Posted on September 2, 2021 10:26 am
 |  Asked by Bhargav Chintakunta
 |  149 views
Tags:
0
1
Print Friendly, PDF & Email

Hello Team,

Hope you’re doing good. I am trying to  build  mLAG between two Arista switches in eve-ng, however system mac is showing 0.0.0.

I tried editing system mac by ([admin@localhost ~]$ echo 1122.3344.5566 > /mnt/flash/system_mac_address) , but it didn’t help

 

Peer-link is UP and I can ping peer ip’s but mLAG is not coming up. I am attaching show-tech files, can you please help me

 

Attachments:
0
Posted by Wei
Answered on September 2, 2021 5:15 pm

etba agent kept crashing.

Can you try another mac, or just reboot to take 00: mac?
11:22:33:44:55:66 mac address, low-order bit of the high-order octet is 1 so it's a multicast.

0
Answered on September 3, 2021 9:22 am

Hi Wei,

I Have changed Mac addresses to

0022.3344.5567 and

0022.3344.5568 on switch1 and 2 respectively.

 

It didn't help me. Am I missing any other things?

0
Posted by Aniket Bhowmick
Answered on September 3, 2021 2:48 pm

Hi Bhargav

As per the configurations, MLAG should come up. To debug further we would request if you can take packet capture on both switches in SVI 4094. Please run the below commands on both switch:

  • On CL1 -->  tcpdump interface vlan 4094 file flash:mlag-capture-cl1.pcap
  • On CL2-->  tcpdump interface vlan 4094 file flash:mlag-capture-cl2.pcap

The capture will be saved in /mnt/flash/ directory. Make sure the vEOS-lab switches has management access so that you can copy the capture to your local machine.

If it is not possible for you to get management access to copy the capture to your local machine, then type the below command on both switch:

  • tcpdump interface vlan 4094 verbose

^ This will dump the packets on the screen itself and you can share the output with us.

Also, we request if you can collect the below outputs from both the switch and share it with us (first one will be a large output):

  • show agent Mlag qtrace | no-more
  • show ip access-list | no-more

Regards,

Aniket

0
Answered on September 4, 2021 5:16 am

Hi Aniket,

Thanks for your response. I am attaching the requested, let me know if you need anything else

0
Answered on September 5, 2021 4:08 am

Hi Aniket,

Please find the attached

0
Answered on September 8, 2021 2:46 am

Hi Team,

Is this some kind of bug or something?

 

I can load old code if that makes sense? Like 4.26M?

0
Posted by Wei
Answered on September 8, 2021 1:54 pm

Hi Bhargav,

If you see new crash file under /var/core folder, please load another EOS code like the latest 4.25M.

0
Answered on September 9, 2021 5:19 am

Nope there are no core files

[admin@UKGLT-P10-PCL001 ~]$ ls -l
total 0
[admin@UKGLT-P10-PCL001 ~]$ cd /var
[admin@UKGLT-P10-PCL001 var]$ ls -l
total 0
drwxr-xr-x 2 root root 3 Apr 11 2018 adm
drwxr-xr-x 4 root root 42 Aug 7 12:37 cache
drwxrwxrwt 2 root root 40 Sep 9 10:45 core
drwxr-xr-x 3 root root 43 Aug 7 12:37 db
drwxr-xr-x 3 root root 27 Aug 7 12:38 empty
drwxr-xr-x 2 root root 3 Apr 11 2018 games
drwxr-xr-x 2 root root 3 Apr 11 2018 gopher
drwxr-xr-x 2 root root 3 Aug 1 09:51 issh
drwxr-xr-x 3 root root 27 Aug 7 12:37 kerberos
drwxr-xr-x 1 root root 80 Aug 7 12:39 lib
drwxr-xr-x 2 root root 3 Apr 11 2018 local
lrwxrwxrwx 1 root root 11 Aug 7 12:37 lock -> ../run/lock
drwxr-xr-x 9 root root 780 Sep 9 10:47 log
lrwxrwxrwx 1 root root 10 Aug 7 12:37 mail -> spool/mail
drwxr-xr-x 2 root root 3 Apr 11 2018 nis
drwxr-xr-x 2 root root 3 Apr 11 2018 opt
drwxr-xr-x 2 root root 3 Apr 11 2018 preserve
lrwxrwxrwx 1 root root 6 Aug 7 12:37 run -> ../run
drwxrwxrwt 6 root root 240 Sep 9 10:47 shmem
drwxr-xr-x 6 root root 65 Aug 7 12:38 spool
drwxrwxrwt 6 root root 160 Sep 9 10:47 tmp
drwxr-xr-x 2 root root 3 Apr 11 2018 yp
[admin@UKGLT-P10-PCL001 var]$ cd core
[admin@UKGLT-P10-PCL001 core]$ ls -l
total 0
[admin@UKGLT-P10-PCL001 core]$ pwd
/var/core

Post your Answer

You must be logged in to post an answer.