On Tue, Oct 19, 2021 at 3:21 PM Numan Siddique wrote:
>
> On Tue, Oct 19, 2021 at 3:28 AM Han Zhou wrote:
> >
> > On Mon, Oct 18, 2021 at 6:35 AM Odintsov Vladislav
> > wrote:
> > >
> > >
> > >
> > > regards,
> > > Vladislav Odintsov
> > >
> > > > On 16 Oct 2021, at 03:20, Han Zhou wrote:
> > >
From: Anton Ivanov
Weak may fail for reasons unrelated to the comparison (not
on x86, where these are practically equivalent).
Thus, it can and should be used only in cases where there will
be a repeat of the event which runs the code (in this case
semaphore post).
On non-x86 architectures usin
When a PMD reload occurs, some PMD cycle measurements are reset.
In order to preserve the full cycles history of an Rxq, the RxQ
cycle measurements were not reset.
These are both used together to display the % of a PMD that an
RxQ is using in the pmd-rxq-show stat.
Resetting one and not the other
On 18/10/2021 17:25, Pai G, Sunil wrote:
Hi Kevin,
I wonder if the logs in the above function be INFO instead of DBG ?
Don't have a strong preference TBH. But if they were info, we need not
remove any test cases no ?
The reason they are debug is so they won't pollute the logs when a rebal
Hi Han,
could you take a look on this, please?
Regards,
Vladislav Odintsov
> On 5 Oct 2021, at 01:02, Vladislav Odintsov wrote:
>
> Hi,
>
> I’ve faced with a next issue using latest OVN (master branch) with OVN-IC
> enabled:
> ovn-northd CPU utilisation was at ~70-80% and ovsdb-server servin
On Tue, Oct 19, 2021 at 8:20 AM wrote:
>
> From: Mohammad Heib
>
> unixctl_client_transact allocate memory using xstrdup and
> this memory must be freed before exiting.
>
> Signed-off-by: Mohammad Heib
Thanks. I applied to the main branch.
Numan
> ---
> utilities/ovn-appctl.c | 3 ++-
> 1 f
On Fri, Oct 15, 2021 at 10:43 AM Mark Gray wrote:
>
> On 11/10/2021 17:52, mh...@redhat.com wrote:
> > From: Mohammad Heib
> >
> > Update old OVN tables numbers to match the correct tables number
> >
> > Signed-off-by: Mohammad Heib
> > ---
> > northd/northd.c | 6 +++---
> > northd/ovn
On Thu, Oct 14, 2021 at 9:22 AM Mark Gray wrote:
>
> When running on Fedora, it may be necessary to allow ESP and IKE traffic
> through the host firewall. If not, this will result in ICMP host
> unreachable messages:
>
> 13:52:10.000695 IP 192.168.122.228 > 192.168.122.125:
> ESP(spi=0xa5830a6b,s
On 20/10/2021 16:00, Numan Siddique wrote:
> On Thu, Oct 14, 2021 at 9:22 AM Mark Gray wrote:
>>
>> When running on Fedora, it may be necessary to allow ESP and IKE traffic
>> through the host firewall. If not, this will result in ICMP host
>> unreachable messages:
>>
>> 13:52:10.000695 IP 192.168
Thank you :)
On Wed, Oct 20, 2021 at 5:59 PM Numan Siddique wrote:
> On Fri, Oct 15, 2021 at 10:43 AM Mark Gray wrote:
> >
> > On 11/10/2021 17:52, mh...@redhat.com wrote:
> > > From: Mohammad Heib
> > >
> > > Update old OVN tables numbers to match the correct tables number
> > >
> > > Signed-
> Previously in OVS, a PMD thread running on cpu X used lcore X.
> This assumption limited OVS to run PMD threads on physical cpu <
> RTE_MAX_LCORE.
>
> DPDK 20.08 introduced a new API that associates a non-EAL thread to a free
> lcore. This new API does not change the thread characteristics (like
Hi Lin,
lin huang writes:
> From: linhuang
>
> Userspace tunnel doesn't have a valid device in the kernel. So
> get_ifindex() function (ioctl) always get error during
> adding a port, deleting a port or updating a port status.
>
> The info log is
> "2021-08-29T09:17:39.830Z|00059|netdev_linux|I
Hi Kevin,
On 10/20/21 11:04, Kevin Traynor wrote:
When a PMD reload occurs, some PMD cycle measurements are reset.
In order to preserve the full cycles history of an Rxq, the RxQ
cycle measurements were not reset.
