First thoughts that come to my mind:
It looks like you are running and EAPS ring as well. The EAPS master switch with blocking port cannot be an MLAG peer as well, see also this thread:
https://community.extremenetworks.com/extreme/topics/eaps_and_mlag
Your ISC should always be redundant, so ideally a sharing group over 2 ports with LACP enabled. Not relevant for your issue, but important for your overall design.
What are HV0 and HV1 (servers?), and how are they connected with the "brown" link to the x690 (port 65) and the x870? Is this supposed to be another LAG / MLAG??
Where is your upstream connectivity connected to, and how?
Overall... I don't get your design with the 2 x870 and the 2 x690, what their roles are and how they are connected and why. What exactly do you want to achieve? At the moment it looks like you have added several redundancy concepts just because, adding way too much complexity.