On Thu, Aug 8, 2024 at 1:55 PM Brian Haley wrote:
> Hi Ilya and others,
>
> Thanks for responding, comments below.
>
> On 8/2/24 7:59 AM, Ilya Maximets wrote:
> > On 6/6/24 18:02, Ihar Hrachyshka wrote:
> >> On Thu, Jun 6, 2024 at 11:12 AM Ihar Hrachyshka <m
Thanks. (Confirmed the other test case that covers tunnel exhaustion
scenarios already uses `eval`.)
Reviewed-By: Ihar Hrachyshka
On Mon, Jul 15, 2024 at 5:17 AM Ales Musil wrote:
> The test tries to setup 4097 LSs via single ovn-nbctl command,
> this is fine however it might l
On Thu, Jun 6, 2024 at 1:27 AM Vladislav Odintsov wrote:
> Thanks Mark for such a detailed answer!
>
> I agree with your points, and also was thinking about them, but could not
> value their importance in terms of I-P logic. You helped with that.
>
> I’d prefer to apply my proposal to revert back
On Thu, Jun 6, 2024 at 11:12 AM Ihar Hrachyshka wrote:
> On Wed, Jun 5, 2024 at 11:17 AM Brian Haley wrote:
>
>> Hi Ilya,
>>
>> On 6/4/24 6:27 AM, Ilya Maximets wrote:
>> > On 6/4/24 11:54, Alin Serdean wrote:
>> >> Does it make sense to make this c
On Wed, Jun 5, 2024 at 11:17 AM Brian Haley wrote:
> Hi Ilya,
>
> On 6/4/24 6:27 AM, Ilya Maximets wrote:
> > On 6/4/24 11:54, Alin Serdean wrote:
> >> Does it make sense to make this configurable via automake in order to
> avoid
> >> future patches if we need to bump the value further?
> >
> > I
On Thu, May 2, 2024 at 5:52 PM Ilya Maximets wrote:
> On 4/26/24 18:54, Ihar Hrachyshka wrote:
> > Remove the notion of cluster/leave --force since it was never
> > implemented. Instead of these instructions, document how a broken
> > cluster can be re-initialized with the
least its stated goal - to simplify the code). I commented
there, and I would like others to vote on patch 1/2.
For this patch (2/2), with capitalized VXLAN mode in docs:
Acked-By: Ihar Hrachyshka
On Thu, May 2, 2024 at 5:51 AM Vladislav Odintsov wrote:
> Commit [1] introduced a "vx
On Thu, May 2, 2024 at 5:51 AM Vladislav Odintsov wrote:
> This simplifies code and subsequent commit to explicitely disable vxlan
>
I personally find it debatable that moving from explicit dependency through
a function argument to implicit dependency through a global variable is a
simplificatio
On Sat, Apr 27, 2024 at 2:14 AM Vladislav Odintsov
wrote:
> Hi Dumitru, Ihar,
>
> Thanks for the review. PSB.
>
> regards,
> Vladislav Odintsov
>
> On 26 Apr 2024, at 23:13, Ihar Hrachyshka wrote:
>
>
> Hi Vladislav,
>
> This looks good to me. I think
asynchronously.
[1]:
https://pubs.opengroup.org/onlinepubs/9699919799.2016edition/functions/connect.html
[2]: see `connect(2)` on Linux for details
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util-unix.c | 8 +--
lib/socket-util.c | 2 +-
lib/socket
The wording above is not completely clear without these scenarios
listed. A confused reader may incorrectly read it as:
```
Only LTS-to-LTS is supported for rolling upgrades.
```
which is wrong.
Signed-off-by: Ihar Hrachyshka
---
Documentation/intro/install/ovn-upgrades.rst | 5 +
1 file
The dependency is during build time, not runtime.
