update: the problem I run over last week was "solved" by using RB1100AHx2 ethernet port 13 (I was using ethernet port 6). For some reason after a period of time OSPF stop sending information over ether6 (even over vlans runnins over ether6). So now, without any reboot, machine is working fine for PPPoE and OSPF:

[admin@CSC-auth-08] > interface ethernet print without-paging 
Flags: X - disabled, R - running, S - slave 
 #    NAME                                    MTU MAC-ADDRESS       ARP        MASTER-PORT                                SWITCH                               
 0 R  ;;; CSC-auth-sw-01.p?? ( CSC-brd / CSC-cpd / CSC-ott ) =[S1]=
      ether1                                 1500 D4:CA:6D:81:E9:1E enabled    none                                       switch2                              
 1    ;;; RESERVA =[S1]=
      ether2                                 1500 D4:CA:6D:81:E9:1F enabled    none                                       switch2                              
 2    ;;; RESERVA =[S1]=
      ether3                                 1500 D4:CA:6D:81:E9:20 enabled    none                                       switch2                              
 3    ;;; RESERVA =[S1]=
      ether4                                 1500 D4:CA:6D:81:E9:21 enabled    none                                       switch2                              
 4    ;;; RESERVA =[S1]=
      ether5                                 1500 D4:CA:6D:81:E9:22 enabled    none                                       switch2                              
 5 X  ;;; MIGROU ether13 --- PROBLEMAS [ ? ? ? ? ] ( OSPF perde comunicacao com outras VLANs sobre esta porta )
      ether6                                 1500 D4:CA:6D:81:E9:23 enabled    none                                       switch1                              
 6    ;;; RESERVA =[S2]=
      ether7                                 1500 D4:CA:6D:81:E9:24 enabled    none                                       switch1                              
 7    ;;; RESERVA =[S2]=
      ether8                                 1500 D4:CA:6D:81:E9:25 enabled    none                                       switch1                              
 8    ;;; RESERVA =[S2]=
      ether9                                 1500 D4:CA:6D:81:E9:26 enabled    none                                       switch1                              
 9    ;;; RESERVA =[S2]=
      ether10                                1500 D4:CA:6D:81:E9:27 enabled    none                                       switch1                              
10    ;;; RESERVA
      ether11                                1500 D4:CA:6D:81:E9:28 enabled   
11    ;;; RESERVA
      ether12                                1500 D4:CA:6D:81:E9:29 enabled   
12 R  ;;; CSC-auth-sw-99.p08
      ether13                                1500 D4:CA:6D:81:E9:2A enabled

No problems found on PPPoE or OSPF after this change. There is still the problem about rebooting RB1100AHx2 which I could not test anymore as I am waiting some time to reboot machine. When rebooting RB1100AHx2 stop responding.

[admin@CSC-auth-08] > system resource print 
             uptime: 6d19h50m49s
            version: 6.0rc14
         build-time: Apr/22/2013 15:10:24
        free-memory: 1458.3MiB
       total-memory: 1518.7MiB
                cpu: e500v2
          cpu-count: 2
      cpu-frequency: 1066MHz
           cpu-load: 7%
     free-hdd-space: 91.6MiB
    total-hdd-space: 128.0MiB
  architecture-name: powerpc
         board-name: RB1100AHx2
           platform: MikroTik

[admin@CSC-auth-08] > interface pppoe-server print without-paging count-only 
151

I will be watching this closely.

natanielklug · Thu May 02, 2013 8:24 pm

I think this will be my last update in this case: we are nun running for 10 days with 6.0rc14 and no problem found.

The problem about routes was something over the ethernet interface (ether6). Changed to ether13 solved the problem.

gallysoft · Fri May 03, 2013 11:23 pm

Very good. Thank you!

natanielklug · Fri May 03, 2013 11:44 pm

Anytime...

natanielklug · Thu Jul 04, 2013 2:23 pm

People in this post should read this new one:

http://forum.mikrotik.com/viewtopic.php ... 02#p376202

Problems on PPPoE over different Beta/Stable versions

Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Re: Problems on PPPoE over different Beta/Stable versions

Who is online