The match wasn't cleared which led to matches being appended together
and the ovn-controller failed to parse them.
Fixes: 3faadc76ad71 ("northd: Fix pmtud for non routed traffic.")
Signed-off-by: Ales Musil
---
northd/northd.c | 3 ++-
tests/ovn-northd.at | 24 +
On Wed, May 29, 2024 at 9:31 AM Dumitru Ceara wrote:
> On 5/29/24 09:03, Ales Musil wrote:
> > The match wasn't cleared which led to matches being appended together
> > and the ovn-controller failed to parse them.
> >
> > Fixes: 3faadc76ad71 ("northd:
review.
> On 5/3/24 03:26, Ales Musil wrote:
> > Use the newly added NAT match and priority column in logical flows.
> > This allows to differentiate between various scenarios and more
> > fine-grained control over the resulting translation. The flows with
> > t
The match wasn't cleared which led to matches being appended together
and the ovn-controller failed to parse them.
Fixes: 3faadc76ad71 ("northd: Fix pmtud for non routed traffic.")
Signed-off-by: Ales Musil
---
northd/northd.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
us if there is any
> utility in printing the priority as well?
>
>
I was thinking about that, however the output seems to be pretty crowded as
is, so I'm not really sure whether it is a good idea. If others disagree I
can include it.
>
> On 5/3/24 03:26, Ales Musil wrote:
> > Ad
On Mon, May 27, 2024 at 9:00 AM Ales Musil wrote:
> Move more code into the new ct-zone module and encapsulate
> functionality that is strictly related to CT zone handling.
>
> Signed-off-by: Ales Musil
> ---
> controller/ct-z
Move the CT zone handling specific bits into it's own module. This
allows for easier changes done within the module and separates the
logic that is unrelated from ovn-controller.
Signed-off-by: Ales Musil
---
controller/automake.mk | 4 +-
controller/ct-zone.c| 377
In order to be able to store CT limits for specified zone, store the
zone inside separate struct instead of simap. This allows to add
the addition of limit without chaning the whole infrastructure again.
This is a preparation step for the CT zone limits.
Signed-off-by: Ales Musil
---
v2: Fix
Move more code into the new ct-zone module and encapsulate
functionality that is strictly related to CT zone handling.
Signed-off-by: Ales Musil
---
controller/ct-zone.c| 156 +---
controller/ct-zone.h| 8 +-
controller/ovn-controller.c | 49
.
The value 0 means unlimited, when the value is not specified it is
derived from OvS default CT limit specified for given OvS datapath.
Reported-at: https://bugzilla.redhat.com/2189924
Signed-off-by: Ales Musil
---
NEWS| 3 +
controller/ct-zone.c| 170
refactor of the CT zone handling logic
which allows us to get the zone limiting more easily.
Ales Musil (4):
controller: Move CT zone handling into separate module.
controller: Further encapsulate the CT zone handling.
controller: Prepare structure around CT zone limiting.
controller, northd
.
The value 0 means unlimited, when the value is not specified it is
derived from OvS default CT limit specified for given OvS datapath.
Reported-at: https://bugzilla.redhat.com/2189924
Signed-off-by: Ales Musil
---
NEWS| 3 +
controller/ct-zone.c| 170
Move the CT zone handling specific bits into it's own module. This
allows for easier changes done within the module and separates the
logic that is unrelated from ovn-controller.
Signed-off-by: Ales Musil
---
controller/automake.mk | 4 +-
controller/ct-zone.c| 377
In order to be able to store CT limits for specified zone, store the
zone inside separate struct instead of simap. This allows to add
the addition of limit without chaning the whole infrastructure again.
This is a preparation step for the CT zone limits.
Signed-off-by: Ales Musil
Move more code into the new ct-zone module and encapsulate
functionality that is strictly related to CT zone handling.
Signed-off-by: Ales Musil
---
controller/ct-zone.c| 156 +---
controller/ct-zone.h| 8 +-
controller/ovn-controller.c | 49
refactor of the CT zone handling logic
which allows us to get the zone limiting more easily.
