An unnecessarily overflow would occurs when the 'value' is longer than
4294967295. So it's required to check ranges to avoid uint32_t overflow.
Reported-by: Nan Zhou
Signed-off-by: Yunjian Wang
---
v2: fix patch code styles
---
lib/ofp-parse.c | 7 ++-
1 file changed, 2 insertions(+), 5 del
> -Original Message-
> From: Simon Horman [mailto:simon.hor...@corigine.com]
> Sent: Thursday, April 20, 2023 9:38 PM
> To: wangyunjian
> Cc: d...@openvswitch.org; i.maxim...@ovn.org; luyicai
> Subject: Re: [ovs-dev] [PATCH] ofp-parse: Check ranges on string to uint32_t
> conversion.
>
>
The test was often failing with a meter of 10 pps while generating 100 packets.
When the 100 packets generation takes more than 100 msec, the test fails.
This happens from time to time in upstream ci.
Reducing the number of generated packets to 40 should reduce number of flakes.
Signed-off-by: Xav
CoPP test failed for this patch, but this is in a flaky test - the failure
is independent of this patch.
A patch for the CoPP test will be proposed later, independently of this
patch
On Thu, Apr 20, 2023 at 8:04 AM Ales Musil wrote:
>
>
> On Wed, Apr 19, 2023 at 2:44 PM Xavier Simonart
> wrote:
The tests were randomly failing as there is no guarentee that
15 nc will reach different load balancers.
Also, running 15 nc made the test quite slow.
Fixed by running loops of 5 nc, and then checking results.
Test duration decreased by 15 to 20 seconds.
Signed-off-by: Xavier Simonart
---
tests/
This test is broken since a long time but passed as it used
OVS_WAIT_UNTIL for checking output, which succeeds whatever the output is ...
- Replaced OVS_WAIT_UNTIL by OVS_WAIT_FOR_OUTPUT
- Update table numbers
- Added back hv$i-vif1 port
- Changed slaves to members
- Added back gw2 (hence could rem
Signed-off-by: Xavier Simonart
---
tests/ovn-controller.at | 19 ---
tests/ovn.at| 24
2 files changed, 24 insertions(+), 19 deletions(-)
diff --git a/tests/ovn-controller.at b/tests/ovn-controller.at
index 597aed00b..137724723 100644
--- a/te
The macro has been used erroneously with the expectation
to take the third parameter as the expected output.
Fail if more than 2 arguments are used.
Also fail if the second argument is an integer (second arg expected to
be a command to execute in case of failure).
Same checks are used in ovs.
Sign
In vxlan mode with more than 2047 lrp in router, build_ports()
prints "all port tunnel ids exhausted", and frees the lrp port.
However, lsp of type "router" connected to the lrp still holds
the pointer in port->peer. This leads to northd crash in
build_lflows().
CallTrace:
build_lswitch_rport_a
On Thu, Apr 20, 2023 at 05:31:23PM +0200, Stefan Hoffmann wrote:
> On Thu, 2023-04-20 at 15:50 +0200, Simon Horman wrote:
> > On Thu, Apr 20, 2023 at 02:18:33PM +0200, David Marchand wrote:
> > > On Thu, Apr 20, 2023 at 2:06 PM Simon Horman
> > > wrote:
> > > > > I do see CI failures, but I think
On Thu, 2023-04-20 at 15:50 +0200, Simon Horman wrote:
> On Thu, Apr 20, 2023 at 02:18:33PM +0200, David Marchand wrote:
> > On Thu, Apr 20, 2023 at 2:06 PM Simon Horman
> > wrote:
> > > > I do see CI failures, but I think these are false negatives:
> > > >
> > > > *
> > > > https://patchwork.o
I don't see how this is any better. Instead of calling to python that
imports GPL libraries, we now call to scapy that calls to python that
imports GPL libraries. In both cases we communicate through shell
pipes and the only "copyright"-able piece here is scapy API. (There
are arguments about wheth
On Thu, Apr 20, 2023 at 03:47:43PM +0200, Simon Horman wrote:
> On Wed, Apr 19, 2023 at 03:49:58PM +0800, Yunjian Wang via dev wrote:
>
> Hi Yunjian,
>
> Please add some text here describing why passing dp and buf is a proble:
> how it leads to a leak.
