Re: [ovs-dev] Restore flows is slow using "ovs-save", is there someone who use RPC to speed up flow restore?

[Timothy M. Redaelli]

On 11/24/2017 04:07 AM, Sam wrote:
> Hi all,
> 
> I'm working on speed up ovs restart, I found that restore flows is slow
> when there are lots of flows, so I want to use RPC to store flows in
> another process, and then restore from it.
> 
> Is there someone who have worked on this? could you share how faster after
> this change? Thank you~
> 
> Or is there some other ways to speed up? Thank you~!

Hi,
I sent a patch [1], already merged on master branch, that makes ovs-save
much faster by using --bundle (if you are supporting at least OF 1.4)
and file (instead of heredoc).

With that patchset you'll be able to restore 1 million flows with
traffic disruption less than 15 seconds.

[1]
https://patchwork.ozlabs.org/project/openvswitch/list/?series=4973=%2A=both
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] rhel: Add support for "systemctl reload openvswitch"

[Timothy M. Redaelli]

On 11/03/2017 07:20 PM, Ben Pfaff wrote:
[...]
>>
>> This looks good to me, my only comment is that the script is in rhel/
>> and uses systemd so maybe we should reflect that in the script's name?
>> Because OVS itself might provide an ovs-reload too and then we will
>> have issues.
>>
>> What do you think?
> 
> Is this script reasonably generic, as opposed to RHEL-specific?  If so
> then we could move it into utilities or make it a command under ovs-ctl.

Unlucky this is specific for RHEL (and derivates) since Debian/Ubuntu
uses only one service file for both ovsdb-server and ovs-vswitchd and so
this script can't work, but I think they can use "ovs-ctl restart" that
have the same behavior when you don't have multiple systemd service files.
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] rhel: fix log directory permissions

[Timothy M. Redaelli]

On 09/22/2017 03:44 PM, Aaron Conole wrote:
> When the logrotate script runs, and Open vSwitch is running as a non-root
> user, the /var/log/openvswitch directory doesn't have other rx bits set.
> This means the reopen attempt will fail with "permission denied", even though
> the default logrotate configuration creates a new log file with the
> appropriate attributes.
> 
> This change sets the r/x bits for other on /var/log/messages
> 
> Signed-off-by: Aaron Conole 
> Tested-by: Jean Hsiao 

Acked-by: Timothy Redaelli 
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] ovs-save: Use --bundle to restore flows (on OpenFlow 1.4+)

[Timothy M. Redaelli]

On 09/23/2017 01:09 AM, Flavio Leitner wrote:
[...]
> 
> Where is $workdir set?  Looks like it's empty and will use root fs.
> 
> Although this is a small patch, it actually does two things. One is
> fixing to use highest enabled OF and another is replacing echo with
> printf.
> 
> I suggest to split them apart so it's easier to review.

All done, thank you.

https://patchwork.ozlabs.org/project/openvswitch/list/?series=4973
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH v3 2/2] rhel: delete transient ports on boot when starting ovsdb-server

[Timothy M. Redaelli]