These are both used together to display the % of a PMD that an
RxQ is using in th
ovsdb-tool join-cluster requires a remote addr, so the existing
code that tried to join a cluster without it when there was an
existing $DB_FILE would fail.
Instead, if we are trying to specifically join a cluster and there
is an existing $DB_FILE, back it up and remove the original before
continu
On Wed, Oct 20, 2021 at 2:45 PM Terry Wilson wrote:
>
> ovsdb-tool join-cluster requires a remote addr, so the existing
> code that tried to join a cluster without it when there was an
> existing $DB_FILE would fail.
>
> Instead, if we are trying to specifically join a cluster and there
> is an ex
On Thu, Oct 14, 2021 at 11:17 AM Lorenzo Bianconi
wrote:
>
> Rely on shash_add_nocopy instead of shash_add in ovs_iface_create in
> order to avoid allocating iface_id twice.
>
> Acked-by: Dumitru Ceara
> Signed-off-by: Lorenzo Bianconi
Thanks. Applied to the main branch.
Numan
> ---
> contr
On Mon, Oct 18, 2021 at 7:36 AM Dumitru Ceara wrote:
>
> On 10/14/21 5:17 PM, Lorenzo Bianconi wrote:
> > Introduce memory accounting for data structures in ovn-controller
> > if_status_mgr module.
> >
> > Signed-off-by: Lorenzo Bianconi
> > ---
>
> Acked-by: Dumitru Ceara
Thanks. Applied this
ovsdb-tool join-cluster requires a remote addr, so the existing
code that tried to join a cluster without it when there was an
existing $DB_FILE would fail.
Instead, if we are trying to specifically join a cluster and there
is an existing $DB_FILE, back it up and remove the original before
continu
Currently the ovs-tcpdump utility creates a virtual tunnel to send
packets to. This method functions perfectly fine, however, it can
greatly impact the performance of the monitored port.
It has been reported to reduce packet throughput significantly. I was
able to reproduce a reduction in throughp
Currently the ovs-tcpdump utility creates a virtual tunnel to send
packets to. This method functions perfectly fine, however, it can greatly
impact performance of the monitored port.
It has been reported to reduce packet throughput significantly. I was
able to reproduce a reduction in throughput o
Currently when a user creates an openflow group with with multiple
buckets without specifying a selection type, the efficient dp_hash is
only selected if the user is creating fewer than 64 buckets. But when
dp_hash is explicitly selected, up to 256 buckets are supported.
While up to 64 buckets see
On Wed, Oct 20, 2021 at 3:48 AM wrote:
>
> From: Anton Ivanov
>
> Weak may fail for reasons unrelated to the comparison (not
> on x86, where these are practically equivalent).
>
> Thus, it can and should be used only in cases where there will
> be a repeat of the event which runs the code (in thi
On Mon, Oct 11, 2021 at 9:11 AM Mark Gray wrote:
>
> On 11/10/2021 13:57, anton.iva...@cambridgegreys.com wrote:
> > From: Anton Ivanov
> >
> > 1. Compute dp group hash only if there will be dp group processing.
> > 2. Remove hmapx interim storage and related hmapx computation for
> > single dp f
On Tue, Oct 19, 2021 at 6:15 AM Frode Nordahl
wrote:
>
> Introduce infrastructure for plug providers and add feature to
> ovn-controller to add and remove ports on the integration bridge as
> directed by CMS through Logical_Switch_Port options.
>
> Traditionally it has been the CMSs responsibility
On Wed, Oct 20, 2021 at 9:32 PM Numan Siddique wrote:
>
> On Tue, Oct 19, 2021 at 6:15 AM Frode Nordahl
> wrote:
> >
> > Introduce infrastructure for plug providers and add feature to
> > ovn-controller to add and remove ports on the integration bridge as
> > directed by CMS through Logical_Switc
On Mon, Oct 18, 2021 at 5:14 AM Mark Gray wrote:
>
> Refactor ENGINE_NODE() macro to not assign function pointers. This
> allows ENGINE_NODE() to be used outside functions, creating an
> engine_node with global scope (but can be statically defined within a
file).
> This allows more flexibility in
On Mon, Oct 18, 2021 at 5:14 AM Mark Gray wrote:
>
> Initial implementation adds a single node (northd). This single
> node executes the northd processing pipeline but does not do so
> incrementally.
>
> In order to develop incremental processing for northd, the code
> will be organised with a .c/
27 matches
Mail list logo
