riginal Message-
> From: Ilya Maximets
> Sent: Saturday, August 20, 2022 2:12 AM
> To: Oleksandr Mykhalskyi
> Cc: i.maxim...@ovn.org; ovs-discuss@openvswitch.org; Dan Williams
> ; Terry Wilson
> Subject: Re: [ovs-discuss] Unreas
Maximets
Sent: Saturday, August 20, 2022 2:12 AM
To: Oleksandr Mykhalskyi
Cc: i.maxim...@ovn.org; ovs-discuss@openvswitch.org; Dan Williams
; Terry Wilson
Subject: Re: [ovs-discuss] Unreasonably often OVS DB compaction
[External Email]
On 8/18/22 11:44, Oleksandr
you can test it out.
Best regards, Ilya Maximets.
>
> Thank you.
>
> Oleksandr Mykhalskyi, System Engineer, IT Servers and Services, LAUT
> Netcracker Technology
>
> +38 (044) 238-8727 ext 6404 | office
>
>
> -Original Message-----
> From: Ilya Maximets
>
@openvswitch.org; Dan Williams
; Terry Wilson
Subject: Re: [ovs-discuss] Unreasonably often OVS DB compaction
[External Email]
> Terry, thank you for update.
> I will check if openstack xena or yoga uses OVS 2.17
>
> Meanwhile, my main concern
> Terry, thank you for update.
> I will check if openstack xena or yoga uses OVS 2.17
>
> Meanwhile, my main concern - why OVS DB triggers compaction every 10-20min
> regardless of conditions, described in
> http://www.openvswitch.org/support/dist-docs/ovsdb-server.1.txt?
> For our case, with
Terry Wilson
Sent: Wednesday, August 3, 2022 8:38 PM
To: Oleksandr Mykhalskyi
Cc: b...@openvswitch.org; Alexander Povaliukhin
; Alexander Stavitskiy
Subject: Re: [ovs-discuss] Unreasonably often OVS DB compaction
[External Email]
You'll need to use ovs 2.1
You'll need to use ovs 2.17 (at least python-ovs 2.17) to use raft
effectively. Leader transfer on compaction will cause disconnected
clients, and before python-ovs 2.17, monitor-cond-since/update3
support was not there to make the reconnect happen using a snapshot
instead of a full dump of the db
On Wed, 2022-08-03 at 13:16 +, Oleksandr Mykhalskyi via discuss
wrote:
> Dear openvswitch developers,
>
> After recent update of our openstack wallaby cloud, when OVS was
> updated from 2.15.0 to 2.15.2, we observe new OVS DB behavior with
> often (every 10-20 min) transferring leadership in
Dear openvswitch developers,
After recent update of our openstack wallaby cloud, when OVS was updated from
2.15.0 to 2.15.2, we observe new OVS DB behavior with often (every 10-20 min)
transferring leadership in raft cluster.
Transferring leadership was implemented by