Ales Musil (4):
controller: Move CT zone handling into separate module.
controller: Further encapsulate the CT zone handling.
controller: Prepare structure around CT zone limiting.
controller, northd
On Wed, May 22, 2024 at 9:59 AM Frode Nordahl wrote:
>
>
> On Wed, May 22, 2024 at 9:48 AM Ales Musil wrote:
>
>>
>>
>> On Mon, May 20, 2024 at 8:56 PM Frode Nordahl
>> wrote:
>>
>>> The ovs-vswitchd daemon may in some situations, typical
63, UDP_HEADER_LEN + 8);
> } else {
> -ovs_be32 ip4_src;
> -ip_parse(svc_mon->sb_svc_mon->src_ip, _src);
> -pinctrl_compose_ipv4(, eth_src, svc_mon->ea,
> - ip4_src,
> in6_addr_get_mapped_ipv4(_mon->ip),
> +pinctrl_compos
]
> --
> 2.45.1
>
> ___
> dev mailing list
> d...@openvswitch.org
> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>
>
Looks good to me, thanks.
Acked-by: Ales Musil
--
Ales Musil
Seni
st
> d...@openvswitch.org
> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>
>
Thanks,
Ales
--
Ales Musil
Senior Software Engineer - OVN Core
Red Hat EMEA <https://www.redhat.com>
amu...@redhat.com
<https://red.ht/sig>
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
| 4 +-
> tests/system-userspace-macros.at | 4 +-
> utilities/ovn-appctl.8.xml | 4 +-
> utilities/ovn-nbctl.c | 2 +-
> 25 files changed, 131 insertions(+), 124 deletions(-)
>
> --
> 2.44.0
>
>
ot;min": 0, "max": 1}}},
> "isRoot": true},
> "Logical_DP_Group": {
> "columns": {
> diff --git a/ovn-sb.xml b/ovn-sb.xml
> index 507a0b571..93a57cd06 100644
> --- a/ovn-sb.xml
> +++ b/ovn-sb.xml
> @
OVS_WAIT_UNTIL([ovn-sbctl list mac_binding | grep
> -q "192.168.10.10"])
> # The record is not deleted
> sleep 5
> AT_CHECK([ovn-sbctl list mac_binding | grep -q "192.168.10.10"])
> +check ovn-sbctl --all destroy mac_binding
> +
> +# Set the aging threshold on both routers and ensure tha
+--
> 3 files changed, 174 insertions(+), 12 deletions(-)
>
> --
> 2.31.1
>
> ___
> dev mailing list
> d...@openvswitch.org
> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>
>
The whole series looks goo
On Thu, May 16, 2024 at 5:57 PM Numan Siddique wrote:
> On Tue, May 14, 2024 at 4:41 AM Ales Musil wrote:
> >
> > The series is pretty small however it is required for the
> > bump to Ubuntu 24.04 and Fedora 40. Both have newer GCC and
> > Clang which brought
lid format
> -check ovn-nbctl set logical_router gw
> options:mac_binding_age_threshold="1;abc/26:0"
> +check ovn-nbctl set logical_router gw-1
> options:mac_binding_age_threshold="1;abc/26:0"
> check ovn-nbctl --wait=sb sync
> -uuid=$(fetch_column datapath _u
On Tue, May 14, 2024 at 3:19 PM Ilya Maximets wrote:
> On 5/14/24 10:38, Ales Musil wrote:
> > There were two things missing for the Fedora builds, 32-bit
> > version of glibc to allows the -m32 compilation on Fedora
> > and numactl-devel package.
> >
&g
There were two things missing for the Fedora builds, 32-bit
version of glibc to allows the -m32 compilation on Fedora
and numactl-devel package.
Signed-off-by: Ales Musil
---
.ci/linux-build.sh | 3 +++
utilities/containers/fedora/Dockerfile | 1 +
2 files changed, 4
Now that all failures were resolved for Fedora 40
we can bump the version in the container.
Signed-off-by: Ales Musil
---
utilities/containers/fedora/Dockerfile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/utilities/containers/fedora/Dockerfile
b/utilities/containers
wget2 has a bug and doesn't return proper exit code on error [0].
Replace wget with curl in places where we expect exit code to
be different from 0.
[0] https://gitlab.com/gnuwget/wget2/-/issues/652
Signed-off-by: Ales Musil
---
tests/system-ovn.at| 15
The container image for scheduled jobs was supposed to be Fedora,
however there was already Ubuntu image in the cache. Make sure
the cache is distinguished also by the event name.
