Re: [ovs-dev] [PATCH ovn v2] Fix load balanced hairpin traffic for fragmented packets.

https://issues.redhat.com/browse/FDP-905 > Fixes: 1139b655c996 ("Don't blindly save original dst IP and Port to avoid > megaflow unwildcarding.") > CC: Han Zhou > Suggested-by: Dumitru Ceara > Signed-off-by: Numan Siddique > --- > > v1 -> v2 > ---

Re: [ovs-dev] [PATCH ovn v2] Fix load balanced hairpin traffic for fragmented packets.

https://issues.redhat.com/browse/FDP-905 > Fixes: 1139b655c996 ("Don't blindly save original dst IP and Port to avoid > megaflow unwildcarding.") > CC: Han Zhou > Suggested-by: Dumitru Ceara > Signed-off-by: Numan Siddique Recheck-request: github-robot-_ovn-kubernet

Re: [ovs-dev] [RFC v2 ovn 1/4] northd: Introduce ECMP_Nexthop table in SB db.

On 11/8/24 11:43, Lorenzo Bianconi wrote: >> Hi Lorenzo, >> >> On 9/19/24 19:08, Lorenzo Bianconi wrote: >>> Introduce ECMP_Nexthop table in the SB db in order to track active >>> ecmp-symmetric-reply connections and flush stale ones. ECMP_Nexthop >>> table contains ip and mac address for each acti

Re: [ovs-dev] [PATCH ovn] northd: Fix issues in RBAC tables recovery.

On 11/8/24 12:26, Aleksandr Smirnov (K2 Cloud) wrote: > On 11/8/24 12:43 PM, Dumitru Ceara wrote: >> On 11/8/24 10:14, Aleksandr Smirnov (K2 Cloud) wrote: >>>> On 10/4/24 13:41, Aleksandr Smirnov wrote: >>>>> From: Aleksandr Smirnov >>>>> >&

Re: [ovs-dev] [PATCH ovn] northd: Fix issues in RBAC tables recovery.

On 11/8/24 10:14, Aleksandr Smirnov (K2 Cloud) wrote: >> On 10/4/24 13:41, Aleksandr Smirnov wrote: >>> From: Aleksandr Smirnov >>> >>> Northd creates hardcoded RBAC role 'ovn-controller' with number of >>> predefined permissions. Then it watches for alternations of the role >>> and permissions an

Re: [ovs-dev] [RFC v2 ovn 4/4] ofctrl: Introduce ecmp_nexthop_monitor.

On 11/7/24 12:22, Lorenzo Bianconi wrote: >> Also, should we instead add a new module? E.g., ecmp-next-hop-monitor.c.? > it is just few lines of code, does it worth to do so? Maybe not, but the problem I have now is that the variable definition is far away from the function and it took me a while

Re: [ovs-dev] [PATCH ovn v1] Fix load balanced hairpin traffic for fragmented packets.

nics, care is taken > to match on these fields only for fragmented packets. > > Reported-at: https://issues.redhat.com/browse/FDP-905 > Fixes: 1139b655c996 ("Don't blindly save original dst IP and Port to avoid > megaflow unwildcarding.") > CC: Han Zhou > Sug

Re: [ovs-dev] [RFC v2 ovn 2/4] pinctrl: Send periodic arp/nd to ecmp next-hops.

On 11/7/24 17:34, Lorenzo Bianconi wrote: > On Nov 06, Dumitru Ceara wrote: >> Hi Lorenzo, >> >> On 9/19/24 19:08, Lorenzo Bianconi wrote: >>> Introduce the capbility to periodically send ARP/ND packets for >>> ECMP nexthops in order to resolve their L2 addr

Re: [ovs-dev] [RFC v2 ovn 4/4] ofctrl: Introduce ecmp_nexthop_monitor.

On 11/7/24 12:28, Lorenzo Bianconi wrote: > [...] +}; + +ovs_be32 lo = get_unaligned_be32((void *)&ea.be16[1]); +ovs_u128 nexthop = { +.u64.hi = ntohs(ea.be16[0]), +.u64.lo = (uint64_t) ntohl(lo) << 32, +}; + +struct o

Re: [ovs-dev] [RFC v2 ovn 4/4] ofctrl: Introduce ecmp_nexthop_monitor.

On 11/7/24 16:07, Lorenzo Bianconi wrote: > On Nov 07, Dumitru Ceara wrote: >> On 11/7/24 15:20, Lorenzo Bianconi wrote: >>> On Nov 07, Dumitru Ceara wrote: >>>> On 11/7/24 12:28, Lorenzo Bianconi wrote: >>>>> [...] >>>>>>>> +

Re: [ovs-dev] [RFC v2 ovn 4/4] ofctrl: Introduce ecmp_nexthop_monitor.

On 11/7/24 15:20, Lorenzo Bianconi wrote: > On Nov 07, Dumitru Ceara wrote: >> On 11/7/24 12:28, Lorenzo Bianconi wrote: >>> [...] >>>>>> +}; >>>>>> + >>>>>> +ovs_be32 lo = get_unaligned_be32((void *)&ea.be16[1])

Re: [ovs-dev] [RFC v2 ovn 3/4] pinctrl: Update ecmp-nexthop mac resolving L2 address.

