Hello people i have been reviewing the differences between MetroCluster TieBreaker and ontap mediator for a MCC IP deployment. After reading some docs and powerpoint, i am not able to perceive the difference from the applications/services/clients point of view how the two differ among them. ->Mediator can be configured from ontap CLI (or System Mgr) , can also use CHAP and certificates, and also have two mailbox disks to store MetroCluster info, monitored from ontap systems itself. ->TieBreaker can autosupport , polls the nodes, log collection and snmp traps, and can initiate a switchover Reading this from manual-documentation, https://docs.netapp.com/us-en/ontap-metrocluster/install-ip/concept_considerations_mediator.html#how-the-ontap-mediator-supports-automatic-unplanned-s... "When a node detects a site failure requiring a switchover, it takes steps to confirm that the switchover is appropriate and, if so, performs the switchover. Mediator Assisted Unplanned SwithOver is only initiated if both SyncMirror mirroring and DR mirroring of each node’s nonvolatile cache is operating and the caches and mirrors are synchronized at the time of the failure" i have some doubts about using the classical Tiebreaker VS Mediator on behalf of failure scenarios. Could anybody shed some light and explain explicit failure scenarios and how differ Tiebreaker and Mediator when is required? Thanks in advance
- supports MAUSO (controllers or complete site failure are covered)
The Mediator came in ONTAP 9.7 (for MCC IP only), and as we can see, it better integrates with ONTAP and also covers controllers or complete site failure (considering that DR mirroring and plex mirrors are in sync). So I would suggest to always go with mediator when MCC IP.
this is a MCC IP scenario bid w 4-node (2localHApair +2remoteHApair nodes-controllers).
In the beginning this will be a active-passive MC layout , 2 data aggregates in sync in local. Nothing complicated. A database host-server in local and another similar waiting in remote to grab and activate the SVM+luns+iSCSI sessions in case of disaster in local site.
Seems Mediator can fit better here avoiding switching over if that DR mirroring and plex mirrors are not in sync.
Always situations and exact use cases pose the solutions to be applied, sure i agree.
Well, in my opinion after reading some docs and pptx, it has never been pretty clear the purposes and differences between the two methods.
This is a MCC IP scenario bid w 4-node (2localHApair +2remoteHApair nodes-controllers).
In the beginning this will be a active-passive MC layout , 2 data aggregates in sync in localHApair. Nothing complicated. A database host-server in local frontend and another similar waiting in remote to grab and activate the SVM+luns+iSCSI sessions in case of disaster in local site. Spanned network layer2 among sites in frontend access.
My specific situations in order to select one MCTB or Mediator for Automatic Unplanned events, are:
-Storage only disaster
I like Mediator because it avoids switching over if that DR mirroring and plex mirrors are not in sync.