On Fri, Mar 15, 2024 at 1:15 PM Ilya Maximets wrote:
>
> Consider the following sequence of events:
>
> 1. Cluster with 2 nodes - A and B. A is a leader.
> 2. C connects to A and sends a join request.
> 3. A sends an append request to C. C is in CATCHUP phase for A.
> 4. A looses leadership
On Fri, Mar 15, 2024 at 1:15 PM Ilya Maximets wrote:
>
> Some of the failure tests can make a single-node cluster to
> loose leadership. In this case the next raft_run() will
> trigger election with a pre-vore enabled. This is causing
s/pre-vore/pre-vote
> an assertion when this server
On Fri, Mar 15, 2024 at 1:15 PM Ilya Maximets wrote:
>
> Consider the following chain of events:
>
> 1. Have a cluster with 2 members - A and B. A is a leader.
> 2. C connects to A, sends a request to join the cluster.
> 3. A catches up C, creates an update for the 'servers' list and sends
>
On Fri, Mar 15, 2024 at 1:15 PM Ilya Maximets wrote:
>
> While joining, ovsdb-server may not wake up for a duration of a join
> timer, which is 1 second and is by default 3x larger than a heartbeat
> timer. This is causing unnecessary warnings from the cooperative
> multitasking module that
Bleep bloop. Greetings Chris Mi, 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 111 characters long (recommended limit is 79)
#39 FILE:
Add three sFlow offload test cases:
3: offloads - sflow with sampling=1 - offloads enabled ok
4: offloads - sflow with sampling=2 - offloads enabled ok
5: offloads - ping over vxlan tunnel with sflow - offloads enabled ok
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
Acked-by: Eelco
Create a unique group ID to map the sFlow info when offloading sample
action to TC. When showing the offloaded datapath flows, translate the
group ID from TC sample action to sFlow info using the mapping.
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
---
NEWS| 2 +
In thread handler 0, add netdev offload recv in normal recv upcalls.
To avoid starvation, introduce a flag to alternate the order of
receiving normal upcalls and offload upcalls based on that flag.
Add similar change for recv_wait.
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
---
Initialize psample socket. Add sample recv API to receive sampled
packets from psample socket. Add sample recv wait API to add psample
socket fd to poll list.
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
---
lib/dpif.h| 6 +-
lib/flow.h| 2 +-
Iterate each registered offload API. It's not a problem for today
since we only have one implementation.
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
---
lib/netdev-offload.c | 38 ++
lib/netdev-offload.h | 5 +
2 files changed, 43 insertions(+)
diff
When offloading sample action to TC, userspace creates a unique ID
to map sample action and tunnel info and passes this ID to kernel
instead of the sample info. Kernel will send this ID and sampled
packet to userspace. Using the ID, userspace can recover the sample
info and send sampled packet to
Update kernel UAPI to support psample and the tc sample action.
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
Acked-by: Eelco Chaudron
---
include/linux/automake.mk| 10 +++---
include/linux/psample.h | 62
include/linux/tc_act/tc_sample.h |
Load kernel module psample to receive sampled packets from TC.
Before removing kernel module psample, remove act_sample first.
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
Acked-by: Eelco Chaudron
---
utilities/ovs-kmod-ctl.in | 14 ++
1 file changed, 14 insertions(+)
diff
This patch set adds offload support for sFlow.
Psample is a genetlink channel for packet sampling. TC action act_sample
uses psample to send sampled packets to userspace.
When offloading sample action to TC, userspace creates a unique ID to
map sFlow action and tunnel info and passes this ID to
On 6/24/2023 4:18 AM, Ilya Maximets wrote:
On 6/19/23 07:05, Chris Mi wrote:
This patch set adds offload support for sFlow.
Psample is a genetlink channel for packet sampling. TC action act_sample
uses psample to send sampled packets to userspace.
When offloading sample action to TC,
On 6/24/2023 4:18 AM, Ilya Maximets wrote:
On 6/19/23 07:05, Chris Mi wrote:
Add three sFlow offload test cases:
3: offloads - sflow with sampling=1 - offloads enabled ok
4: offloads - sflow with sampling=2 - offloads enabled ok
5: offloads - ping over vxlan tunnel with sflow -
On 6/24/2023 4:18 AM, Ilya Maximets wrote:
On 6/19/23 07:05, Chris Mi wrote:
In thread handler 0, add netdev offload recv in normal recv upcalls.
