Re: [ovs-dev] [PATCH ovn v2 4/4] vif-plug.c: Use daemon_started_recently() to replace the prime counter.

On Sat, Aug 6, 2022 at 1:49 AM Han Zhou wrote: > > > > On Thu, Aug 4, 2022 at 7:09 AM Dumitru Ceara wrote: > > > > On 7/25/22 23:34, Han Zhou wrote: > > > Also remove the reset mechanism when DB is reconnected, because at DB > > > reconnection the data in IDL would not reset. > > > > > > Signed-o

Re: [ovs-dev] [PATCH ovn branch-21.12 0/4] Avoid unnecessary deletion & recreation during restart.

On Tue, Aug 9, 2022 at 4:22 AM Mark Michelson wrote: > > On 8/4/22 15:55, Numan Siddique wrote: > > On Thu, Aug 4, 2022 at 10:12 AM Mark Michelson wrote: > >> > >> This is a backport of Han Zhou's patch series "Avoid unnecessary > >> deletion & recreation during restart" for branch-21.12. The ori

Re: [ovs-dev] [PATCH ovn branch-22.03 1/5] ofctrl: Wakeup when entering S_UPDATE_FLOWS.

On Tue, Aug 9, 2022 at 4:29 AM Mark Michelson wrote: > > From: Han Zhou > > In theory it is possible that ovn-controller tried to install flows by > calling ofctrl_put() while ofctrl is at a state other than > S_UPDATE_FLOWS thus not able to install. So, when entering S_UPDATE_FLOWS, > we should

Re: [ovs-dev] [PATCH v2] datapath-windows: Add ERSPAN v1 support.

Sorry for the late response. Patch looks good just a few nits below. -Original Message- From: William Tu Sent: Monday, May 30, 2022 6:42 AM To: d...@openvswitch.org Cc: pweis...@vmware.com; Alin-Gabriel Serdean Subject: [PATCH v2] datapath-windows: Add ERSPAN v1 support. Similar to Li

Re: [ovs-dev] [PATCH ovn] northd: add support to make l3dgw ports fully distributed

On Mon, Aug 8, 2022 at 11:29 PM Vladislav Odintsov wrote: > > Hi Numan, > > thanks for the review. > My comments below. > > Regards, > Vladislav Odintsov > > > On 8 Aug 2022, at 03:24, Numan Siddique wrote: > > > > On Wed, Jul 20, 2022 at 12:15 AM Vladislav Odintsov > >

Re: [ovs-dev] [PATCH v1 1/2] datapath-windows:Fix icmp related error code.

Series applied on master thank you for the patches! Alin. -Original Message- From: dev On Behalf Of ldejing via dev Sent: Friday, June 24, 2022 2:09 PM To: d...@openvswitch.org Cc: ldejing Subject: [ovs-dev] [PATCH v1 1/2] datapath-windows:Fix icmp related error code. From: ldejing W

[ovs-dev] [PATCH 2/2] ofproto-dpif-upcall.c: Add counters for reasons of DP flow deletions.

Signed-off-by: Han Zhou --- ofproto/ofproto-dpif-upcall.c | 13 - 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/ofproto/ofproto-dpif-upcall.c b/ofproto/ofproto-dpif-upcall.c index 23b52ef40..f26a035f4 100644 --- a/ofproto/ofproto-dpif-upcall.c +++ b/ofproto/ofproto-dpi

[ovs-dev] [PATCH 1/2] Revalidator: Allow min-revalidator-pps to be 0 (disabled).

Today the minimum value for this setting is 1. This patch allows it to be 0, meaning not checking pps at all, and always do revalidation. This is particularly useful for environments where some of the applications with long-lived connections may have very low traffic for certain period but have hi

Re: [ovs-dev] [PATCH ovn v2] controller: Fix misaligned access to ip6_hdr

Thanks for the clarification Ales. Acked-by: Mark Michelson On 7/26/22 01:23, Ales Musil wrote: I forgot to add the v2 comment, so for clarification just the commit title has changed between the versions. Thanks, Ales On Mon, Jul 25, 2022 at 5:19 PM Ales Musil wrote: The ip6_hdr is aligne

[ovs-dev] [PATCH] github: Move CI to ubuntu 20.04 base image.

