Hello ALL,
Could you hint how else I can troubleshoot problem with UDLD ?
Switches: 9702E
AOS: 6.4.4.603.R01
I have static lacp between two 9702E switches.
This aggregation has 8x10Gb optical links.
Problem:
I have flapping some links in the aggregation daily (3-10 times per day) by UDLD protocol.
As UDLD works in agressive mode so these links goes down.
Before we had the same problem on other switches, but we cleaned optical links and problem is still existing between two switches.
I collected some logs but for me this information is not so clear.
Thank you in advance.
UDLD problem
UDLD problem
You do not have the required permissions to view the files attached to this post.
Re: UDLD problem
Hi,
if you use LACP instead of static lag than you don't need udld (in my opinion).
In your log I'm wondering about the LBD messages. Have you activated LBD at the OS9702? LBD is for loop detection at the edge side.
UDLD:
did you have the following output during the issue?
> show interfaces port
Slot/ Admin Link Violations Recovery Recovery WTRPort Status Status Time Max (sec)
------+----------+---------+----------+----------+----------+--------
2/6 enable down UDLD 300 10 0
recovery time is 300 = 5Min. thats why at 23:27 the port joins again the lag.
So you have to exchange all patchcable between the port 2/6 of both switches (or clean it also).
regards
Silvio
if you use LACP instead of static lag than you don't need udld (in my opinion).
In your log I'm wondering about the LBD messages. Have you activated LBD at the OS9702? LBD is for loop detection at the edge side.
UDLD:
did you have the following output during the issue?
> show interfaces port
Slot/ Admin Link Violations Recovery Recovery WTRPort Status Status Time Max (sec)
------+----------+---------+----------+----------+----------+--------
2/6 enable down UDLD 300 10 0
recovery time is 300 = 5Min. thats why at 23:27 the port joins again the lag.
So you have to exchange all patchcable between the port 2/6 of both switches (or clean it also).
regards
Silvio
Re: UDLD problem
Hello silvio.
Thank you once again for your help !
I was wrong. There is static LAG between switches and not LACP.
LBD is not activated on switches, you can see it in attached file.
You are right that every time when port goes down it will recovered in 300 sec(5 min).
According to Recovery Timer on port.
Also, we tried to exclude patch panel and made direct connection of one link in LAG between switches but without success.
So, may be we should make cleaning fo ports once again ?
Thank you once again for your help !
I was wrong. There is static LAG between switches and not LACP.
LBD is not activated on switches, you can see it in attached file.
You are right that every time when port goes down it will recovered in 300 sec(5 min).
According to Recovery Timer on port.
Also, we tried to exclude patch panel and made direct connection of one link in LAG between switches but without success.
So, may be we should make cleaning fo ports once again ?
You do not have the required permissions to view the files attached to this post.
Re: UDLD problem
yes, clean your sfp, or exchange it, there could be the reason.
Sometimes its helpfull (if possible) to exchange at one site (f.e. only at switch A) the LAG-Ports. So if the error changes from one port to the other or it keeps at the old port ... you can see at which site (switch A or switch B) the defect connection is located.
Silvio
Sometimes its helpfull (if possible) to exchange at one site (f.e. only at switch A) the LAG-Ports. So if the error changes from one port to the other or it keeps at the old port ... you can see at which site (switch A or switch B) the defect connection is located.
Silvio