To avoid starvation, introduce a flag to alternate the order of
receiving normal upcalls and offload upcalls based on that flag.
Add similar change
On 6/24/2023 4:18 AM, Ilya Maximets wrote:
On 6/19/23 07:05, Chris Mi wrote:
Create a unique group ID to map the sFlow info when offloading sample
action to TC. When showing the offloaded datapath flows, translate the
group ID from TC sample action to sFlow info using the mapping.
On 6/24/2023 4:18 AM, Ilya Maximets wrote:
On 6/19/23 07:05, Chris Mi wrote:
Iterate each registered offload API. It's not a problem for today
since we only have one implementation.
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
---
lib/netdev-offload.c | 38
On 6/24/2023 4:17 AM, Ilya Maximets wrote:
On 6/19/23 07:05, Chris Mi wrote:
Initialize psample socket. Add sample recv API to receive sampled
packets from psample socket. Add sample recv wait API to add psample
socket fd to poll list.
Signed-off-by: Chris Mi
Reviewed-by: Roi Dayan
---
On 6/24/2023 4:17 AM, Ilya Maximets wrote:
On 6/19/23 07:05, Chris Mi wrote:
When offloading sample action to TC, userspace creates a unique ID
to map sample action and tunnel info and passes this ID to kernel
instead of the sample info. Kernel will send this ID and sampled
packet to userspace.
The ethernet addresses of two ICMP request packets are indeed different . One
is original packet and the other is modified. It is an expected behavior
according to the code.
Actually, when a packet sent by port A is changed by flow table and then is
sent to itself, we expect to capture this
On Mon, 25 Mar 2024 19:45:05 +0200 Andy Shevchenko wrote:
> > +/* Non-parallel generic netlink requests are serialized by a global lock.
> > */
>
> While at it, maybe drop extra space? (I noticed it was originally like this.
Fair, I'll post v3.
> > +#define MODULE_ALIAS_GENL_FAMILY(family)
On Mon, Mar 25, 2024 at 10:37:16AM -0700, Jakub Kicinski wrote:
> genetlink.h is a shell of what used to be a combined uAPI
> and kernel header over a decade ago. It has fewer than
> 10 lines of code. Merge it into net/genetlink.h.
> In some ways it'd be better to keep the combined header
> under
The "pass" verdict was not explicitly defined in the list of verdicts
for ACL logging. This resulted in logs saying "Syntax error at `pass'
unknown verdict."
This change adds the "pass" verdict explicitly so that it shows up as a
proper log in ovn-controller.
Reported-at:
The only legit reason I could think of for net/genetlink.h
and linux/genetlink.h to be separate would be if one was
included by other headers and we wanted to keep it lightweight.
That is not the case, net/openvswitch/meter.h includes
linux/genetlink.h but for no apparent reason (for struct
genetlink.h is a shell of what used to be a combined uAPI
and kernel header over a decade ago. It has fewer than
10 lines of code. Merge it into net/genetlink.h.
In some ways it'd be better to keep the combined header
under linux/ but it would make looking through git history
harder.
Acked-by:
Bleep bloop. Greetings Jacob Tanenbaum, 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: The subject summary should start with a capital.
WARNING: The subject summary
Bleep bloop. Greetings Jacob Tanenbaum, 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: The subject summary should end with a dot.
Subject: Merge QoS logical pipelines
currently there are 2 QoS pipelines for ingress (ls_in_qos_mark,
ls_in_qos_meter) and egress (ls_out_qos_mark, ls_out_qos_meter). This is
not necessary as there are no actions across the two pipelines that
depend on each other. The two pipelines can be merged.
commit: 1807499 changed the testing to get the table numbers themselves.
The table number was left in here. This patch corrects that and removes
the table number.