Signed-off-by: Ales Musil
---
.github/workflows/test.yml | 5 +++--
1 file changed, 3 insertions(+), 2 deletions
The series is pretty small however it is required for the
bump to Ubuntu 24.04 and Fedora 40. Both have newer GCC and
Clang which brought up some issues that needed to be fixed.
The series also includes fix for weekly runs to use Fedora
because the cache string wasn't specific enough.
Ales Musil
when
it is just set to UTC. Setting it to Etc/UTC fixes the issue:
ValueError: ZoneInfo keys may not be absolute paths, got: /UTC
Signed-off-by: Ales Musil
---
utilities/containers/ubuntu/Dockerfile | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/utilities/containers
t;tests: Fix build failure with Clang 18 due to
-Wformat-truncation.")
Signed-off-by: Ales Musil
---
ovs | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/ovs b/ovs
index f19448b86..bf1b16364 16
--- a/ovs
+++ b/ovs
@@ -1 +1 @@
-Subproject commit f19448b8618967a108ec6f34713
Keep the Fedora version pinned to 40 for the build-rpm.
Fixes: 8f18b3b6c52f ("ci: Keep the container version pinned.")
Signed-off-by: Ales Musil
---
.github/workflows/test.yml | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/.github/workflows/test.yml b/.github
On Sun, May 12, 2024 at 11:17 PM Sri kor wrote:
> Hi @num...@ovn.org @Ales Musil ,
> Currently we are on OVN 23.09. We are facing the following syntax
> error with the address_set.From the diff, looks like you address this fix.
> Is this something expected?
>
> *2024-0
On Fri, May 10, 2024 at 1:34 PM Ilya Maximets wrote:
> On 5/10/24 12:57, Ales Musil wrote:
> > The bitmap used in the update_ct_zones was uninitialized, and it could
> > contain any value besides 0. Use the bitmap_allocate() function instead,
> > to allocate the bitmap in h
)
by 0x46954C: engine_run (inc-proc-eng.c:528)
by 0x40AE9D: main (ovn-controller.c:5776)
Uninitialised value was created by a stack allocation
at 0x440313: update_ct_zones (ovn-controller.c:747)
Fixes: f9cab11d5fab ("Allow explicit setting of the SNAT zone on a gateway
router.")
Signed-of
On Fri, May 10, 2024 at 11:23 AM Ilya Maximets wrote:
> On 5/10/24 10:44, Ales Musil wrote:
> > The bitmap used in the update_ct_zones was uninitialized and it could
> > contain any value besides 0, make sure we initialize it to 0 instead.
> > This was caught by valgrind
was created by a stack allocation
at 0x440313: update_ct_zones (ovn-controller.c:747)
Fixes: f9cab11d5fab ("Allow explicit setting of the SNAT zone on a gateway
router.")
Signed-off-by: Ales Musil
---
controller/ovn-controller.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
)
Reported-at: https://issues.redhat.com/browse/FDP-393
Acked-by: Mark Michelson
Signed-off-by: Ales Musil
Signed-off-by: Numan Siddique
(cherry picked from commit ddf051cbc6af24c303bf88970750e5c5fe285400)
---
controller/binding.c| 13 --
controller/ovn-controller.c | 27 +++
not properly remove flows when the last logical port of
a datapath is deleted.
Signed-off-by: Xavier Simonart
Acked-by: Ales Musil
Signed-off-by: Dumitru Ceara
(cherry picked from commit 69ec36eba074263c43051ade7578792b44518d2c)
---
tests/ovn-macros.at | 44
On Tue, May 7, 2024 at 6:49 PM Numan Siddique wrote:
> On Tue, May 7, 2024 at 2:25 AM Ales Musil wrote:
> >
> > There were two modules in controller mac_cache and mac-learn, both of
> > them did very similar thing with pretty big overlap. The goal of the
> > series
Merge mac-cache and mac-learn into single module. Both of those
modules contained very similar functionality with some small
differences. By merging those we have unified interface to deal
with FDB and MAC binding.
Acked-by: Mark Michelson
Signed-off-by: Ales Musil
---
v2: Rebase on top of main
to datapath's tunnel key requires somehow costly
synchronization when the tunnel key changes. However, this is fine as
the key shouldn't change very often in some cases it won't change at
all.
Also fix wrong check in the aging tests that would ignore failure.