On 11/7/24 16:06, Lorenzo Bianconi wrote: > On Nov 07, Dumitru Ceara wrote: >> On 11/7/24 15:17, Lorenzo Bianconi wrote: >>>> Hi Lorenzo, >>>> >>>> On 9/19/24 19:08, Lorenzo Bianconi wrote: >>>>> Set the mac address column in the ECMP_

Re: [ovs-dev] [RFC v2 ovn 3/4] pinctrl: Update ecmp-nexthop mac resolving L2 address.

On 11/7/24 15:17, Lorenzo Bianconi wrote: >> Hi Lorenzo, >> >> On 9/19/24 19:08, Lorenzo Bianconi wrote: >>> Set the mac address column in the ECMP_Nexthop table updating >>> sb MAC_Binding table. This is a preliminary patch to introduce the >>> capability to flush stale ECMP CT entries. >>> > > [

Re: [ovs-dev] [RFC v2 ovn 2/4] pinctrl: Send periodic arp/nd to ecmp next-hops.

Hi Lorenzo, On 9/19/24 19:08, Lorenzo Bianconi wrote: > Introduce the capbility to periodically send ARP/ND packets for > ECMP nexthops in order to resolve their L2 address. This is a preliminary > patch to introduce the capability to flush stale ECMP CT entries. > > Signed-off-by: Lorenzo Bianco

Re: [ovs-dev] [PATCH ovn v2] tests: Wait for controller exit before restart.

On 10/2/24 10:39, Xavier Simonart wrote: > In some rare cases, and despite "recent" changes to wait for cleanup > before replying to exit, ovn-controller was still running when trying > to restart it. > > Signed-off-by: Xavier Simonart > > --- > v2: - Renamed OVN_CONTROLLER_EXIT_RESTART to OVN_C

Re: [ovs-dev] [RFC v2 ovn 3/4] pinctrl: Update ecmp-nexthop mac resolving L2 address.

Hi Lorenzo, On 9/19/24 19:08, Lorenzo Bianconi wrote: > Set the mac address column in the ECMP_Nexthop table updating > sb MAC_Binding table. This is a preliminary patch to introduce the > capability to flush stale ECMP CT entries. > > Signed-off-by: Lorenzo Bianconi > --- > controller/ovn-cont

Re: [ovs-dev] [RFC v2 ovn 4/4] ofctrl: Introduce ecmp_nexthop_monitor.

Hi Lorenzo, On 9/19/24 19:08, Lorenzo Bianconi wrote: > Introduce ecmp_nexthop_monitor in ovn-controller in order to track and > flush ecmp-symmetric reply ct entires when requested by the CMS (e.g > removing the related static ecmp routes). CT entries are flushed using > the ethernet mac address

Re: [ovs-dev] [PATCH ovn v8] ovn-controller: Reserve zones for upcoming ports.

Hi Mansi, On 10/7/24 06:38, Mansi Sharma via dev wrote: > Ping for review. > Sorry for the delay. > From: Mansi Sharma > Date: Monday, 30 September 2024 at 12:41 PM > To: d...@openvswitch.org , num...@ovn.org > , amu...@redhat.com > Cc: Mansi Sharma > Subject: [PATCH ovn v8] ovn-controller:

Re: [ovs-dev] [PATCH ovn] northd: Fix issues in RBAC tables recovery.

On 10/4/24 13:41, Aleksandr Smirnov wrote: > From: Aleksandr Smirnov > > Northd creates hardcoded RBAC role 'ovn-controller' with number of > predefined permissions. Then it watches for alternations of the role > and permissions and recover them if they were changed. > An original code have issue

Re: [ovs-dev] [RFC v2 ovn 4/4] ofctrl: Introduce ecmp_nexthop_monitor.

On 11/6/24 12:38, Dumitru Ceara wrote: > Hi Lorenzo, > > On 9/19/24 19:08, Lorenzo Bianconi wrote: >> Introduce ecmp_nexthop_monitor in ovn-controller in order to track and >> flush ecmp-symmetric reply ct entires when requested by the CMS (e.g >> removing the rela

Re: [ovs-dev] [RFC v2 ovn 1/4] northd: Introduce ECMP_Nexthop table in SB db.

Hi Lorenzo, On 9/19/24 19:08, Lorenzo Bianconi wrote: > Introduce ECMP_Nexthop table in the SB db in order to track active > ecmp-symmetric-reply connections and flush stale ones. ECMP_Nexthop > table contains ip and mac address for each active nexthop. > > Signed-off-by: Lorenzo Bianconi > ---

Re: [ovs-dev] [RFC v2 ovn 0/4] Introduce ECMP_nexthop monitor in ovn-controller

On 9/19/24 19:08, Lorenzo Bianconi wrote: > With respect to the previous implementation, OVN is now tracking and > flushing ECMP next-hop ct entries with the next-hop mac address instead > of using an identification number committed by ovn-northd since the previous > approach had some limitations w

Re: [ovs-dev] [PATCH ovn v3] controller: Cleanup FDB entries when a VIF goes down.

ntrollers in the deployment will > remove the FDB flows from table OFTABLE_GET_FDB as soon as the port > binding status goes down. > > If the above approach seems too complicated I'm fine with this patch > as an interim solution. > I vote for avoiding an interim solution. If

Re: [ovs-dev] [PATCH ovn v2] Revert "controller: Properly handle localnet flows in I+P.".

