Re: [ovs-dev] [PATCH ovn v4] ofctrl: Add a predictable resolution for conflicting flow actions.

Hi Dumitru, Thanks for the revision. It looks good overall. The major problems left are: 1. it missed updating the active desired_flow in the installed_flow. 2. when a tracked flow deletion is handled, if there are other desired flows linked to the same installed flow, it should use the same

[ovs-dev] [PATCH ovn] ofctrl.c: Avoid repeatedly linking an installed flow and a desired flow.

In update_installed_flows_by_compare() there are two loops. The first loop iterates the installed flows and find its peer in desired flows to 1. uninstall flows that are not needed anymore 2. update flows if needed At the same time, it links the desired flow found for the installed flow which also

[ovs-dev] [RFC net-next] net: openvswitch: Add support to lookup invalid packet in ct action.

From: Numan Siddique For a tcp packet which is part of an existing committed connection, nf_conntrack_in() will return err and set skb->_nfct to NULL if it is out of tcp window. ct action for this packet will set the ct_state to +inv which is as expected. But a controller cannot add an OVS flow

[ovs-dev] [PATCH ovn] ofctrl.c: Fix duplicated flow handling in I-P while merging opposite changes.

In a scenario in I-P when a desired flow is removed and an exactly same flow is added back in the same iteration, the merge_tracked_flows() function will merge the operation without firing any OpenFlow action. However, if there are multiple desired flows (e.g. F1 and F2) matching the same

[ovs-dev] [PATCH 2/2] python: set ovs.dirs variables with build system values

ovs/dirs.py should be auto-generated using the template ovs/dirs.py.template at build time. This will set the ovs.dirs python variables with a value specified by the environment or, if the environment variable is not set, from the build system. Signed-off-by: Mark Gray --- lib/automake.mk

Re: [ovs-dev] [PATCH ovn] northd: properly reconfigure ipam when subnet is changed

> > On 9/30/20 9:46 AM, Lorenzo Bianconi wrote: [...] > > Adding this side effect to ipam_get_unused_ip() makes it less clear what > this function's purpose is. Before this change, this function would > attempt to get the next unused IP and return it. Now, it does that but > it can also change

[ovs-dev] [PATCH 1/2] Documentation: update IPsec tutorial for F32

F32 requires the "python3-openvswitch" package now. Also, the iptables chain "IN_FedoraServer_allow" does not exist on Fedora 32. Signed-off-by: Mark Gray --- Documentation/tutorials/ipsec.rst | 27 --- 1 file changed, 12 insertions(+), 15 deletions(-) diff --git

[ovs-dev] [PATCH 0/2] Some fixes for OVS IPsec on Fedora

In the process of testing OVS IPsec on Fedora, I discovered some small issues to be resolved. I am following the tutorial at https://docs.openvswitch.org/en/latest/tutorials/ipsec/ Mark Gray (2): Documentation: update IPsec tutorial for F32 python: set ovs.dirs variables with build system

[ovs-dev] [PATCH v2 ovn] northd: properly reconfigure ipam when subnet is changed

Reconfigure dynamic assigned addresses if subnet is modified removing IP out of configured netmask if present Signed-off-by: Lorenzo Bianconi --- Changes since v1: - do not remove invalid address in ipam_get_unused_ip() --- northd/ovn-northd.c | 2 +- tests/ovn.at| 14 ++ 2

Re: [ovs-dev] [PATCH ovn v4] ofctrl: Add a predictable resolution for conflicting flow actions.

re-adding my ack to this version Acked-by: Mark Michelson On 9/30/20 3:38 AM, Dumitru Ceara wrote: Until now, in case the ACL configuration generates openflows that have the same match but different actions, ovn-controller was using the following approach: 1. If the flow being added contains

Re: [ovs-dev] [PATCH V3 11/24] datapath: fix possible memleak on destroy flow-table

On 9/29/2020 2:43 PM, Yi-Hung Wei wrote: On Wed, Sep 16, 2020 at 10:33 AM Greg Rose wrote: From: Tonghao Zhang Upstream commit: commit 50b0e61b32ee890a75b4377d5fbe770a86d6a4c1 Author: Tonghao Zhang Date: Fri Nov 1 22:23:52 2019 +0800 net: openvswitch: fix possible

Re: [ovs-dev] [PATCH V3 05/24] datapath: Set OvS recirc_id from tc chain index

On 9/28/2020 2:46 PM, Yi-Hung Wei wrote: On Wed, Sep 16, 2020 at 10:33 AM Greg Rose wrote: From: Paul Blakey Upstream commit: commit 95a7233c452a58a4c2310c456c73997853b2ec46 Author: Paul Blakey Date: Wed Sep 4 16:56:37 2019 +0300 net: openvswitch: Set OvS recirc_id

[ovs-dev] Alibaba Cloud Users Contact info

*Hi,* I really think you can Increase your sales growth by acquiring *Alibaba Cloud Users Contact info* We also have*: Cloudflare Users, Amazon CloudFront Users,* *Imperva Cloud Users, Akamai Users. And many more.* *Please let me know your interest so that I can provide you more information.*

Re: [ovs-dev] [PATCH ovn v2] lflow.c: Refactor function convert_match_to_expr.

On 10/2/20 9:39 AM, Mark Gray wrote: > > LGTM > > Acked-by: Mark Gray > > Thanks, > > Mark > Thanks Mark, Mark, and Han for reviews and applying this! ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH ovn v2] lflow.c: Refactor function convert_match_to_expr.

On 01/10/2020 12:22, Dumitru Ceara wrote: > To make it easier to understand what the function does we now explicitly > pass the input/output arguments instead of the complete > lflow_ctx_in/lflow_ctx_out context. > > Also: > - change the error handling to avoid forcing the caller to supply >