Acked-by: Mark Michelson
Signed-off-by: Ales
For consistency rename the mac_cache.c/.h to mac-cache.c/.h.
Acked-by: Mark Michelson
Signed-off-by: Ales Musil
---
v2: Rebase on top of main.
---
controller/automake.mk | 4 ++--
controller/{mac_cache.c => mac-cache.c} | 2 +-
controller/{mac_cache.h => mac-cache.
)
Reported-at: https://issues.redhat.com/browse/FDP-393
Acked-by: Mark Michelson
Signed-off-by: Ales Musil
---
v2: Rebase on top of main.
---
controller/binding.c| 13 --
controller/ovn-controller.c | 27 +++
northd/northd.c | 7 +
tests/ovn.at
changed during runtime.
Ales Musil (4):
northd, controller: Handle tunnel_key change consistently.
controller: Rename mac_cache to to mac-cache.
controller: Merge the mac-cache and mac-learn.
controller: Use datapath key for the mac cache thresholds.
controller/automake.mk | 6
On Mon, May 6, 2024 at 8:41 PM Han Zhou wrote:
>
>
> On Thu, May 2, 2024 at 10:35 PM Ales Musil wrote:
> >
> > On Thu, May 2, 2024 at 6:23 PM Han Zhou wrote:
> > >
> > >
> > >
> > > On Thu, May 2, 2024 at 6:29 AM Ales Musil wrote:
&
node")
Reported-at: https://issues.redhat.com/browse/FDP-610
Signed-off-by: Ales Musil
---
controller/ovn-controller.c | 20 +--
tests/ovn-controller.at | 38 +
2 files changed, 48 insertions(+), 10 deletions(-)
diff --git a/controller/ovn-contr
of the flows with match, the priority
is calculated as 300 + priority column.
Reported-at: https://issues.redhat.com/browse/FDP-433
Signed-off-by: Ales Musil
---
northd/northd.c | 31 +++--
northd/ovn-northd.8.xml | 31 +
tests/ovn-northd.at | 79
tests/system-ovn.at
logical IP, but different external IP. Also,
when the match is specified allow addition of priority to order the
NAT rule valuation as needed.
Signed-off-by: Ales Musil
---
ovn-nb.ovsschema | 8 +-
ovn-nb.xml| 15 +++
tests/ovn-nbctl.at| 220
The priority calculation was scattered in multiple places which
could result in errors when the code is being updated. Move it
to common function that makes it very clear how is the priority
calculated.
Signed-off-by: Ales Musil
---
northd/northd.c | 82
re is also priority in addition to the match which controls the
evaluation order of the NAT with match, as the priority can be used
only in combination with match.
Ales Musil (3):
nothd: Unify the priority calculation for NAT flows.
nb: Add support for match and priority in NAT.
northd: Use the
t; state->paused = false;
> +poll_immediate_wake();
>
> unixctl_command_reply(conn, NULL);
> }
> --
> 2.31.1
>
> ___
> dev mailing list
> d...@openvswitch.org
> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>
onfiguration file
> + Mutually exclusive with
> --db-ic-nb-use-remote-in-db=yes.
> + --db-ic-sb-config-file=FILE OVN_IC_Southbound ovsdb-server
> configuration file
> + Mutually exclusive with
> --db-ic-sb-use-remote-in-d
1:5
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior lib/hash.c:46:22
[0] https://github.com/llvm/llvm-project/issues/90848
Fixes: db5a101931c5 ("clang: Fix the alignment warning.")
Signed-off-by: Ales Musil
---
v3: Do partial revert of db5a101931c5 instead of simple cast.
---
lib/ha
> +
> The first LTS version of OVN was 22.03. If you want to upgrade between
> other
> versions, you can use the `Fail-safe upgrade`_ procedure.
>
> --
> 2.41.0
>
> ___
> dev mailing list
> d...@openvswitch.org
>
__
> dev mailing list
> d...@openvswitch.org
> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>
>
Looks good to me, thanks.
