Thanks for filing an FFe. I think this broadly makes sense and ovn releases are pretty stable and tested. As for now, I see the following changelog entries:
Post v23.06.0
-------------
- Added FDB aging mechanism, that is disabled by default.
It can be enabled per logical switch with other_config
"fdb_age_threshold".
- Add DHCPv6 "fqdn" (39) option, that works similarly to
DHCPv4 "hostname" (12) option.
- Support to create/update MAC_Binding when GARP received from VTEP (RAMP)
switch on l3dgw port.
- To allow optimizing ovn-controller's monitor conditions for the regular
VIF case, ovn-controller now unconditionally monitors all sub-ports
(ports with parent_port set).
- ECMP routes use L4_SYM dp-hash by default if the datapath supports it.
Existing sessions might get re-hashed to a different ECMP path when
OVN detects the algorithm support in the datapath during an upgrade
or restart of ovn-controller.
Given this is not yet final, I don't think we can take that call right now. I mean, as I said, in broader terms it should be OK but how about this - once this is released and the changelog is finalized, we can circle back on this and take a decision?
Feel free to ping me or release team members on IRC and we'll take another look.
Just to be clear - I am not an official release team member yet. :)
Hi,
Thanks for filing an FFe. I think this broadly makes sense and ovn releases are pretty stable and tested. As for now, I see the following changelog entries:
Post v23.06.0 age_threshold" .
-------------
- Added FDB aging mechanism, that is disabled by default.
It can be enabled per logical switch with other_config
"fdb_
- Add DHCPv6 "fqdn" (39) option, that works similarly to
DHCPv4 "hostname" (12) option.
- Support to create/update MAC_Binding when GARP received from VTEP (RAMP)
switch on l3dgw port.
- To allow optimizing ovn-controller's monitor conditions for the regular
VIF case, ovn-controller now unconditionally monitors all sub-ports
(ports with parent_port set).
- ECMP routes use L4_SYM dp-hash by default if the datapath supports it.
Existing sessions might get re-hashed to a different ECMP path when
OVN detects the algorithm support in the datapath during an upgrade
or restart of ovn-controller.
Given this is not yet final, I don't think we can take that call right now. I mean, as I said, in broader terms it should be OK but how about this - once this is released and the changelog is finalized, we can circle back on this and take a decision?
Feel free to ping me or release team members on IRC and we'll take another look.
Just to be clear - I am not an official release team member yet. :)