>
> Also, FYI, I notice that the CI run fai
On Thu, Apr 13, 2023 at 10:23:31AM +0200, David Marchand wrote:
> Let's separate DPDK compilation from the rest of OVS build:
> - this avoids multiple jobs building DPDK in parallel, which especially
> affects builds in the dpdk-latest branch,
> - we separate concerns about DPDK build requirement
On Fri, Apr 14, 2023 at 04:44:53PM +0100, Kevin Traynor wrote:
> By default OVS configures 2048 descriptors for tx and rx queues
> on DPDK devices. It also allows the user to configure those values.
>
> If the values used are not acceptable to the device then queue setup
> would fail.
>
> The dev
On Thu, Apr 20, 2023 at 03:13:20PM +0200, Dumitru Ceara wrote:
> Hi Ihar, Simon,
>
> This is not very pretty but would it improve the situation (not even
> importing scapy anymore)?
IANAL, but I would be more comfortable with this approach.
>
> fmt_pkt() {
> echo "import binascii; \
>
On Thu, Apr 20, 2023 at 02:18:33PM +0200, David Marchand wrote:
> On Thu, Apr 20, 2023 at 2:06 PM Simon Horman
> wrote:
> > > I do see CI failures, but I think these are false negatives:
> > >
> > > *
> > > https://patchwork.ozlabs.org/project/openvswitch/patch/3f70ca7bafad296e18ed9579f30fd7044c
On Wed, Apr 19, 2023 at 03:49:58PM +0800, Yunjian Wang via dev wrote:
Hi Yunjian,
Please add some text here describing why passing dp and buf is a proble:
how it leads to a leak.
Also, FYI, I notice that the CI run failed, which seems completely
unrelated to your patch:
* https://github.com/ovs
The normalize_prefix_str returns NULL when it fails to
normalize the specified string. Prevent UB with formatting
NULL with %s and simplify the free pattern.
Signed-off-by: Ales Musil
---
utilities/ovn-nbctl.c | 12 +---
1 file changed, 5 insertions(+), 7 deletions(-)
diff --git a/utili
On Fri, Apr 14, 2023 at 05:47:59PM +0800, Yunjian Wang via dev wrote:
> An unnecessarily overflow would occurs when the 'value' is longer than
> 4294967295. So it's required to check ranges to avoid uint32_t overflow.
Thanks, I see this is correct.
Perhaps it would be worth noting that it occurs
As a note:
The newly added system test sometimes fails as it is finding some warnings
in the controller.log (e.g.
2023-04-19T13:18:29.536Z|00087|if_status|WARN|Trying to release unknown
interface lsp1)
This can happen in the following scenario:
- logical port is created
- OVS interfaces is added w
Hi Ihar, Simon,
This is not very pretty but would it improve the situation (not even
importing scapy anymore)?
fmt_pkt() {
echo "import binascii; \
out = binascii.hexlify(raw($1)); \
out.decode()" | scapy -H 2> /dev/null | awk '{print $4}' |
cut -f 2 -d "'"
}
Re
IANAL but
I don't think it's more of a problem than e.g. autoconf being GPL.
(It's not.) We don't link with scapy, and it's only used in test
suite. (Also, scapy was already used in the test suite before the
patch.)
On Thu, Apr 20, 2023 at 8:36 AM Simon Horman wrote:
>
> On Wed, Apr 12, 2023 at
On Wed, Apr 12, 2023 at 01:29:48AM +, Ihar Hrachyshka wrote:
> Scapy allows to define packets in descriptive form that is easier to
> digest and debug.