Acked-by: Ales Musil
--
Ales Musil
Senior Software Engineer - OVN Core
Red Hat EMEA <https://www.redh
On Thu, May 2, 2024 at 6:23 PM Han Zhou wrote:
>
>
>
> On Thu, May 2, 2024 at 6:29 AM Ales Musil wrote:
> >
> > Instead of tracking address set per struct expr_constant_set track it
> > per individual struct expr_constant. This allows more fine grained
> > c
On Thu, May 2, 2024 at 8:03 PM Ilya Maximets wrote:
> On 5/2/24 15:28, Ales Musil wrote:
> > The pointer was passed to memcpy as uin32_t *, however the hash bytes
> > might be unaligned at that point. Case it to uint8_t * instead
>
> 'Case' ?
>
> > which
-at: https://issues.redhat.com/browse/FDP-243
Signed-off-by: Ales Musil
---
v4: Rebase on top of current main.
v3: Rebase on top of current main.
Fix the copy-paste error in ovn-controller documentation.
v2: Rebase on top of current main.
Make the probe interval accept milliseconds to be aligned
4.dst == $as2"
The memory impact should be minimal as there is only increase of 8 bytes
per the struct expr_constant.
Reported-at: https://issues.redhat.com/browse/FDP-509
Signed-off-by: Ales Musil
---
v4: Rebase on top of current main.
Update the "lflow_handle_addr_set_update&quo
0x507987 in parse_options ovs/ovsdb/ovsdb-server.c:2659:13
7 0x507987 in main ovs/ovsdb/ovsdb-server.c:751:5
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior lib/hash.c:46:22
[0] https://github.com/llvm/llvm-project/issues/90848
Signed-off-by: Ales Musil
---
lib/hash.c | 2 +-
lib/jhash.c
On Wed, May 1, 2024 at 6:38 PM Han Zhou wrote:
>
>
> On Tue, Apr 30, 2024 at 9:56 AM Ales Musil wrote:
> >
> > Instead of tracking address set per struct expr_constant_set track it
> > per individual struct expr_constant. This allows more fine grained
> > cont
On Thu, May 2, 2024 at 1:22 PM Ilya Maximets wrote:
> On 5/2/24 12:22, Ales Musil wrote:
> > The has was passed to memcpy as uin32_t *, however the hash bytes
>
> 'The has was passed' ? :)
>
Oops :)
>
> > might be unaligned at that point. Case it to uint8_t * instea
-off-by: Ales Musil
---
include/sparse/immintrin.h | 4
1 file changed, 4 insertions(+)
diff --git a/include/sparse/immintrin.h b/include/sparse/immintrin.h
index dd742be9f..36b41d352 100644
--- a/include/sparse/immintrin.h
+++ b/include/sparse/immintrin.h
@@ -26,5 +26,9 @@
#define
: UndefinedBehaviorSanitizer: undefined-behavior lib/hash.c:46:22
Signed-off-by: Ales Musil
---
lib/hash.c | 2 +-
lib/jhash.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/hash.c b/lib/hash.c
index c722f3c3c..986fa6643 100644
--- a/lib/hash.c
+++ b/lib/hash.c
@@ -43,7 +43,7 @@ hash_bytes
4.dst == $as2"
The memory impact should be minimal as there is only increase of 8 bytes
per the struct expr_constant.
Reported-at: https://issues.redhat.com/browse/FDP-509
Signed-off-by: Ales Musil
---
v3: Rebase on top of current main.
Address comments from Han:
- Adjust the com
The Ubuntu 24.04 brought some issues that are not really straight
forward to fix. Keep the Ubuntu version on 22.04 for now to keep
the CI working.
At the same time Fedora updated Clang to version 18, which is
throwing compilation error that need to be fixed in OvS first.
Signed-off-by: Ales
ompiler changes. */
> +#define _AVX512BF16INTRIN_H_INCLUDED
> +#define _AVX512BF16VLINTRIN_H_INCLUDED
> +#define _AVXNECONVERTINTRIN_H_INCLUDED
> +#define _KEYLOCKERINTRIN_H_INCLUDED
> +#define __AVX512FP16INTRIN_H_INCLUDED
> +#define
On Tue, Apr 30, 2024 at 3:31 PM Ales Musil wrote:
> The Ubuntu 24.04 brought some issues that are not realyl stright
>
typo: s/realyl stright/really straight :(
forward to fix. Keep the Ubuntu version on 22.04 for now to keep
> the CI working.
>
> At the same time Fedor
The Ubuntu 24.04 brought some issues that are not realyl stright
forward to fix. Keep the Ubuntu version on 22.04 for now to keep
the CI working.