18.04 image is deprecated and will disappear soon. Also some slowdowns and brownouts are planned to push users away from this deprecated version: https://github.com/actions/virtual-environments/issues/6002 Moving to 20.04. Can't move to 22.04 at the moment because of deprecation warnings from

Re: [ovs-dev] [PATCH] netdev-offload-tc: Disable offload of IPv6 fragments.

On 8/5/22 01:57, Marcelo Leitner wrote: > On Fri, Jul 29, 2022 at 10:38:36PM +0200, Ilya Maximets wrote: >> OVS kernel datapath and TC are parsing IPv6 fragments differently. >> For IPv6 later fragments, according to the original design [1], OVS >> always sets nw_proto=44 (IPPROTO_FRAMENT), regardl

Re: [ovs-dev] [PATCH v1] ovs-save: Use right OpenFlow version for add-tlv-map

On 8/3/22 20:42, Mike Pattrick wrote: > On Fri, Jul 1, 2022 at 6:53 AM Han Ding wrote: >> >> >> When the bridge protocols is not included Openflow10, printing an error >> message "version negotiation failed" when doing "Restoring saved flows". >> >> Signed-off-by: Han Ding > > Looks good to me!

Re: [ovs-dev] [PATCH ovn branch-22.03 4/5] ofctrl.c: Use bundle to avoid data plane downtime during the first flow installation.

Bleep bloop. Greetings Mark Michelson, I am a robot and I have tried out your patch. Thanks for your contribution. I encountered some error that I wasn't expecting. See the details below. checkpatch: WARNING: Comment with 'xxx' marker #138 FILE: controller/ofctrl.c:2610: * XXX: Ideal

Re: [ovs-dev] [PATCH ovn branch-22.03 2/5] ofctrl: Support ovn-ofctrl-wait-before-clear to reduce down time during upgrade.

Bleep bloop. Greetings Mark Michelson, I am a robot and I have tried out your patch. Thanks for your contribution. I encountered some error that I wasn't expecting. See the details below. checkpatch: WARNING: Line is 84 characters long (recommended limit is 79) #323 FILE: controller/ovn-contr

[ovs-dev] [PATCH ovn branch-22.03 5/5] extend-table: Fix table ID double allocation after OVS restart.

From: Han Zhou There were problems observed occasionally after OVS restart, the OVS flow bundle installation from ovn-controller was failed because of "GROUP_EXISTS" error, which end up with missing flows/groups/meters in OVS until ovn-controller is restarted. Example error logs in OVS: 2022-07-

[ovs-dev] [PATCH ovn branch-22.03 4/5] ofctrl.c: Use bundle to avoid data plane downtime during the first flow installation.

From: Han Zhou Now in ofctrl we wait in S_WAIT_BEFORE_CLEAR for the initial flow compute to complete before clearing the existing flows, to reduce the data plane down time during ovn-controller restart. However, the flow installation takes a long time when the flow number is huge, which still lea

[ovs-dev] [PATCH ovn branch-22.03 3/5] ofctrl.c: Include group changes to bundle.

From: Han Zhou Move the group changes to the same bundle as the OF flow changes. The steps in ofctrl_put were: - add groups - add meters - bundle - change flows - delete groups - delete meters Now it becomes: - add meters - bundle - add groups - change flows - del

[ovs-dev] [PATCH ovn branch-22.03 2/5] ofctrl: Support ovn-ofctrl-wait-before-clear to reduce down time during upgrade.

From: Han Zhou Whenever OpenFlow connection between ovn-controller and OVS is connected/reconnected, typically during ovn-controller/OVS restart/upgrade, ovn-controller would: 1. clears all the existing flows after the initial hand-shaking 2. compute the new flows 3. install the new flows to OVS.

[ovs-dev] [PATCH ovn branch-22.03 1/5] ofctrl: Wakeup when entering S_UPDATE_FLOWS.

From: Han Zhou In theory it is possible that ovn-controller tried to install flows by calling ofctrl_put() while ofctrl is at a state other than S_UPDATE_FLOWS thus not able to install. So, when entering S_UPDATE_FLOWS, we should immediately wakeup the main loop so that any pending flows can be i

Re: [ovs-dev] [PATCH ovn branch-21.12 0/4] Avoid unnecessary deletion & recreation during restart.