Signed-off-by: Ihar Hrachyshka
---
Documentation/intro/install/ovn-upgrades.rst | 18 --
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/Documentation/intro/install/ovn-upgrades.rst
b/Documentation/intro/install
Hi Vladislav,
This looks good to me. I think documentation should be adjusted to not
explain in detail ID allocations for VXLAN in two separate places. Assuming
the documentation is de-duplicated and typos Dumitru spotted are addressed,
Acked-by: Ihar Hrachyshka
On Thu, Apr 4, 2024 at 2:41 PM
It's always NULL.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 2fef95f38..5e12fd1e8 100644
--- a/northd/northd.c
+++ b/northd/northd.c
@@ -4364,7 +4364,6 @@ ls_port_c
It's always NULL.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 934c22dac..2fef95f38 100644
--- a/northd/northd.c
+++ b/northd/northd.c
@@ -4342,7 +4342,7 @@ ls_port
This allows callers to avoid cleanup of the record in case the function
fails.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 24
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 4800a44bd..934c22dac 100644
Let the callers do the cleanup as needed.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 24 +++-
1 file changed, 15 insertions(+), 9 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index b541e8455..3bdd0648c 100644
--- a/northd/northd.c
+++ b/northd
By: Vladislav Odintsov
Signed-off-by: Vladislav Odintsov
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-northd.at | 43 ---
1 file changed, 40 insertions(+), 3 deletions(-)
diff --git a/tests/ovn-northd.at b/tests/ovn-northd.at
index 27183c551..680d96675
;list is never left attached to a list, we can remove
ovs_list_remove calls for it elsewhere, including where op was never
attached in the first place.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 23 +--
1 file changed, 13 insertions(+), 10 deletions(-)
diff --git
op->list in build_ports (and never elsewhere.)
v7: simplify pb cleanup logic in ls_port_init.
Ihar Hrachyshka (6):
northd: Don't cleanup op in ovn_port_allocate_key.
northd: Don't detach op->list when it wasn't used.
tests: Correct tunnel ids exhaustion scenario.
northd: Don
On Tue, Apr 23, 2024 at 4:59 PM Mark Michelson wrote:
> On 4/11/24 21:57, Ihar Hrachyshka wrote:
> > This allows callers to avoid cleanup of the record in case the function
> > fails.
> >
> > Signed-off-by: Ihar Hrachyshka
> > ---
> > northd/northd.c
Remove the notion of cluster/leave --force since it was never
implemented. Instead of these instructions, document how a broken
cluster can be re-initialized with the old database contents.
Signed-off-by: Ihar Hrachyshka
---
v1: initial version.
v2: remove --force mentioned in ovsdb-server(1
On Fri, Apr 12, 2024 at 10:33 AM Dumitru Ceara wrote:
> On 4/11/24 10:32, Ales Musil wrote:
> > On Thu, Apr 4, 2024 at 11:44 PM Ihar Hrachyshka
> wrote:
> >
> >> Strictly speaking, this is not *essential* to start from MIN and not
> >> MIN+1 (once the hint r
Remove the notion of cluster/leave --force since it was never
implemented. Instead of these instructions, document how a broken
cluster can be re-initialized with the old database contents.
Signed-off-by: Ihar Hrachyshka
---
v1: initial version.
v2: remove --force mentioned in ovsd-server(1
]
https://github.com/ovn-org/ovn/commit/0baca3e519756cbe98a32526ccc637bb73468743
[2] https://mail.openvswitch.org/pipermail/ovs-discuss/2024-April/053049.html
Signed-off-by: Ihar Hrachyshka
Acked-by: Eelco Chaudron
---
lib/socket-util.c| 2 +-
python/ovs/stream.py | 2 +-
2 files changed, 2
Remove the notion of cluster/leave --force since it was never
implemented. Instead of these instructions, document how a broken
cluster can be re-initialized with the old database contents.