Hi Xavier, On 11/5/24 13:07, Xavier Simonart wrote: > Hi > > The test needs two small changes: > - ovs-appctl vlog/set dbg should be removed (Thanks Dumitru for catching > this) > - the two AT_CHECK in the test should be replaced with OVS_WAIT_UNTIL to > avoid a potential flaky test as waiting fo

Re: [ovs-dev] [PATCH ovn 0/4] controller: FDB/MAC_Binding learning/aging fixes.

On 11/4/24 10:45, Ales Musil wrote: > On Fri, Nov 1, 2024 at 2:29 PM Dumitru Ceara wrote: > >> This series is a collection of FDB and MAC_Binding fixes for >> ovn-controller's learning and aging code. >> >> The first patch in the code fixes a bug that was in

Re: [ovs-dev] [PATCH ovn 2/4] mac-cache: Properly handle deletion of SB mac_bindings.

On 11/4/24 10:44, Ales Musil wrote: > On Fri, Nov 1, 2024 at 2:29 PM Dumitru Ceara wrote: > >> SB.MAC_Binding doesn't store a strong reference to SB.Port_Binding. >> Instead it stores the port binding 'logical_port' name. As the >> mac-binding cache in o

Re: [ovs-dev] [PATCH ovn] Revert "controller: Properly handle localnet flows in I+P.".

On 11/4/24 16:29, Xavier Simonart wrote: > Hi Dumitru > > Thanks for the review and the comments. > > On Mon, Nov 4, 2024 at 12:34 PM Dumitru Ceara wrote: > >> Hi Xavier, >> >> On 11/1/24 15:03, Mark Michelson wrote: >>> Thanks for the fix Xavier.

Re: [ovs-dev] [PATCH ovn] Revert "controller: Properly handle localnet flows in I+P.".

Hi Xavier, On 11/1/24 15:03, Mark Michelson wrote: > Thanks for the fix Xavier. > +1 > Acked-by: Mark Michelson > > On 10/31/24 13:10, Xavier Simonart wrote: >> This reverts commit edc064b4c589ab1bb69352523481bd6d997aa1ca. >> >> In setups with multiple localnet ports w/o vlans (or multiple lo

Re: [ovs-dev] [PATCH] github: Remove ASLR entropy workaround.

On 10/30/24 17:44, Ilya Maximets wrote: > On 10/25/24 22:40, Ilya Maximets wrote: >> The issue should be fixed by now. >> >> Signed-off-by: Ilya Maximets >> --- >> .github/workflows/build-and-test.yml | 8 >> 1 file changed, 8 deletions(-) >> >> diff --git a/.github/workflows/build-and-t

[ovs-dev] [PATCH ovn 2/4] mac-cache: Properly handle deletion of SB mac_bindings.

.com/browse/FDP-906 Fixes: b57f60a617b4 ("controller: Add MAC cache I-P node") Signed-off-by: Dumitru Ceara --- controller/mac-cache.c | 77 ++--- controller/mac-cache.h | 30 +++--- controller/ovn-controller.c | 67 ++ controller/pinctrl.c

[ovs-dev] [PATCH ovn 3/4] pinctrl: Skip deleted mac bindings in run_buffered_binding().

There's no real point to try to send any buffered packets for these MAC bindings, they're stale anyway. Fixes: bbd5e0d81c16 ("controller: improve buffered packets management") Signed-off-by: Dumitru Ceara --- controller/pinctrl.c | 4 1 file changed, 4 inserti

[ovs-dev] [PATCH ovn 1/4] pinctrl: Use correct map size in pinctrl_handle_put_fdb().

Fixes: fb96ae36793a ("controller: Merge the mac-cache and mac-learn.") Signed-off-by: Dumitru Ceara --- controller/pinctrl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/controller/pinctrl.c b/controller/pinctrl.c index b891435c12..dfb3130aa2 100644 --- a/

[ovs-dev] [PATCH ovn 0/4] controller: FDB/MAC_Binding learning/aging fixes.

pdate deleted SB.MAC_Binding IDL records when the port they're learnt on is being deleted. The last two patches are fixes that restrict the range of SB.MAC_Bindings that ovn-controller's pinctrl processes (skips deleted and non-local ones). Dumitru Ceara (4): pinctrl: Use corr

[ovs-dev] [PATCH ovn 4/4] pinctrl: Skip non-local mac bindings in run_buffered_binding().

There's no point to process MAC binding updates for non-local datapaths. If they were local before but not anymore we don't really have a reason to send any buffered packets for them anymore. Fixes: bbd5e0d81c16 ("controller: improve buffered packets management") Signed-o

[ovs-dev] [PATCH ovn] Documentation: Clarify OVS submodule bump strategy.

-October/417702.html Signed-off-by: Dumitru Ceara --- Documentation/internals/ovs_submodule.rst | 62 +-- 1 file changed, 35 insertions(+), 27 deletions(-) diff --git a/Documentation/internals/ovs_submodule.rst b/Documentation/internals/ovs_submodule.rst index 51f2a8c758

Re: [ovs-dev] [PATCH] github: Remove ASLR entropy workaround.