At the same time Fedora updated Clang to version 18, which is
throwing compilation error that need to be fixed in OvS first.
Signed-off-by: Ales Musil
The Ubuntu 24.04 brought some issues that are not realyl stright
forward to fix. Keep the Ubuntu version on 22.04 for now to keep
the CI working.
Signed-off-by: Ales Musil
---
utilities/containers/ubuntu/Dockerfile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/utilities
On Mon, Apr 29, 2024 at 11:01 PM Han Zhou wrote:
> On Fri, Apr 12, 2024 at 8:20 AM Ales Musil wrote:
> >
> > Instead of tracking address set per struct expr_constant_set track it
> > per individual struct expr_constant. This allows more fine grained
> > control fo
_act_pedit(request, , keys_ex,
> flower->csum_update_flags ? TC_ACT_PIPE :
> action_pc);
>
> return 0;
> --
> 2.44.0
>
> ___
> dev mailing list
> d...@o
_cmdl_context *ctx
> OVS_UNUSED)
> ovs_assert(!strcmp(s, "abcd"));
> #if __GNUC__ >= 7
> #pragma GCC diagnostic pop
> +#endif
> +#if __clang_major__ >= 18
> +#pragma clang diagnostic pop
> #endif
>
> ovs_assert(snprintf(s, 6, "abcde"
gt;
> if [ "$CC" = "clang" ]; then
> set make CFLAGS="$CFLAGS -Wno-error=unused-command-line-argument"
> --
> 2.44.0
>
> ___
> dev mailing list
> d...@openvswitch.org
> https://mail.openvswitch.o
-at: https://issues.redhat.com/browse/FDP-243
Signed-off-by: Ales Musil
---
v4: Rebase on top of current main.
v3: Rebase on top of current main.
Fix the copy-paste error in ovn-controller documentation.
v2: Rebase on top of current main.
Make the probe interval accept milliseconds to be aligned
15 +0200, Dumitru Ceara wrote:
> > On 4/23/24 11:12, martin.kal...@canonical.com wrote:
> > > Hi Ales,
> > > Sorry that these new tests are causing problems. Just out of
> > > curiosity,
> > > do you have link to some failing test runs? I'll add few thou
IPv4 when
the there isn't any server address specified and fails to do so. Add
-6 flag to indicate that this is pure IPv6 connection.
[0] https://github.com/nmap/nmap/commit/4e6c8feb153c0c9ff8a68cd841669d650319ab45
Fixes: 40136a2f2c84 ("northd: Fix direct access to SNAT network.")
Signe
On Fri, Apr 19, 2024 at 6:38 PM Numan Siddique wrote:
>
>
> On Fri, Apr 19, 2024 at 7:09 AM Ales Musil wrote:
>
>> The current packet injection loses ct_state in the process. When
>> the ct_state is lost we might commit to DNAT zone and perform
>> zero S
p && ip4.src ==
>> 20.0.0.10 && outport == "DR-S3" && is_chassis_resident("cr-DR-S3") &&
>> (!ct.trk || !ct.rpl)), action=(ct_snat(192.168.0.10);)
>> ])
>>
>> +AT_CHECK([grep lr_out_post_snat lrflows | ovn_strip_lflows],
/* ct(nat) implies ct(). */
> if (!(ct_flow.ct_state & CS_TRACKED)) {
> -ct_flow.ct_state |= next_ct_state();
> +ct_flow.ct_state |= next_ct_state(action);
> }
>
> struct ovntrace_node *node = ovntrace_node_append(
> -super, OVNTRACE_NO
Merge mac-cache and mac-learn into single module. Both of those
modules contained very similar functionality with some small
differences. By merging those we have unified interface to deal
with FDB and MAC binding.
Signed-off-by: Ales Musil
---
controller/automake.mk | 2 -
controller
)
Reported-at: https://issues.redhat.com/browse/FDP-393
Signed-off-by: Ales Musil
---
controller/binding.c| 13 --
controller/ovn-controller.c | 27 +++
northd/northd.c | 7 +
tests/ovn.at| 52 +
to datapath's tunnel key requires somehow costly
synchronization when the tunnel key changes. However this is fine as
the key shouldn't change very often in some cases it won't change at
all.
Also fix wrong check in the aging tests that would ignore failure.