Signed-off-by: Ihar Hrachyshka
---
Documentation/ref/ovsdb.7.rst | 50
It's always NULL.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 6a8ace52f..78028131f 100644
--- a/northd/northd.c
+++ b/northd/northd.c
@@ -4336,7 +4336,7 @@ ls_port
It's always NULL.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 78028131f..6a51e5bad 100644
--- a/northd/northd.c
+++ b/northd/northd.c
@@ -4358,7 +4358,6 @@ ls_port_c
By: Vladislav Odintsov
Signed-off-by: Vladislav Odintsov
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-northd.at | 43 ---
1 file changed, 40 insertions(+), 3 deletions(-)
diff --git a/tests/ovn-northd.at b/tests/ovn-northd.at
index be006fb32..1a4e7274d
This allows callers to avoid cleanup of the record in case the function
fails.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 4cea669cf..6a8ace52f 100644
--- a/northd
;list is never left attached to a list, we can remove
ovs_list_remove calls for it elsewhere, including where op was never
attached in the first place.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 23 +--
1 file changed, 13 insertions(+), 10 deletions(-)
diff --git
Let the callers do the cleanup as needed.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 24 +++-
1 file changed, 15 insertions(+), 9 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 02cf5b234..f2406890c 100644
--- a/northd/northd.c
+++ b/northd
op->list in build_ports (and never elsewhere.)
Ihar Hrachyshka (6):
northd: Don't cleanup op in ovn_port_allocate_key.
northd: Don't detach op->list when it wasn't used.
tests: Correct tunnel ids exhaustion scenario.
northd: Delete pb if tunnel is not allocated.
n
Sorry, this is incorrect, the order of steps here is important. Instead, we
should clean up the newly created record on failure. (Other patches in the
series are ok; will send an updated series.)
On Thu, Apr 11, 2024 at 6:59 PM Ihar Hrachyshka wrote:
> This allows to avoid cleanup of the rec
Series with this patch that also fixes ubsan failure here:
https://patchwork.ozlabs.org/project/ovn/list/?series=402694
On Mon, Apr 8, 2024 at 9:49 AM Ihar Hrachyshka wrote:
> Note to reviewers: looks like the port tunnel id test case revealed a
> number of undefined behaviors and leaks
It's always NULL.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index adc8930c8..044a69264 100644
--- a/northd/northd.c
+++ b/northd/northd.c
@@ -4353,7 +4353,6 @@ ls_port_c
Let the caller do the cleanup, as needed.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 23 +--
1 file changed, 13 insertions(+), 10 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index f2406890c..ace663d54 100644
--- a/northd/northd.c
+++ b/northd
The `op` here is a new port structure that is not, yet, attached
anywhere, and the attempt to detach it will result in undefined
behavior.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/northd/northd.c b/northd/northd.c
index ace663d54
By: Vladislav Odintsov
Signed-off-by: Vladislav Odintsov
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-northd.at | 43 ---
1 file changed, 40 insertions(+), 3 deletions(-)
diff --git a/tests/ovn-northd.at b/tests/ovn-northd.at
index be006fb32..1a4e7274d
It's always NULL.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 5a0225189..adc8930c8 100644
--- a/northd/northd.c
+++ b/northd/northd.c
@@ -4331,7 +4331,7 @@ ls_port
This allows to avoid cleanup of the record in case tunnel id fails to
allocate.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 3d2715911..5a0225189 100644
--- a/northd
both cases of hint = 0 and hint > max.
v3: reduce the number of ports to create in the hint > max scenario needed to
trigger the problem.
v4: remove spurious lib/ovn-util.c change.
Ihar Hrachyshka (7):
northd: Don't cleanup op in ovn_port_allocate_key.
northd: Don't cleanup o
Let the callers do the cleanup as needed.
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 24 +++-
1 file changed, 15 insertions(+), 9 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 02cf5b234..f2406890c 100644
--- a/northd/northd.c
+++ b/northd
is fine.)
On Sat, Apr 6, 2024 at 2:30 AM Vladislav Odintsov wrote:
> Hi Ihar,
>
> Thanks for cooperation and enhancements in the testcases!
> The patch looks good to me.
>
> On 5 Apr 2024, at 19:14, Ihar Hrachyshka wrote:
>
> The original version of the scenario passed
By: Vladislav Odintsov
Signed-off-by: Vladislav Odintsov
Signed-off-by: Ihar Hrachyshka
---
v1: initial version.
v2: cover both cases of hint = 0 and hint > max.
v3: reduce the number of ports to create in the hint > max scenario needed to
trigger the problem.
v4: remove spurious li
Sorry for forgetting to include changelogs in these submissions.
v1: initial version.
v2: cover both cases of hint = 0 and hint > max.
v3: reduce the number of ports to create in the hint > max scenario needed
to trigger the problem.