On 10/25/24 22:40, Ilya Maximets wrote: > The issue should be fixed by now. > > Signed-off-by: Ilya Maximets > --- Looks good to me! Acked-by: Dumitru Ceara If you don't mind I'll cherry pick this to OVN too once it gets merged in

[ovs-dev] OVN technical community meeting - Dec 2nd (Was: Re: OVN technical community meeting - Oct 21st)

On 10/22/24 13:18, Dumitru Ceara wrote: > On 10/22/24 12:55, Ilya Maximets wrote: >> On 10/22/24 11:40, Dumitru Ceara wrote: [...] >>> As discussed at the end of the meeting I'm thinking of suggesting >>> November 25th (5 weeks from now), 3PM UTC as a new date/tim

Re: [ovs-dev] [PATCH ovn 0/5] Reuse UUID for SB rows if there is 1:1 mapping

On 10/18/24 12:09, Ales Musil wrote: > Some of the NB rows has only one corresponding row in SB, we can > reuse the UUID from NB for the SB row. This makes it easier to handle > changes as we can directly get the corrsponding row by UUID and is also > better for debugging, because the relation is o

Re: [ovs-dev] OVN technical community meeting - Oct 21st

On 10/22/24 12:55, Ilya Maximets wrote: > On 10/22/24 11:40, Dumitru Ceara wrote: >> Hi everyone, >> >> On 10/18/24 15:56, Dumitru Ceara wrote: >>> Just a quick reminder for anyone interested in joining the next OVN >>> technical meeting. It's schedule

Re: [ovs-dev] OVN technical community meeting - Oct 21st

Hi everyone, On 10/18/24 15:56, Dumitru Ceara wrote: > Just a quick reminder for anyone interested in joining the next OVN > technical meeting. It's scheduled to happen Monday, October 21st, at > 3PM UTC. > > Meeting details: > Date/Time: Monday, October 21st

Re: [ovs-dev] OVN technical community meeting - Oct 21st

Hi Frode, Felix, On 10/21/24 08:07, Felix Huettner wrote: > On Sat, Oct 19, 2024 at 08:52:39AM +0200, Frode Nordahl wrote: >> fre. 18. okt. 2024, 15:56 skrev Dumitru Ceara : >> >>> Hi everyone, >>> >>> Just a quick reminder for anyone interested in

[ovs-dev] OVN technical community meeting - Oct 21st

Hi everyone, Just a quick reminder for anyone interested in joining the next OVN technical meeting. It's scheduled to happen Monday, October 21st, at 3PM UTC. Meeting details: Date/Time: Monday, October 21st, 3PM UTC Link: https://meet.google.com/zns-gqsd-jdn Agenda: https://docs.google.com/docu

Re: [ovs-dev] [PATCH ovn branch-24.03] ovs: Bump submodule to latest OVS v3.3.2.

Slight correction, the commit log will be "ovs: Bump submodule to OVS v3.3.2.". I'll fix that up when merging the patch. Regards, Dumitru On 10/18/24 09:50, Dumitru Ceara wrote: > From: Vladislav Odintsov > > This picks up the following relevant OVS changes: >

Re: [ovs-dev] [PATCH ovn branch-24.09] ovs: Bump submodule to latest OVS branch-3.4.

On 10/18/24 07:42, Vladislav Odintsov wrote: > Hi Dumitru, > > On 15.10.2024 18:34, Dumitru Ceara wrote: >> On 10/14/24 16:41, Vladislav Odintsov wrote: >>> On 14.10.2024 16:13, Dumitru Ceara wrote: >>>> On 10/14/24 15:01, Vladislav Odintsov wrote: >>&g

Re: [ovs-dev] [PATCH ovn branch-22.03] ovs: Bump submodule to OVS 3.0.7.

t; > On 18.10.2024 10:55, Dumitru Ceara wrote: >> From: Vladislav Odintsov >> >> This picks up the following relevant OVS changes: >>4198bcdfb compiler: Fix errors in Clang 17 ubsan checks. >>ec048ab62 vlog: Destroy async_append first then close log_fd. >

Re: [ovs-dev] [PATCH ovn branch-22.06] ovs: Bump submodule to OVS 3.0.7.

On 10/18/24 10:13, Vladislav Odintsov wrote: > Nit: commit subject should be adjusted to: "ovs: Bump submodule to OVS > v3.0.7.". > Thanks for pointing that out! I fixed it up and applied the patch to 22.06. Thanks, Dumitru > On 18.10.2024 10:54, Dumitru Ceara wro

Re: [ovs-dev] [PATCH ovn branch-22.09] ovs: Bump submodule to OVS 3.0.7.

On 10/18/24 10:13, Vladislav Odintsov wrote: > Nit: commit subject should be adjusted to: "ovs: Bump submodule to OVS > v3.0.7.". > Thanks for pointing that out! I fixed it up and applied the patch to 22.09. Thanks, Dumitru > On 18.10.2024 10:54, Dumitru Ceara wro

Re: [ovs-dev] [PATCH ovn branch-22.12] ovs: Bump submodule to OVS v3.1.5.

On 10/18/24 09:53, Dumitru Ceara wrote: > From: Vladislav Odintsov > > This picks up the following relevant OVS changes: > 7694dfacb compiler: Fix errors in Clang 17 ubsan checks. > faf175155 vlog: Destroy async_append first then close log_fd. > 483bc24e4 hash, jhash: Fi

Re: [ovs-dev] [PATCH ovn branch-23.03] ovs: Bump submodule to OVS v3.1.5.

