Community discussions

MikroTik App
 
ulink
Member Candidate
Member Candidate
Topic Author
Posts: 128
Joined: Fri Mar 05, 2010 5:07 am

Diagnosing this "possible" loop

Mon Nov 23, 2015 11:27 pm

Have a router log telling me the following:
"vpls250: Bridge port received a packet with own address as source address: probably loop"

I torched the interface and awaited the log entry to occur again then scoured it for a src. address that may match one of the many allocated to this unit. But, no go.

For anyone that has tracked this sort of issue before, can you share a few ways I may be able to find the source of this error?

Thanks in advance for your time and assistance.
 
Ape
Member Candidate
Member Candidate
Posts: 177
Joined: Sun Oct 06, 2013 3:32 pm
Location: Freiburg, Germany
Contact:

Re: Diagnosing this "possible" loop

Wed Nov 25, 2015 12:15 pm

In some scenarios with MPLS/VPLS, it might look like a loop but it isn't.
It can happen when paths are switching.

IMHO, as long as your log is not filling up with this message, you should be okay.

Additionally you can enable STP/RSTP on the bridge.

For debugging: Capture the traffic on a mirrored port and look at the source and destination MAC addresses. A loop can be detected, by looking for packets arriving at the bridge which where originally sent by the bridge itself (as the log message says).

Ape
 
ulink
Member Candidate
Member Candidate
Topic Author
Posts: 128
Joined: Fri Mar 05, 2010 5:07 am

Re: Diagnosing this "possible" loop

Wed Nov 25, 2015 4:18 pm

Thanks Ape for the time to provide the info.
 
ulink
Member Candidate
Member Candidate
Topic Author
Posts: 128
Joined: Fri Mar 05, 2010 5:07 am

Re: Diagnosing this "possible" loop

Tue Jan 19, 2016 11:25 pm

Still fighting this loop warning.
Is there a combination of log settings that may help shed more light on what the Mikrotik is seeing as a loop?
To date, I have reviewed the bridge Hosts list where 12 entries relating/connecting to port vpls250 are labeled. Going to each router and interface that attach to this tunnel, I see no duplicate MAC that may be finding its way back.
I have ran the packet sniffer on the interface and reviewed in wireshark the output for the interface using eth.src==MAC with no success. The result is null. Nothing is displayed indicating that nothing in that capture was sourced from the offending MAC. I even coordinated the error time stamp with the capture.
Regarding my setup. I am using split horizon bridging but have RSTP disabled (just disabled RSTP within the last 48 hours to test result).
Perhaps a temporary work around would be to use a bridge filter on the forward chain that blocks the offending MAC as a source address to the VPLS interface.
Thoughts from the forum appreciated. Thanks in advance.
vplserrorBS.png
You do not have the required permissions to view the files attached to this post.
 
pjinkcc
just joined
Posts: 12
Joined: Thu Aug 29, 2013 12:05 am

Re: Diagnosing this "possible" loop

Wed Mar 09, 2016 3:02 am

v6.32.3 here and I am having a similar issue but just on a very basic bridge setup with no special anything. No split horizon or filtering on any port or interface. I have not done a thorough packet analyzation like you but if there seems to be a better and quicker way to pinpoint this, it would be great to know.
 
User avatar
isolnet
newbie
Posts: 45
Joined: Sat Jan 30, 2016 7:28 am
Location: India

Re: Diagnosing this "possible" loop

Wed Mar 09, 2016 6:51 am

I think your are restore another device bkp

Kindly reset and fresh configure...

Who is online

Users browsing this forum: CHUPAPEE, faxxe, Google [Bot], holla29, Majestic-12 [Bot] and 217 guests