On 8/4/22 15:55, Numan Siddique wrote: On Thu, Aug 4, 2022 at 10:12 AM Mark Michelson wrote: This is a backport of Han Zhou's patch series "Avoid unnecessary deletion & recreation during restart" for branch-21.12. The original series can be found at https://patchwork.ozlabs.org/project/ovn/lis

[ovs-dev] [PATCH v2] python: Do not send non-zero flag for a SSL socket

Python std library SSLSocket.send does not allow non-zero value for the optional flag. pyOpenSSL was recently switched for the Python standard library ssl module commit 68543dd523bd00f53fa7b91777b962ccb22ce679 (python: Replace pyOpenSSL with ssl). Python SSLsocket.send() does not allow non-zero

[ovs-dev] [PATCH 1/2] netdev-offload: Expose error when init_flow_api() returns EBUSY.

init_flow_api() can fail due to errors, such as netlink "Device or resource busy". Today the errors are ignored except just an INFO log. This ends up with an interface created without offload enabled, and application won't notice it. This patch expose the error to the OVSDB record, and reverts the

[ovs-dev] [PATCH 2/2] bridge.c: Retry when iface_create encounters transient failures.

Starts a timer when EBUSY is returned, to let it retry, because EBUSY is considered a temporary error when the device is operated by something else (e.g. Network Manager) at the moment, and retry should succeed. Signed-off-by: Han Zhou --- vswitchd/bridge.c | 13 - 1 file changed, 12

Re: [ovs-dev] [PATCH ovn] tests: Avoid matching on the OVS flow key contents.

On 8/8/22 17:29, Dumitru Ceara wrote: > These may change over time and, as a matter of fact, have recently changed > in OVS master branch with commit fcdf8ae4a350 ("lib: Print nw_frag in flow > key."). > > Instead, strip the flow key from the output. To achieve that, the calls > to ovn-trace had

Re: [ovs-dev] [PATCH] python: Send non-zero flag for a SSL socket

On 8/8/22 17:22, Miro Tomaska wrote: > Yes, you may build python without ssl support (if ssl libraries are not > available when you build it) and so try is still necessary (they also > do that in cpython, for example in asyncio > https://github.com/python/cpython/blob/main/Lib/async

[ovs-dev] [PATCH ovn] tests: Avoid matching on the OVS flow key contents.

These may change over time and, as a matter of fact, have recently changed in OVS master branch with commit fcdf8ae4a350 ("lib: Print nw_frag in flow key."). Instead, strip the flow key from the output. To achieve that, the calls to ovn-trace had to go through a bit of refactoring too. OVN has w

Re: [ovs-dev] [PATCH] python: Send non-zero flag for a SSL socket

> > Yes, you may build python without ssl support (if ssl libraries are not > available when you build it) and so try is still necessary (they also > do that in cpython, for example in asyncio > https://github.com/python/cpython/blob/main/Lib/asyncio/sslproto.py) Good point on the import. I did n

Re: [ovs-dev] [PATCH] system-traffic: Fix IPv4 fragmentation test sequence for check-kernel.

On 8/5/22 23:49, Paolo Valerio wrote: > Ilya Maximets writes: > >> On 8/5/22 17:08, Paolo Valerio wrote: >>> The following test sequence: >>> >>> conntrack - IPv4 fragmentation incomplete reassembled packet >>> conntrack - IPv4 fragmentation with fragments specified >>> >>> leads to a systematic

Re: [ovs-dev] [PATCH] system-traffic: Don't run IPv6 header modification test on kernels < 5.19.

On 8/5/22 20:13, Aaron Conole wrote: > Ilya Maximets writes: > >> OVS kernel module is incorrectly updating checksums while changing >> IPv6 fields of later fragments that doesn't really have L4 headers. >> >> This makes the 'ping6 between two ports with header modify' test >> fail on most of the

Re: [ovs-dev] [PATCH] system-traffic: Fix incorrect neigh entry in ipv6 header modification test.