On 10/18/24 09:52, Dumitru Ceara wrote: > From: Vladislav Odintsov > > This picks up the following relevant OVS changes: > 7694dfacb compiler: Fix errors in Clang 17 ubsan checks. > faf175155 vlog: Destroy async_append first then close log_fd. > 483bc24e4 hash, jhash: Fi

Re: [ovs-dev] [PATCH ovn branch-23.09 1/2] Set release date for 23.09.6.

ce..9cc5d7274 100644 > --- a/debian/changelog > +++ b/debian/changelog > @@ -2,7 +2,7 @@ ovn (23.09.6-1) unstable; urgency=low > [ OVN team ] > * New upstream version > > - -- OVN team Fri, 16 Aug 2024 13:51:01 -0400 > + -- OVN team Thu, 17 Oct 2024 15:07:00 -040

Re: [ovs-dev] [PATCH ovn branch-23.09] ovs: Bump submodule to OVS v3.2.3.

On 10/18/24 10:59, Dumitru Ceara wrote: > This needs proper authorship, my bad: I'll change the author to: > > Author: Vladislav Odintsov > I fixed this up and applied the patch to 23.09. Thanks, Dumitru > On 10/18/24 09:51, Dumitru Ceara wrote: >> This picks up

Re: [ovs-dev] [PATCH ovn branch-24.03] ovs: Bump submodule to latest OVS v3.3.2.

Applied to 24.03. Thanks, Dumitru On 10/18/24 10:00, Dumitru Ceara wrote: > Slight correction, the commit log will be "ovs: Bump submodule to OVS > v3.3.2.". I'll fix that up when merging the patch. > > Regards, > Dumitru > > On 10/18/24 09:50, Dumitru Cear

Re: [ovs-dev] [PATCH ovn branch-24.09 2/2] Prepare for 24.09.2.

* New upstream version > + > + -- OVN team Thu, 17 Oct 2024 15:12:04 -0400 > + > ovn (24.09.1-1) unstable; urgency=low > [ OVN team ] > * New upstream version Acked-by: Dumitru Ceara Thanks! ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH ovn branch-23.09 2/2] Prepare for 23.09.7.

* New upstream version > + > + -- OVN team Thu, 17 Oct 2024 15:07:00 -0400 > + > ovn (23.09.6-1) unstable; urgency=low > [ OVN team ] > * New upstream version Acked-by: Dumitru Ceara Thanks! ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH ovn branch-24.03 1/2] Set release date for 24.03.4.

94..6476191d6 100644 > --- a/debian/changelog > +++ b/debian/changelog > @@ -2,7 +2,7 @@ ovn (24.03.4-1) unstable; urgency=low > [ OVN team ] > * New upstream version > > - -- OVN team Fri, 16 Aug 2024 13:51:03 -0400 > + -- OVN team Thu, 17 Oct 2024 15:12:02 -040

Re: [ovs-dev] [PATCH ovn branch-24.03 2/2] Prepare for 24.03.5.

* New upstream version > + > + -- OVN team Thu, 17 Oct 2024 15:12:02 -0400 > + > ovn (24.03.4-1) unstable; urgency=low > [ OVN team ] > * New upstream version Acked-by: Dumitru Ceara Thanks! ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH ovn branch-24.09 1/2] Set release date for 24.09.1.

2c..a50569bb6 100644 > --- a/debian/changelog > +++ b/debian/changelog > @@ -2,7 +2,7 @@ ovn (24.09.1-1) unstable; urgency=low > [ OVN team ] > * New upstream version > > - -- OVN team Fri, 13 Sep 2024 19:09:31 -0400 > + -- OVN team Thu, 17 Oct 2024 15:12:04 -040

Re: [ovs-dev] [PATCH ovn branch-23.09] ovs: Bump submodule to OVS v3.2.3.

This needs proper authorship, my bad: I'll change the author to: Author: Vladislav Odintsov On 10/18/24 09:51, Dumitru Ceara wrote: > This picks up the following relevant OVS changes: > 27841b2c184b compiler: Fix errors in Clang 17 ubsan checks. > a622f6cc6324 vlog: Destro

[ovs-dev] [PATCH ovn branch-22.12] ovs: Bump submodule to OVS v3.1.5.

From: Vladislav Odintsov This picks up the following relevant OVS changes: 7694dfacb compiler: Fix errors in Clang 17 ubsan checks. faf175155 vlog: Destroy async_append first then close log_fd. 483bc24e4 hash, jhash: Fix unaligned access to the hash remainder. bd5b5d3b3 ovs-atomic: Fix in

[ovs-dev] [PATCH ovn branch-22.03] ovs: Bump submodule to OVS 3.0.7.

From: Vladislav Odintsov This picks up the following relevant OVS changes: 4198bcdfb compiler: Fix errors in Clang 17 ubsan checks. ec048ab62 vlog: Destroy async_append first then close log_fd. dbaf7271c hash, jhash: Fix unaligned access to the hash remainder. 85285fb45 ovs-atomic: Fix in

[ovs-dev] [PATCH ovn branch-22.06] ovs: Bump submodule to OVS 3.0.7.

From: Vladislav Odintsov This picks up the following relevant OVS changes: 4198bcdfb compiler: Fix errors in Clang 17 ubsan checks. ec048ab62 vlog: Destroy async_append first then close log_fd. dbaf7271c hash, jhash: Fix unaligned access to the hash remainder. 85285fb45 ovs-atomic: Fix in

[ovs-dev] [PATCH ovn branch-22.09] ovs: Bump submodule to OVS 3.0.7.