>
> Signed-off-by: Ihar Hrachyshka
> ---
> tests/ovn-macros.at | 23 +++
> 1 file changed, 23 insertions(+)
>
> diff --
On Thu, Apr 20, 2023 at 2:06 PM Simon Horman wrote:
> > I do see CI failures, but I think these are false negatives:
> >
> > *
> > https://patchwork.ozlabs.org/project/openvswitch/patch/3f70ca7bafad296e18ed9579f30fd7044c47fc61.ca...@cloudandheat.com/
> >
> > I'm retrying the GitHub based jobs her
On Thu, Apr 20, 2023 at 01:35:59PM +0200, Simon Horman wrote:
> On Thu, Apr 20, 2023 at 10:14:14AM +0200, Stefan Hoffmann wrote:
> > In some cases ovsdb server or relay gets restarted, ovsdb python clients
> > may keep the local socket open. Instead of reconnecting a lot of failures
> > will be log
On Thu, Apr 20, 2023 at 11:23 AM wangchuanlei
wrote:
> Signed-off-by: wangchuanlei
> ---
> controller/pinctrl.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/controller/pinctrl.c b/controller/pinctrl.c
> index 795847729..97a5e392f 100644
> --- a/controller/pinctrl.
On Thu, Apr 20, 2023 at 10:14:14AM +0200, Stefan Hoffmann wrote:
> In some cases ovsdb server or relay gets restarted, ovsdb python clients
> may keep the local socket open. Instead of reconnecting a lot of failures
> will be logged.
> This can be reproduced with ssl connections to the server/relay
Aaron Conole writes:
> Paolo Valerio writes:
>
>> Connections that need to be removed, e.g. while forcing a direction,
>> were invalidated forcing them to be expired.
>> This is not actually needed, as it's typically a one-time
>> operation.
>> The patch replaces a call to conn_force_expire() wi
Signed-off-by: wangchuanlei
---
controller/pinctrl.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/controller/pinctrl.c b/controller/pinctrl.c
index 795847729..97a5e392f 100644
--- a/controller/pinctrl.c
+++ b/controller/pinctrl.c
@@ -7190,7 +7190,9 @@ bfd_monitor_send_ms
On Sun, Apr 09, 2023 at 07:15:09PM +0800, mit...@outlook.com wrote:
> From: Lin Huang
>
> OvS has supported packet-per-second policer which can be set at ingress and
> egress
> side in kernel datapath. But the userspace datapath dosen't support for
> ingress and
> egress packet-per-second polic
Hi Lin Huang,
thanks for your patch.
On Sun, Apr 09, 2023 at 07:15:08PM +0800, mit...@outlook.com wrote:
> From: Lin Huang
>
> OvS has supported packet-per-second policer which can be set at ingress and
> egress
> side in kernel datapath. But the userspace datapath dosen't support for
> ingre
In some cases ovsdb server or relay gets restarted, ovsdb python clients
may keep the local socket open. Instead of reconnecting a lot of failures
will be logged.
This can be reproduced with ssl connections to the server/relay and
restarting it, so it has the same IP after restart.
This patch catc
On Mon, Apr 17, 2023 at 10:16???AM wangchuanlei wrote:
when create bfd entry, set min_tx to 1, (tx_timeout * 25)/100 is
zero, which cause controller to restart!
Signed-off-by: wangchuanlei
Hi,
---
controller/pinctrl.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/controller/pinctrl.c b/
On Tue, Apr 26, 2022 at 07:15:59AM -0400, 0-day Robot wrote:
> Bleep bloop. Greetings lin huang, 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.
>
>
> git-am:
> error: corrupt patch at
When doing performance testing with OVS v3.1 we ran into a deadlock
situation with the netdev_hmap_rwlock read/write lock. After some
debugging, it was discovered that the netdev_hmap_rwlock read lock
was taken recursively. And well in the folowing sequence of events:
netdev_ports_flow_get()
I
36 matches
Mail list logo
