[Bugs] [Bug 1351949] management connection loss when volfile-server goes down

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1351949

Atin Mukherjee <amukherj at redhat.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Doc Type|If docs needed, set a value |Bug Fix


--- Doc Text *updated* ---
Problem:
Lets say from the pool of nodes, we have mounted the volume with Node1's IP which means that Node1 is playing as the volfile server role. In case of any changes to volume by 'vol set' or brick add/remove/replace those changes should send to client from a connection (management connection) between Node1 and client.

Currently in case if Node1 goes down the management connection is broken (expected), because of which any of 'volume set', 'add-brick/remove-brick'or any changes to volume will not be seen by client until Node1 comes back.

Fix:
In case Node1 goes down, switch between pool/list of other nodes to establish the management connection.

So from now on, if a volfile server goes down, there will an immediate switch in management connection path (i.e between client and Node1 to another node) and any changes to volume will be notified through the new connection. So we really don't have to wait for Node1 to come back alive.


-- 
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=ouhS6pOJtC&a=cc_unsubscribe