Signed-off-by: Ales Musil
---
controller/mac
For consistency rename the mac_cache.c/.h to mac-cache.c/.h.
Signed-off-by: Ales Musil
---
controller/automake.mk | 4 ++--
controller/{mac_cache.c => mac-cache.c} | 2 +-
controller/{mac_cache.h => mac-cache.h} | 2 +-
controller/ovn-controller.c | 2 +-
cont
changed during runtime.
Ales Musil (4):
northd, controller: Handle tunnel_key change consistently.
controller: Rename mac_cache to to mac-cache.
controller: Merge the mac-cache and mac-learn.
controller: Use datapath key for the mac cache thresholds.
controller/automake.mk | 6
iated
metering.")
Reported-at: https://issues.redhat.com/browse/FDP-439
Signed-off-by: Ales Musil
Acked-by: Mark Michelson
Signed-off-by: Numan Siddique
(cherry picked from commit c141d8b1961459a6e9a1834f33613d8be079310e)
---
controller/lflow.c | 1 +
controller/lflow.h
There are some improvements that are needed for further work on OVN
side. Most notably the commits mentioned below:
f4a4416b3807 ("netdev-dummy: Add local route entries for IP addresses.")
9719f64a34c0 ("ofproto-dpif-xlate: Fix continuations with associated metering.")
Signe
iated
metering.")
Reported-at: https://issues.redhat.com/browse/FDP-439
Signed-off-by: Ales Musil
Acked-by: Mark Michelson
Signed-off-by: Numan Siddique
(cherry picked from commit c141d8b1961459a6e9a1834f33613d8be079310e)
---
controller/lflow.c | 1 +
controller/lflow.h
/browse/FDP-571
Signed-off-by: Ales Musil
---
northd/northd.c | 5 +
tests/ovn-northd.at | 7 +++
2 files changed, 12 insertions(+)
diff --git a/northd/northd.c b/northd/northd.c
index 37f443e70..b834f1274 100644
--- a/northd/northd.c
+++ b/northd/northd.c
@@ -4689,6 +4689,11 @@ fail
the option as with the local_ip being enforced
also for single encap it became "true" in all scenarios, and it's not
needed anymore.
Reported-at: https://issues.redhat.com/browse/FDP-570
Signed-off-by: Ales Musil
---
NEWS| 3 +++
controller/encaps.c
iated
metering.")
Reported-at: https://issues.redhat.com/browse/FDP-439
Signed-off-by: Ales Musil
---
v5: Rebase on top of current main.
v4: Fix copy paste error in the global_handler.
v3: Rebase on top of current main.
Add flag to ensure that the paused action works during upgrade.
v2: Fi
t;ofproto-dpif-xlate: Fix continuations with associated metering.")
6448c1b697d7 ("vlog: Log stack trace on vlog_abort.")
Signed-off-by: Ales Musil
---
ovs | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/ovs b/ovs
index fe55ce37a..f19448b86 16
--- a/ovs
On Thu, Apr 18, 2024 at 11:45 AM Martin Kalcok
wrote:
> Thanks for another round of review.
>
> On Wed, Apr 17, 2024 at 1:45 PM Ales Musil wrote:
>
>>
>>
>> On Wed, Apr 17, 2024 at 1:05 PM Dumitru Ceara wrote:
>>
>>> On 4/17/24 09:04, Ales Musil
On Wed, Apr 17, 2024 at 1:05 PM Dumitru Ceara wrote:
> On 4/17/24 09:04, Ales Musil wrote:
> > On Mon, Apr 15, 2024 at 2:15 PM Martin Kalcok <
> martin.kal...@canonical.com>
> > wrote:
> >
> >> This change adds a new action 'ct_commit_to_zone' that ena
On Wed, Apr 17, 2024 at 9:09 AM Ales Musil wrote:
> The current packet injection loses ct_state in the process. When
> the ct_state is lost we might commit to DNAT zone and perform
> zero SNAT after the packet injection. This causes the first session
> to be wrong as the r
On Wed, Apr 17, 2024 at 9:09 AM Ales Musil wrote:
> The ovn-fake-multinode workflow can be triggered manually,
> however the definition didn't respect the branch for the manual
> run and always used main branch. Make sure that the correct
> branch is used for the ovn-fake-multin
1 - 100 of 1454 matches
Mail list logo