On Fri, Apr 5, 2024 at 12:10 PM Ihar Hrachyshka wrote:
By: Vladislav Odintsov
Signed-off-by: Vladislav Odintsov
Signed-off-by: Ihar Hrachyshka
---
lib/ovn-util.c | 11 ---
tests/ovn-northd.at | 43 ---
2 files changed, 44 insertions(+), 10 deletions(-)
diff --git a/lib/ovn-util.c b/lib/ovn
On Thu, Apr 4, 2024 at 4:02 PM Vladislav Odintsov wrote:
>
>
> > On 4 Apr 2024, at 22:51, Mark Michelson wrote:
> >
> > On 4/4/24 12:46, Dumitru Ceara wrote:
> >> On 4/4/24 17:52, Vladislav Odintsov wrote:
> >>> Thanks Dumitru!
> >>> I’m totally fine with your change.
> >>> Should I send backpor
On Fri, Apr 5, 2024 at 11:38 AM Vladislav Odintsov
wrote:
>
>
> On 5 Apr 2024, at 18:35, Ihar Hrachyshka wrote:
>
> On Thu, Apr 4, 2024 at 3:56 PM Vladislav Odintsov
> wrote:
>
>> Thanks Ihar for the patch.
>>
>> It definitely triggers the bug mentioned
By: Vladislav Odintsov
Signed-off-by: Vladislav Odintsov
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-northd.at | 43 ---
1 file changed, 40 insertions(+), 3 deletions(-)
diff --git a/tests/ovn-northd.at b/tests/ovn-northd.at
index be006fb32..29c4a93a0
> OVS_WAIT_UNTIL([grep "all datapath tunnel ids exhausted"
> northd/ovn-northd.log])
>
>
> On 4 Apr 2024, at 20:13, Ihar Hrachyshka wrote:
>
> The original version of the scenario passed with or without the fix.
>
> Fixes: a1f165a7b807 ("northd: fix infi
d check that all "vxlan peer chassis" hosts only type=vtep
bindings... I agree that a config knob here is the way to go.
>
> regards,
> Vladislav Odintsov
>
> On 3 Apr 2024, at 20:44, Ihar Hrachyshka wrote:
>
>
> Thank you Vladislav.
>
> FYI it was repo
Strictly speaking, this is not *essential* to start from MIN and not
MIN+1 (once the hint reaches max, it will wrap back to MIN anyway), but
this is inconsistent with how we handle datapath and port keys (we start
with hint = 0 there).
Signed-off-by: Ihar Hrachyshka
---
northd/northd.c | 3
heck_row_count nb:Logical_Switch 4097
> +check_row_count nb:Logical_Switch 4095
> +wait_row_count sb:Datapath_Binding 4095
> +
> +ovn-nbctl ls-add lsw-exhausted
> +
> +check_row_count nb:Logical_Switch 4096
> wait_row_count sb:Datapath_Binding 4095
>
> OVS_WAIT_UNTIL([gr
e an option that statically sets vxlan mode is
> more resilient.
>
> In general we try to avoid new config knobs.
> .
> > What do you think?
> >
>
> But in this case it make actually be easier if we offload the work of
> determining vxlan-mode to the CMS.
>
>
On Thu, Apr 4, 2024 at 2:36 AM Eelco Chaudron wrote:
>
>
> On 3 Apr 2024, at 23:18, Ihar Hrachyshka wrote:
>
> > Before the patch, the size of the backlog depended on the type of socket
> > (UNIX vs INET) as well as on the language (C vs Python), specifically:
> >
On Thu, Apr 4, 2024 at 1:46 PM Dumitru Ceara wrote:
> On 4/4/24 19:17, Ihar Hrachyshka wrote:
> > I tried to revert the util change and the test case passed just fine.
> >
>
> I had done that before pushing the patch but.. I got tricked by the fact
> that northd was sp
I tried to revert the util change and the test case passed just fine.