From: Vladislav Odintsov This picks up the following relevant OVS changes: 4198bcdfb compiler: Fix errors in Clang 17 ubsan checks. ec048ab62 vlog: Destroy async_append first then close log_fd. dbaf7271c hash, jhash: Fix unaligned access to the hash remainder. 85285fb45 ovs-atomic: Fix in

[ovs-dev] [PATCH ovn branch-23.03] ovs: Bump submodule to OVS v3.1.5.

From: Vladislav Odintsov This picks up the following relevant OVS changes: 7694dfacb compiler: Fix errors in Clang 17 ubsan checks. faf175155 vlog: Destroy async_append first then close log_fd. 483bc24e4 hash, jhash: Fix unaligned access to the hash remainder. bd5b5d3b3 ovs-atomic: Fix in

[ovs-dev] [PATCH ovn branch-23.09] ovs: Bump submodule to OVS v3.2.3.

This picks up the following relevant OVS changes: 27841b2c184b compiler: Fix errors in Clang 17 ubsan checks. a622f6cc6324 vlog: Destroy async_append first then close log_fd. 34b4675aa639 hash, jhash: Fix unaligned access to the hash remainder. ... and others Reported-at: https://mail.ope

[ovs-dev] [PATCH ovn branch-24.03] ovs: Bump submodule to latest OVS v3.3.2.

From: Vladislav Odintsov This picks up the following relevant OVS changes: f02dc3cfe vlog: Destroy async_append first then close log_fd. 01eca18be hash, jhash: Fix unaligned access to the hash remainder. ... and others. Reported-at: https://mail.openvswitch.org/pipermail/ovs-dev/2024-Octo

Re: [ovs-dev] [OVN] Issues with lb_force_snat_ip=router_ip and multiple networks per port.

Hi Surya, On 10/16/24 15:24, Surya Seetharaman wrote: > > On Wed, Oct 16, 2024 at 1:35 PM Ilya Maximets > wrote: > > > I still think that issue 1 should be addressed though by choosing the > appropriate network instead of the random one.  If not as a bug fix,

Re: [ovs-dev] [OVN] Issues with lb_force_snat_ip=router_ip and multiple networks per port.

On 10/16/24 12:43, Ilya Maximets wrote: > On 10/16/24 09:53, Dumitru Ceara wrote: >> Thanks Ilya and Numan, for your replies! >> >> On 10/15/24 18:57, Numan Siddique wrote: >>> On Tue, Oct 15, 2024 at 11:41 AM Ilya Maximets wrote: >>>> >>>>

Re: [ovs-dev] [OVN] Issues with lb_force_snat_ip=router_ip and multiple networks per port.

Thanks Ilya and Numan, for your replies! On 10/15/24 18:57, Numan Siddique wrote: > On Tue, Oct 15, 2024 at 11:41 AM Ilya Maximets wrote: >> >> On 10/15/24 15:51, Dumitru Ceara wrote: >>> Hi all, >>> >>> We received a bug report from the

Re: [ovs-dev] [PATCH ovn branch-24.09] ovs: Bump submodule to latest OVS branch-3.4.

On 10/14/24 16:41, Vladislav Odintsov wrote: > > On 14.10.2024 16:13, Dumitru Ceara wrote: >> On 10/14/24 15:01, Vladislav Odintsov wrote: >>> On 14.10.2024 14:47, Dumitru Ceara wrote: >>>> On 10/13/24 10:19, Vladislav Odintsov wrote: >>>>>

[ovs-dev] [OVN] Issues with lb_force_snat_ip=router_ip and multiple networks per port.

Hi all, We received a bug report from the ovn-kubernetes development team: https://issues.redhat.com/browse/FDP-871 Given a gateway router with load balancers attached to it. If: - the gateway router has a router port attached to multiple IP networks and - the router is configured with lb_for

Re: [ovs-dev] [ovn-dev] Port mirroring filter support in ovn.

On 10/14/24 15:38, Aleksandra Rukomoinikova wrote: > Hey there, > Hi Aleksandra, Thanks for reaching out! > I am writing to seek clarification on the feasibility of implementing > traffic filters for port mirroring in ovn, specifically regarding the > encapsulation of traffic to overlay using

Re: [ovs-dev] [PATCH ovn branch-24.09] ovs: Bump submodule to latest OVS branch-3.4.

On 10/14/24 15:01, Vladislav Odintsov wrote: > > On 14.10.2024 14:47, Dumitru Ceara wrote: >> On 10/13/24 10:19, Vladislav Odintsov wrote: >>> This picks up the following relevant OVS changes: >>>a15ce086d ofproto-dpif: Improve load balancing in dp_hash select

Re: [ovs-dev] [PATCH ovn branch-24.09] ovs: Bump submodule to latest OVS branch-3.4.

On 10/13/24 10:19, Vladislav Odintsov wrote: > This picks up the following relevant OVS changes: > a15ce086d ofproto-dpif: Improve load balancing in dp_hash select groups. > 76ba41b5c vconn: Always properly free flow stats reply. > 64cb90507 ovsdb-idl: Fix IDL memory leak. > ... and others.

Re: [ovs-dev] [PATCH ovn branch-23.03] ovs: Bump submodule to tip of OVS branch-3.2.

On 10/11/24 21:20, Odintsov Vladislav wrote: > On 11.10.2024 14:00, Dumitru Ceara wrote: > >> On 10/7/24 12:45, Vladislav Odintsov wrote: >>> From: Dumitru Ceara >>> >>> This picks up the following relevant commit: >>> cd8ffc956c3c ovs

Re: [ovs-dev] [PATCH ovn v2] ci: Add timeout command to test execution.