On 08/11/2017 07:12 PM, Aaron Conole wrote:
> Timothy Redaelli  writes:
> 
>> Use ovs-ctl --delete-transient-ports-on-boot to start ovsdb-server.
>>
>> This feature can be disabled by appending --no-delete-transient-ports-on-boot
>> to OPTIONS in /etc/sysconfig/openvswitch
>>
>> Signed-off-by: Timothy Redaelli 
>> ---
>>  rhel/usr_lib_systemd_system_ovsdb-server.service | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/rhel/usr_lib_systemd_system_ovsdb-server.service 
>> b/rhel/usr_lib_systemd_system_ovsdb-server.service
>> index 7acd25f78..42473161e 100644
>> --- a/rhel/usr_lib_systemd_system_ovsdb-server.service
>> +++ b/rhel/usr_lib_systemd_system_ovsdb-server.service
>> @@ -10,14 +10,14 @@ Type=forking
>>  Restart=on-failure
>>  EnvironmentFile=/etc/openvswitch/default.conf
>>  EnvironmentFile=-/etc/sysconfig/openvswitch
>> +ExecStartPre=/usr/bin/mkdir -p /var/run/openvswitch
>>  ExecStartPre=/usr/bin/chown ${OVS_USER_ID} /var/run/openvswitch
>>  ExecStart=/usr/share/openvswitch/scripts/ovs-ctl \
>>--no-ovs-vswitchd --no-monitor --system-id=random \
>> +  --delete-transient-ports-on-boot \
>>--ovs-user=${OVS_USER_ID} \
>>start $OPTIONS
>>  ExecStop=/usr/share/openvswitch/scripts/ovs-ctl --no-ovs-vswitchd stop
>>  ExecReload=/usr/share/openvswitch/scripts/ovs-ctl --no-ovs-vswitchd \
>> --ovs-user=${OVS_USER_ID} \
>> --no-monitor restart $OPTIONS
>> -RuntimeDirectory=openvswitch
>> -RuntimeDirectoryMode=0755
> 
> I'm not sure about this change.  One thing that's nice is the way
> systemd will cleanup the runtime directories when this is done.  I think
> this can leave it around.
> 
> I realize that under rhel (and fedora), the /run mountpoint is tmpfs.
> 
> Is there another way we could accomodate this?  Does a db flag make
> sense?

I don't think we can use db since we need that flags to be reset on each
reboot (that's the reason I use /run).

In the previous patchset this run directory was created by using
tmpfiles.d, but unlucky tmpfiles.d cannot use environment files, so you
can't use ${OVS_USER_ID}.

If cleanup is needed, but imho it should be done by ovs-ctl or
ovs-vswitchd/ovsdb-server directly, we may do the cleanup on ExecStopPost.

___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] netdev-dpdk: include dpdk PCI header directly

[Timothy M. Redaelli]

On 08/09/2017 10:00 PM, Aaron Conole wrote:
> As part of a devargs rework in DPDK, the PCI header file was removed, and
> needs to be directly included.  This isn't required to build with 17.05 or
> earlier, but will be required should a future update happen.
> 
> Signed-off-by: Aaron Conole 

Tried to build with dpdk 17.08

Tested-By: Timothy Redaelli 
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] rhel: Add support for "systemctl reload openvswitch"

[Timothy M. Redaelli]

On 08/08/2017 08:24 PM, Guru Shetty wrote:
> 
> 
> On 8 August 2017 at 11:05, Russell Bryant  > wrote:
> 
> On Tue, Aug 8, 2017 at 1:32 PM, Guru Shetty  > wrote:
> > On 8 August 2017 at 07:42, Timothy Redaelli  > wrote:
> >
> >> The reload procedure will trigger a script that saves the flows and tlv
> >> maps then it restarts ovsdb-server, it stops ovs-vswitchd, it sets
> >> other_config:flow-restore-wait=true (to wait till flow restore is
> >> finished), it starts ovs-vswitchd, it restore the backupped flows/tlv
> >> maps and it removes other_config:flow-restore-wait=true (logic mostly
> >> ripped
> >> from ovs-ctl).
> >>
> >> It uses systemctl with --job-mode=ignore-dependencies to restart
> >> ovsdb-server
> >> and stop and start ovs-vswitchd in order to avoid systemd to restart 
> the
> >> other components due to dependencies (as explained in
> >> rhel/README.RHEL.rst).
> >>
> >> It also uses --bundle, when available, in order to minimize the 
> downtime.
> >>
> >
> > The core script can probably be added to ovs-ctl, reuse some functions 
> and
> > then you can call it? This way, other platforms can benefit too.
> 
> Sounds like a good suggestion.

The script is actually "specialized" to how Fedora/RHEL starts
openvswitch since, I think, only Fedora/RHEL have ovsdb-server and
ovs-vswitchd as splitted systemd unit files.

Ubuntu uses only one .service file (openvswitch-nonetwork.service) that
starts both ovsdb-server and ovs-vswitch at the same time, so this
approach can't work for them.

(It seems) Debian only has /etc/init.d/openvswitch-switch and so this
doesn't apply either.