On 8/8/22 16:17, Aaron Conole wrote: > Ilya Maximets writes: > >> On 8/5/22 20:32, Aaron Conole wrote: >>> Ilya Maximets writes: >>> The permanent neighbor entry for fc00::1 is added into a wrong namespace, so in order to reply to a ping from at_ns1, the address of fc00::1 has to

Re: [ovs-dev] [PATCH] system-traffic: Fix incorrect neigh entry in ipv6 header modification test.

Ilya Maximets writes: > On 8/5/22 20:32, Aaron Conole wrote: >> Ilya Maximets writes: >> >>> The permanent neighbor entry for fc00::1 is added into a wrong >>> namespace, so in order to reply to a ping from at_ns1, the >>> address of fc00::1 has to be discovered. >> >> This is strange - how di

Re: [ovs-dev] OVSDB echo requests/replies in a background thread: should we?

On 8/5/22 08:28, Han Zhou wrote: > On Thu, Jun 30, 2022 at 12:55 PM Mark Michelson wrote: >> >> Hi OVS and OVN devs, >> >> The OVN team has considered the idea of moving inactivity probes (i.e. >> OVSDB echo requests/replies) into a background thread. >> >> OVN logical networks can be very large,

Re: [ovs-dev] [PATCH ovn] northd: add support to make l3dgw ports fully distributed

Hi Numan, thanks for the review. My comments below. Regards, Vladislav Odintsov > On 8 Aug 2022, at 03:24, Numan Siddique wrote: > > On Wed, Jul 20, 2022 at 12:15 AM Vladislav Odintsov > wrote: >> >> This is used when traffic from HW VTEP goes to >> routable network

Re: [ovs-dev] [PATCH] system-traffic: Fix incorrect neigh entry in ipv6 header modification test.

> -Original Message- > From: Ilya Maximets > Sent: Friday 5 August 2022 19:41 > To: Aaron Conole > Cc: i.maxim...@ovn.org; ovs-dev@openvswitch.org; Mike Pattrick > ; Salem Sol ; Phelan, Michael > > Subject: Re: [PATCH] system-traffic: Fix incorrect neigh entry in ipv6 header > modificati

Re: [ovs-dev] [PATCH] python: Send non-zero flag for a SSL socket

On 8/8/22 12:19, Timothy Redaelli wrote: > On Mon, 8 Aug 2022 11:42:55 +0200 > Ilya Maximets wrote: > >> On 8/8/22 04:27, Miro Tomaska wrote: >>> Python std library SSLSocket.send does not allow non-zero value for the >>> optional flag. >>> >>> pyOpenSSL was recently switched for the Python stan

Re: [ovs-dev] [PATCH] python: Send non-zero flag for a SSL socket

On Mon, 8 Aug 2022 11:42:55 +0200 Ilya Maximets wrote: > On 8/8/22 04:27, Miro Tomaska wrote: > > Python std library SSLSocket.send does not allow non-zero value for the > > optional flag. > > > > pyOpenSSL was recently switched for the Python standard library ssl module > > commit 68543dd523bd

Re: [ovs-dev] [PATCH] python: Send non-zero flag for a SSL socket

On 8/8/22 04:27, Miro Tomaska wrote: > Python std library SSLSocket.send does not allow non-zero value for the > optional flag. > > pyOpenSSL was recently switched for the Python standard library ssl module > commit 68543dd523bd00f53fa7b91777b962ccb22ce679 (python: Replace pyOpenSSL > with ssl).

Re: [ovs-dev] OVN 22.09 Soft Freeze 5 August, 2022

On 8/5/22 22:22, Mark Michelson wrote: > On 8/5/22 12:36, Dumitru Ceara wrote: >> On 7/29/22 17:50, Mark Michelson wrote: >>> Hi everyone, >>> >> >> Hi Mark, >> >>> Soft freeze for the OVN 22.09.0 release is in one week, 5 August. As a >>> reminder, soft freeze is the deadline for new feature patch

[ovs-dev] reply: [PATCH ovn] pinctrl: fix bug of monitor status in health check

Ok, I understand what you mean, numans. But here still has a problem, if service_monitor does not get a reply due to backends is not work. Thus, health status is ok. If client sends request for vip, even vip forwords request to backend, client would not get a reply. So, here may be needs another