On 10/11/24 13:53, Ales Musil wrote: > The timeout command allows us to specify a timeout for each job in CI. > This is handy because there might be a test that is not able to > finish for various reasons, last time it was forking nc process. > At the same time having timeout over the make command

Re: [ovs-dev] [PATCH ovn] ci: Add timeout command to test execution.

Hi Ales, Thanks for the patch! On 10/10/24 14:20, Ales Musil wrote: > The timeout command allows us to specify a timeout for each job in CI. > This is handy because there might be a test that is not able to > finish fro various reasons, last time it was forking nc process. Typo: "fro". > At the

Re: [ovs-dev] [PATCH ovn branch-23.03] ovs: Bump submodule to tip of OVS branch-3.2.

On 10/7/24 12:45, Vladislav Odintsov wrote: > From: Dumitru Ceara > > This picks up the following relevant commit: > cd8ffc956c3c ovs-atomic: Fix inclusion of Clang header by GCC 14. > > Without this builds on Fedora 40 (rawhide) are broken due to failing to >

Re: [ovs-dev] [PATCH ovn] northd: Allow multichassis port to be bound on remote chassis.

On 10/9/24 10:06, Ales Musil wrote: > In order to support tunneling with multichassis port across > multiple AZ allow the port to be bound to remote chassis if > there is multiple chassis requested. > > Reported-at: https://issues.redhat.com/browse/FDP-850 > Signed-off-by: Ales Musil > --- With

Re: [ovs-dev] [PATCH ovn] northd: Commit ct_label.obs_point_id for blocked connections.

On 10/8/24 13:16, Adrián Moreno wrote: > On Tue, Oct 08, 2024 at 11:51:54AM GMT, Dumitru Ceara wrote: >> On 10/8/24 10:59, Lorenzo Bianconi wrote: >>> Considering the following configuration: >>> >>> $ovn-nbctl acl-list sw01 >>> from-lport 100 (inp

Re: [ovs-dev] [PATCH ovn] northd: Commit ct_label.obs_point_id for blocked connections.

On 10/8/24 10:59, Lorenzo Bianconi wrote: > Considering the following configuration: > > $ovn-nbctl acl-list sw01 > from-lport 100 (inport == "sw01-port1" && udp.dst == 5201) allow-related > [after-lb] > from-lport10 (inport == "sw01-port1" && udp) drop [after-lb] > > $ovn-nbctl list acl >

Re: [ovs-dev] [PATCH ovn v2] northd: Respect --ecmp-symmetric-reply for single routes.

On 9/26/24 11:38, Dumitru Ceara wrote: > On 9/20/24 21:00, Dumitru Ceara wrote: >> On 9/20/24 17:36, Rosemarie O'Riorden wrote: >>> When preparing to build ECMP and static route flows, routes are sorted >>> into unique routes (meaning they are not part of a grou

Re: [ovs-dev] [PATCH ovn v2] northd: Respect --ecmp-symmetric-reply for single routes.

On 9/20/24 21:00, Dumitru Ceara wrote: > On 9/20/24 17:36, Rosemarie O'Riorden wrote: >> When preparing to build ECMP and static route flows, routes are sorted >> into unique routes (meaning they are not part of a group) or they are >> added to EMCP groups. Then, ECMP ro

Re: [ovs-dev] [PATCH ovn] northd, controller: Recompute immediately after commit failure.

On 9/10/24 15:29, Ales Musil wrote: > The recompute was requested when the commit failed, however the > engine run might have happened later than expected. This was caused > by poll_block() call, which was waiting for any event in the loop. > Make sure that the engine runs right away by calling > p

[ovs-dev] OVN technical community meeting - Oct 21st (Was: Re: OVN technical community meeting - Sep 16th)

On 9/18/24 11:23, Frode Nordahl wrote: >> I'd also like to start the discussion about scheduling the next instance >> of the meeting. Would October 21st, 3PM UTC, work for most people? > Works for me, thank you for organizing! Hi everyone, I went ahead and scheduled a new instance of the OVN tec

Re: [ovs-dev] [PATCH ovn v2] northd: Respect --ecmp-symmetric-reply for single routes.

t; Edited documentation to support this change. > Also updated incorrect actions in documentation. > > Fixes: 4fdca656857d ("Add ECMP symmetric replies.") > Reported-at: https://issues.redhat.com/browse/FDP-786 > Signed-off-by: Rosemarie O'Riorden > --- Looks good to

Re: [ovs-dev] [PATCH ovn] ovn-controller: Fix potential assert when exiting.

On 9/19/24 14:11, Ales Musil wrote: > On Thu, Jul 25, 2024 at 3:16 PM Xavier Simonart wrote: > >> There was a potential ovs_assert when exiting ovn-controller: >> controller/if-status.c:261: assertion >> shash_is_empty(&mgr->ovn_uninstall_hash) failed in if_status_mgr_clear() >> >> Fixes: 395

Re: [ovs-dev] [PATCH ovn] Documentation: Add the OVN project charter page.

