- Description has changed:
Diff:
~~~~
--- old
+++ new
@@ -4,6 +4,12 @@
---------
Instead of having specialized system controller nodes, the idea is that all
nodes are born equal and they all have the potential of taking on the system
controller role. This makes sense in (but its usefulness is not limited to) a
cloud deployment, where OpenSAF runs on a collection of compute nodes, and from
an outside perspective no particular node is any different from the other nodes.
+
+Expectations
+---------
+
+* The cluster can start and restart as long as any non-empty subset of the
configured nodes is present (i.e. there are no dedicated controller nodes that
are required to be present in order to start the system)
+* If one of the controller nodes dies, a payload node (if there is any
available) will be promoted to become a controller node. This promotion is done
without requiring the payload node to reboot.
~~~~
---
** [tickets:#79] Dynamically changing the opensaf node type
(controller/payload)**
**Status:** unassigned
**Milestone:** future
**Created:** Mon May 13, 2013 04:33 AM UTC by Nagendra Kumar
**Last Updated:** Tue Jul 14, 2015 10:43 AM UTC
**Owner:** nobody
Migrated from http://devel.opensaf.org/ticket/2101
Rationale
---------
Instead of having specialized system controller nodes, the idea is that all
nodes are born equal and they all have the potential of taking on the system
controller role. This makes sense in (but its usefulness is not limited to) a
cloud deployment, where OpenSAF runs on a collection of compute nodes, and from
an outside perspective no particular node is any different from the other nodes.
Expectations
---------
* The cluster can start and restart as long as any non-empty subset of the
configured nodes is present (i.e. there are no dedicated controller nodes that
are required to be present in order to start the system)
* If one of the controller nodes dies, a payload node (if there is any
available) will be promoted to become a controller node. This promotion is done
without requiring the payload node to reboot.
We can add support for the following at run time:
1. Changing a controller node into payload node.
2. Changing a payload node into a controller node.
As of now, some portion of it can be implemented by locking middleware 2N SU on
standby controller node and unlocking 2N SU on payload node[Assumption is 2N SU
are configured on all the nodes and unlocked on one node.].
For locking/unlocking the middleware 2N SU, refer the patch floated for
http://devel.opensaf.org/ticket/2023
Any thoughts ??
------------------------------
Changed 15 months ago by hafe ¶
■summary changed from Dynamically changing the node capability
(controller/payload) to Dynamically changing the opensaf node type
(controller/payload)
Changed slogan to not confuse with node capacity which is an AMF concept (not
yet implemented)
---
Sent from sourceforge.net because [email protected] is
subscribed to https://sourceforge.net/p/opensaf/tickets/
To unsubscribe from further messages, a project admin can change settings at
https://sourceforge.net/p/opensaf/admin/tickets/options. Or, if this is a
mailing list, you can unsubscribe from the mailing list.
------------------------------------------------------------------------------
_______________________________________________
Opensaf-tickets mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets
