What is strange is that ISL is seen as down.
Could you try toggling the LAG used for ISL on both primary and secondary ?
like checking LACP state.
A very important note:
In November 2020, Aruba communicate that all AOS-CX systems should be upgraded to at least 10.05.0021 for 10.05 (or 10.04.3031 for 10.04).
due to a bug impacting the life of SSD. I see you still run 10.05.0020. I would strongly recommend to upgrade to the latest maintenance release of 10.08
(which brings lot of new features and bug fixes).
------------------------------
Vincent Giles
------------------------------
Original Message:
Sent: Mar 10, 2022 05:36 AM
From: Timo Krjukoff
Subject: 8320 VSX Split recovery
Hi
Had a split brain on our 8320 cluster running 10.05.0020.
Split Recovery Mode: Enabled
All MC-LAG interfaces on the secondary node remain down.
ISL channel: Out-Of-Sync
ISL mgmt channel: inter_switch_link_down
Config Sync Status: out-of-sync
NAE: peer_unreachable
...
ISL version: same in Primary & secondary
Software version: ame in Primary & secondary
...
Device Role: Primary = Primary & Secondary = Secondary (Device roles inconsistent)
I have tried to find examples on how to manually recover the current split situation but cannot find any information.
Since the secondary node is obviously not forwarding any traffic I thought that i could just go ahead and reboot it, BUT not sure what's gonna happen if nodes are not in sync.
Appreciate any relevant advice.
Cheers
Timo Krjukoff
------------------------------
Timo Krjukoff
------------------------------