On 9/18/24 17:09, Russell Bryant wrote: > On Wed, Sep 18, 2024 at 10:21 AM Dumitru Ceara wrote: > >> Per [0] OVN has been a Linux Foundation Collaborative Project since >> November 4th, 2019. The OVN charter is linked in the same email thread >> [1]. For better visib

[ovs-dev] [PATCH ovn] Documentation: Add the OVN project charter page.

ss/2020-January/049702.html [1] https://mail.openvswitch.org/pipermail/ovs-discuss/attachments/20200116/c648d433/attachment-0001.pdf Signed-off-by: Dumitru Ceara --- Documentation/automake.mk | 1 + Documentation/index.rst | 3 +- Documentati

Re: [ovs-dev] [PATCH ovn] northd: Respect --ecmp-symmetric-reply for single routes.

On 9/18/24 13:14, Ilya Maximets wrote: > On 9/18/24 12:39, Dumitru Ceara wrote: >> On 9/18/24 12:31, Ilya Maximets wrote: >>> On 9/18/24 12:26, Ilya Maximets wrote: >>>> On 9/18/24 12:06, Dumitru Ceara wrote: >>>>> On 9/17/24 19:00, Rosemarie O'

Re: [ovs-dev] [PATCH ovn] northd: Respect --ecmp-symmetric-reply for single routes.

On 9/18/24 12:31, Ilya Maximets wrote: > On 9/18/24 12:26, Ilya Maximets wrote: >> On 9/18/24 12:06, Dumitru Ceara wrote: >>> On 9/17/24 19:00, Rosemarie O'Riorden wrote: >>>> When preparing to build ECMP and static route flows, routes are sorted >>>>

Re: [ovs-dev] [PATCH ovn] northd: Respect --ecmp-symmetric-reply for single routes.

On 9/17/24 19:00, Rosemarie O'Riorden wrote: > When preparing to build ECMP and static route flows, routes are sorted > into unique routes (meaning they are not part of a group) or they are > added to EMCP groups. Then, ECMP route flows are built out of the > groups, and static route flows are buil

Re: [ovs-dev] OVN technical community meeting - Sep 16th

On 9/13/24 12:18, Dumitru Ceara wrote: > Hi everyone, > > Just a quick reminder for anyone interested in joining the next OVN > technical meeting. It's scheduled to happen Monday, September 16th, at > 3PM UTC. > Thanks to everyone who could join. For those who want

Re: [ovs-dev] RFC OVN: fabric integration

On 9/12/24 19:45, Roberto Bartzen Acosta wrote: > >          - The result of this synchronization is basically: > >          - SB->NB: creating/deleting/updating > Logical_Router_Static_Route > >          entries for learned routes in the Routing_Information_Base > > table (usin

Re: [ovs-dev] [PATCH ovn v2] controller: Cleanup FDB entries when a VIF goes down.

On 8/23/24 19:02, Shibir Basak wrote: > When a VIF port goes down, all FDB entries associated > with it are deleted. > This helps is reducing traffic outage due to stale FDB > entries present until the lsp is removed. > > Acked-by: Naveen Yerramneni > Signed-off-by: Shibir Basak > --- > v2: Reba

Re: [ovs-dev] [PATCH ovn] controller: Cleanup FDB entries when a VIF goes down.

t;> >>> On 16-Aug-2024, at 7:35 PM, Shibir Basak >>> wrote: >>> >>> >>> >>>> On 15-Aug-2024, at 1:20 AM, Dumitru Ceara wrote: >>>> >>>> !---

[ovs-dev] OVN technical community meeting - Sep 16th

On 8/8/24 10:21, Dumitru Ceara wrote: > > 2nd try: Monday, Sep 16th, 3PM UTC > > It brings us back to the 6-7 weeks cadence but I guess we can discuss > any technical issues on-list until then. > Hi everyone, Just a quick reminder for anyone interested in joining the

Re: [ovs-dev] [PATCH ovn branch-24.09 2/2] Prepare for 24.09.1.

On 9/12/24 20:18, Mark Michelson wrote: > The date shown in this patch reflects when I made the patch. I will > re-generate the patches before pushing so that the date will reflect the > actual release date of ovn 24.09.0. > > Signed-off-by: Mark Michelson > --- Acked-by: Dumi

Re: [ovs-dev] [PATCH ovn branch-24.09 1/2] Set release date for 24.09.0.

On 9/12/24 20:18, Mark Michelson wrote: > The date shown in this patch reflects when I made the patch. I will > re-generate the patches before pushing so that the date will reflect the > actual release date of ovn 24.09.0. > > Signed-off-by: Mark Michelson > --- Acked-by: Dumi

Re: [ovs-dev] [PATCH ovn] ovs: Move the submodule to the official v3.4.0 release.

On 9/9/24 19:35, Mark Michelson wrote: > Acked-by: Mark Michelson > Thanks, Ilya and Mark! Applied to main and 24.09. Regards, Dumitru > On 8/20/24 07:03, Ilya Maximets wrote: >> OVS v3.4.0 contains several bug fixes in comparison to the current >> submodule commit.  They may not be very impo

Re: [ovs-dev] [PATCH ovn] controller: Do not remove snat-ct-zone requested by the CMS.

On 9/10/24 19:25, Mark Michelson wrote: > Excellent, thanks Lorenzo! > > Acked-by: Mark Michelson > Thanks, Lorenzo, Ales and Mark! I applied this to main and 24.09 with a small change to the test (I think the original version had a chance of being flaky): diff --git a/tests/ovn-controller.at

  1   2   3   4   5   6   7   8   9   10   >