I think the scenario that may get the hint out of bounds is 1) start with
no vxlan chassis; 2) create 4097 DPs; 3) add a vxlan chassis - this makes
northd downgrade its max key to 4096. Now when we create a DP, it will spin
in c
Sorry, missing 'ovn' in the subject. Resent to ovn queue.
On Thu, Apr 4, 2024 at 1:13 PM Ihar Hrachyshka wrote:
> The original version of the scenario passed with or without the fix.
>
> Fixes: a1f165a7b807 ("northd: fix infinite loop in ovn_allocate_tnlid()")
>
The original version of the scenario passed with or without the fix.
Fixes: a1f165a7b807 ("northd: fix infinite loop in ovn_allocate_tnlid()")
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-northd.at | 17 +++--
1 file changed, 11 insertions(+), 6 deletions(-)
diff --git a
The original version of the scenario passed with or without the fix.
Fixes: a1f165a7b807 ("northd: fix infinite loop in ovn_allocate_tnlid()")
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-northd.at | 17 +++--
1 file changed, 11 insertions(+), 6 deletions(-)
diff --git a
effectively bumps the
number of simultaneous connections to python unixctl servers to 64. Also
for INET C servers too.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util.c| 2 +-
python/ovs/stream.py | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/socket-util.c b
s support for explicit disabling of vxlan mode via
> Northbound database.
>
> 0: https://github.com/ovn-org/ovn/commit/b07f1bc3d068
>
> CC: Ihar Hrachyshka
> Fixes: b07f1bc3d068 ("Add VXLAN support for non-VTEP datapath bindings")
> Signed-off-by: Vladislav Odintso
On Tue, Jan 30, 2024 at 10:52 PM Naveen Yerramneni <
naveen.yerramn...@nutanix.com> wrote:
>
>
> > On 29-Jan-2024, at 9:11 PM, Ihar Hrachyshka wrote:
> >
> > On Mon, Jan 22, 2024 at 12:22 PM Naveen Yerramneni <
> naveen.yerramn...@nutanix.com> wrote:
> &
On Mon, Jan 22, 2024 at 12:22 PM Naveen Yerramneni <
naveen.yerramn...@nutanix.com> wrote:
> This option can be used to enable/disable arp/nd reply flows.
>
> Usecase:
> =
> It is useful to reduce packet loss when VM is being migrated to
>
It may indeed be useful to be able to disable ARP
start_scapy_server doesn't necessarily start a server (only the
first lucky subshell will do), wait for server socket file before
proceeding with ovs-appctl call.
Signed-off-by: Ihar Hrachyshka
Acked-by: Mark Michelson
---
tests/ovn-macros.at | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
execution after [1], [2] as well as this series are all merged.)
---
v1: initial
v2: use perf_counter in scapy-server time logging
log payload request before any processing
Ihar Hrachyshka (4):
fmt_pkt: don't subshell when calling ovs-appctl
fmt_pkt: use -S check to wait for scapy sock
Signed-off-by: Ihar Hrachyshka
Acked-by: Mark Michelson
---
tests/ovn-macros.at | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/ovn-macros.at b/tests/ovn-macros.at
index 8dc4ec75c..3191bb6ad 100644
--- a/tests/ovn-macros.at
+++ b/tests/ovn-macros.at
@@ -876,7 +876,7
The daemon will now log to scapy.log file.
Log messages include stats on request processing time as well as any
errors that may happen during processing.
If you'd like to see even more logs (e.g. for debugging purposes), just
pass --verbose to scapy-server.
Signed-off-by: Ihar Hrach
While either check works, it's better to use a more explicit check.
Signed-off-by: Ihar Hrachyshka
Acked-by: Mark Michelson
---
tests/ovn-macros.at | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/ovn-macros.at b/tests/ovn-macros.at
index 3191bb6ad..0b15bcc80 1
On Mon, Nov 20, 2023 at 11:22 AM Simon Horman wrote:
> On Sat, Nov 18, 2023 at 01:07:02AM +0000, Ihar Hrachyshka wrote:
> > In nonblocking mode, a POSIX socket may temporarily fail to connect(),
> > which is indicated by returning EINPROGRESS. This patch will loop
> > wai
The daemon will now log to scapy.log file.
Log messages include stats on request processing time as well as any
errors that may happen during processing.