Signed-off-by: Jacob Tanenbaum
---
tests/ovn.at | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/ovn.at
On Mon, Mar 25, 2024 at 3:48 AM Zhangweiwei wrote:
>
> Hi,
> I have tried this patch, however, there are still some issues when the
> packets contents are changed across recirculation. On the follow example,
> packets are modified in recirc_id(0) after mirror, the mirror context reset.
>
On 25 Mar 2024, at 13:37, Ilya Maximets wrote:
> On 3/25/24 13:22, Aaron Conole wrote:
>> Eelco Chaudron writes:
>>
>>> On 22 Mar 2024, at 20:06, Aaron Conole wrote:
>>>
Open vSwitch is originally intended to switch at layer 2, only dealing with
Ethernet frames. With the
On 3/25/24 08:19, Han Zhou wrote:
>
>
> On Tue, Mar 19, 2024 at 12:05 AM Felix Huettner via dev
> mailto:ovs-dev@openvswitch.org>> wrote:
>>
>> On Mon, Mar 18, 2024 at 05:52:12PM +0100, Ilya Maximets wrote:
>> > On 3/18/24 17:15, Felix Huettner wrote:
>> > > On Fri, Mar 15, 2024 at 09:14:49PM
On 3/25/24 13:22, Aaron Conole wrote:
> Eelco Chaudron writes:
>
>> On 22 Mar 2024, at 20:06, Aaron Conole wrote:
>>
>>> Open vSwitch is originally intended to switch at layer 2, only dealing with
>>> Ethernet frames. With the introduction of l3 tunnels support, it crossed
>>> into the realm of
Eelco Chaudron writes:
> On 22 Mar 2024, at 20:06, Aaron Conole wrote:
>
>> Open vSwitch is originally intended to switch at layer 2, only dealing with
>> Ethernet frames. With the introduction of l3 tunnels support, it crossed
>> into the realm of needing to care a bit about some routing
Ilya Maximets writes:
> On 3/22/24 14:40, Aaron Conole wrote:
>> Open vSwitch supports the ability to invoke a controller action by way
>> of a sample action with a specified meter. In the normal case, this
>> sample action is transparently generated during xlate processing. However,
>> when
The ovn-ctl utility was assuming that the user/group is always root,
when not specified otherwise by the --ovn-user/--ovn-group options.
This has the consequence of trying to change permissions of OVN
directories to root:root even though the script might be run as
completely different user.
Take
Hi,
I have tried this patch, however, there are still some issues when the packets
contents are changed across recirculation. On the follow example, packets are
modified in recirc_id(0) after mirror, the mirror context reset. Therefore,
there are two ICMP request packets are mirrored on port
On Mon, Mar 25, 2024 at 9:47 AM Dumitru Ceara wrote:
> On 3/22/24 19:34, Mark Michelson wrote:
> > On 3/21/24 19:15, Dumitru Ceara wrote:
> >> Most of the steps were inaccurate. Instead, use latest Ubuntu, use
> >> OVS from the submodule inside the OVN repo.
> >>
> >> Signed-off-by: Dumitru
On 3/22/24 19:34, Mark Michelson wrote:
> On 3/21/24 19:15, Dumitru Ceara wrote:
>> Most of the steps were inaccurate. Instead, use latest Ubuntu, use
>> OVS from the submodule inside the OVN repo.
>>
>> Signed-off-by: Dumitru Ceara
>> ---
>> utilities/docker/Makefile | 4 ++--
>>
On 22 Mar 2024, at 20:06, Aaron Conole wrote:
> Open vSwitch is originally intended to switch at layer 2, only dealing with
> Ethernet frames. With the introduction of l3 tunnels support, it crossed
> into the realm of needing to care a bit about some routing details when
> making forwarding
On Tue, Mar 19, 2024 at 12:05 AM Felix Huettner via dev <
ovs-dev@openvswitch.org> wrote:
>
> On Mon, Mar 18, 2024 at 05:52:12PM +0100, Ilya Maximets wrote:
> > On 3/18/24 17:15, Felix Huettner wrote:
> > > On Fri, Mar 15, 2024 at 09:14:49PM +0100, Ilya Maximets wrote:
> > >> Each cluster member
43 matches
Mail list logo