If you'd like to see even more logs (e.g. for debugging purposes), just
pass --verbose to scapy-server.
Signed-off-by: Ihar Hrach
While either check works, it's better to use a more explicit check.
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-macros.at | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/ovn-macros.at b/tests/ovn-macros.at
index 3191bb6ad..0b15bcc80 100644
--- a/tests/ovn-macr
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-macros.at | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/ovn-macros.at b/tests/ovn-macros.at
index 8dc4ec75c..3191bb6ad 100644
--- a/tests/ovn-macros.at
+++ b/tests/ovn-macros.at
@@ -876,7 +876,7 @@ fmt_pkt() {
if
start_scapy_server doesn't necessarily start a server (only the
first lucky subshell will do), wait for server socket file before
proceeding with ovs-appctl call.
Signed-off-by: Ihar Hrachyshka
---
tests/ovn-macros.at | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/test
execution after [1], [2] as well as this series are all merged.)
[1] https://patchwork.ozlabs.org/project/openvswitch/list/?series=382739
[2] https://patchwork.ozlabs.org/project/ovn/list/?series=382220
Ihar Hrachyshka (4):
fmt_pkt: don't subshell when calling ovs-appctl
fmt_pkt: use -S che
While POSIX document EINPROGRESS, Linux connect() system call returns
EAGAIN for nonblocking Unix sockets instead. This patch handles this
case by repeating connect() attempts.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util-unix.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
stream_class to support connect API, that would
allow the caller to repeat connect() attempts asynchronously.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util-unix.c | 14 ++
1 file changed, 10 insertions(+), 4 deletions(-)
diff --git a/lib/socket-util-unix.c b/lib/socket-util-unix.c
Unix sockets also have listener backlog. Returning EAGAIN is valid to
allow the caller handle the non-blocking temporary failure to connect()
gracefully.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util-unix.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/lib/socket-util-unix.c b/lib
effectively bumps the
number of simultaneous connections to python unixctl servers to 64. Also
for INET C servers too.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util.c| 2 +-
python/ovs/stream.py | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/socket-util.c b
() are handled by looping until
success.
- conversion from EAGAIN to EPROTO is removed.
---
v1: initial version
v2: fixed commit message for one of patches
Ihar Hrachyshka (4):
Use listen backlog = 64 for all connections
Don't convert EAGAIN to EPROTO for unix sockets
Loop to connect
While POSIX document EINPROGRESS, Linux connect() system call returns
EAGAIN for nonblocking Unix sockets instead. This patch handles this
case by repeating connect() attempts.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util-unix.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
stream_class to support connect API, that would
allow the caller to repeat connect() attempts asynchronously.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util-unix.c | 14 ++
1 file changed, 10 insertions(+), 4 deletions(-)
diff --git a/lib/socket-util-unix.c b/lib/socket-util-unix.c
Unix sockets also have listener backlog. Returning EAGAIN is valid to
allow the caller handle the non-blocking temporary failure to connect()
gracefully.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util-unix.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/lib/socket-util-unix.c b/lib
effectively bumps the
number of simultaneous connections to python unixctl servers to 64.
Signed-off-by: Ihar Hrachyshka
---
lib/socket-util.c| 2 +-
python/ovs/stream.py | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/socket-util.c b/lib/socket-util.c
index
from EAGAIN to EPROTO is removed
Ihar Hrachyshka (4):
Use listen backlog = 64 for all connections
Don't convert EAGAIN to EPROTO for unix sockets
Loop to connect unix socket until success
Handle EAGAIN on unix socket connect() failure
lib/socket-util-unix.c | 17 ++---
I confirm the changes in v2 match the cover letter description.
For the whole series, +1. I will follow up with a patch on background
execution of fmt_pkt.
Reviewed-By: Ihar Hrachyshka
On Tue, Nov 14, 2023 at 3:54 PM Mark Michelson wrote:
> Note: v1 and v2 differences are listed at
rming they work as advertised.
Signed-off-by: Ihar Hrachyshka
---
v1: - initial version.
v2: - convert from appctl to ovstest.
- add --bad-csum support.
- drop separate test case and man for the tool.
v3: - fix build warnings on restricted ovs_be16 ++.
v4: - migrate to compose-packe
On Wed, Nov 1, 2023 at 5:42 PM Ilya Maximets wrote:
> On 10/24/23 19:04, Ihar Hrachyshka wrote:
> > With --hexified, it will produce a bare hexified payload with no spaces
> > or offset indicators inserted, which is useful in tests to produce
> > frames to pass to e.g. `
+1, thanks for the cleanup.
Reviewed-By: Ihar Hrachyshka
On Thu, Nov 2, 2023 at 8:50 AM Ales Musil wrote:
> The empty initializer (" = {}") is allowed only from
> C23 and further [0]. Use the supported C99 compatible
> method instead ("= {0}"). This shouldn'
+1 for the series.
Reviewed-By: Ihar Hrachyshka
On Mon, Oct 30, 2023 at 4:10 PM Ilya Maximets wrote:
> The main goal of this patch set is to enable builds without warnings
> on systems with Python 3.12. Python 3.12 turned having invalid escape
> sequences into a SyntaxWarning. In t
This is the standard clang JSON Compilation Database file name, as
documented in:
https://clang.llvm.org/docs/JSONCompilationDatabase.html#build-system-integration
FYI the file is used by Language Server Protocol implementations and
other AST tools.
Signed-off-by: Ihar Hrachyshka
This is the standard clang JSON Compilation Database file name, as
documented in:
https://clang.llvm.org/docs/JSONCompilationDatabase.html#build-system-integration
FYI the file is used by Language Server Protocol implementations and
other AST tools.
Signed-off-by: Ihar Hrachyshka
E721: "do not compare types, for exact checks use `is` / `is not`, for
instance checks use `isinstance()`"
This fixes `make flake8-check` target when running with
pycodestyle>=1.2.
Signed-off-by: Ihar Hrachyshka
---
python/ovs/jsonrpc.py | 2 +-
tests/test-jsonrpc.py | 4 ++-
rming they work as advertised.
Signed-off-by: Ihar Hrachyshka
---
v1: - initial version.
v2: - convert from appctl to ovstest.
- add --bad-csum support.
- drop separate test case and man for the tool.
v3: - fix build warnings on restricted ovs_be16 ++.
v4: - migrate to compose-packet.
---
On Fri, Oct 20, 2023 at 6:53 PM Ilya Maximets wrote:
> On 10/21/23 00:40, Ihar Hrachyshka wrote:
> > On Fri, Oct 20, 2023 at 6:02 PM Ilya Maximets <mailto:i.maxim...@ovn.org>> wrote:
> >
> > On 10/20/23 15:15, Ihar Hrachyshka wrote:
> > > On Thu
On Fri, Oct 20, 2023 at 5:42 PM Ilya Maximets wrote:
> On 10/20/23 21:43, Ihar Hrachyshka wrote:
> > On Fri, Oct 20, 2023 at 12:26 PM Mark Michelson
> wrote:
> >
> >> The "IPv6 ND Router Solicitation responder" test was altered to test
> >> mor
On Fri, Oct 20, 2023 at 6:02 PM Ilya Maximets wrote:
> On 10/20/23 15:15, Ihar Hrachyshka wrote:
> > On Thu, Oct 19, 2023 at 6:29 PM Ilya Maximets <mailto:i.maxim...@ovn.org>> wrote:
> >
> > On 10/19/23 16:11, Ihar Hrachyshka wrote:
> > > The comm
Hi.
To confirm, are you trying to implement a (permutation of?)
https://www.rfc-editor.org/rfc/rfc3046.html here? If so, the feature
should probably be renamed into DHCP Relay Agent support. Please clarify
if that's your intent here.
Thanks,
Ihar
On 8/22/23 10:36 PM, naveen.yerramneni wrot
On 10/6/23 3:02 AM, Ales Musil wrote:
The unixctl exit command would receive reply immediately
which is confusing and can cause some issues in some tests
if the cleanup takes longer than expected. To avoid that
make sure we reply to the exit command only after the
main cleanup was done so there s
1 - 100 of 753 matches
Mail list logo