> What would you think about updating the stop/start behavior to do
> this, as well?  It looks like this makes "reload" actually just an
> "intelligent restart".  systemctl has a "restart" command too, which
> looks like it just does stop/start, and it'd be nice to make that work
> better too.

You can't change the behavior of "systemctl restart", systemd will
always call ExecStop and ExecStart, that's the reason I add the new
script as "reload" target, but that makes sense since it's not a
"standard" restart (it keeps the flows). Moreover I'd like to keep user
to have the choice if reloading or restarting the daemon.

> We had this discussion when we added intelligent restart
> to rhel/etc_init.d_openvswitch and debian/openvswitch-switch.init
> 
> Currently both rhel and debian will take a "--save-flows=yes" option to
> restart and it will do the same things as the script proposed here does
> (except for --bundle) via ovs-ctl's restart function . We wanted
> "--save-flows=yes" to be the default behavior instead of explicitly
> calling it. But a long term user of OVS said that they use "restart" for
> a fresh start. That is, they bake in the assumption that a "restart"
> means the added flows disappear.

My implementation doesn't change this assumption, the only assumption is
that if you use "reload" you want to keep all the flows (that's the same
assumption you have when you use "systemctl reload" on nginx or apache)

> What this patch does in addition to the "restart" function in ovs-ctl is
> that it uses --bundle. "restart" function in ovs-ctl calls functions in
> "ovs-save". So it may make sense for this patch to add the "--bundle"
> feature to ovs-save and just re-use it?

Sounds like a good suggestion.

I'll do another patchset with a commit that adds the --bundle option +
get_highest_ofp_version in ovs-save and another commit that changes
ovs-reload script to use it.

get_highest_ofp_version is used to find if we can use --bundle and to
make ovs-ofctl work when you have a bridge with only OpenFlow15+
protocols enabled.

Any other suggestions?
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH v4] Update relevant artifacts to add support for DPDK 17.05.1.

[Timothy M. Redaelli]

On 07/25/2017 02:34 PM, Michal Weglicki wrote:

Upgrading to DPDK 17.05.1 stable release adds new
significant features relevant to OVS, including,
but not limited to:
- tun/tap PMD,
- VFIO hotplug support,
- Generic flow API.

Following changes are applied:
- netdev-dpdk: Changes required by DPDK API modifications.
- doc: Because of DPDK API changes, backward compatibility
   with previous DPDK releases will be broken, thus all
   relevant documentation entries are updated.
- .travis: DPDK version change from 16.11.1 to 17.05.1.
- rhel/openvswitch-fedora.spec.in: DPDK version change
   from 16.11 to 17.05.1

v1->v2: Patch rebase.
v2->v3: Fixed wrong formating after v2 patch rebase.
v3->v4: Minor documentation changes.

Signed-off-by: Michal Weglicki 
Reviewed-by: Aaron Conole 
---
  .travis/linux-build.sh   |   2 +-
  Documentation/faq/releases.rst   |   1 +
  Documentation/howto/dpdk.rst |   6 +-
  Documentation/intro/install/dpdk.rst |  14 +--
  Documentation/topics/dpdk/vhost-user.rst |  12 +--
  NEWS |   1 +
  lib/netdev-dpdk.c| 144 +++
  rhel/openvswitch-fedora.spec.in  |   2 +-
  tests/dpdk/ring_client.c |   6 +-
  9 files changed, 114 insertions(+), 74 deletions(-)


[...]

diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in
index 3a045d3..2bb7102 100644
--- a/rhel/openvswitch-fedora.spec.in
+++ b/rhel/openvswitch-fedora.spec.in
@@ -84,7 +84,7 @@ BuildRequires: libcap-ng libcap-ng-devel
  %endif
  %if %{with dpdk}
  BuildRequires: libpcap-devel numactl-devel
-BuildRequires: dpdk-devel >= 16.11
+BuildRequires: dpdk-devel >= 17.05.1
  Provides: %{name}-dpdk = %{version}-%{release}
  %endif
 



Hi,
on Fedora there is no dpdk-devel 17.05.1 package, the last rawhide 
version is dpdk-devel 17.05 (and usually Fedora doesn't package minor 
releases of dpdk).


I tried to build openvswitch with the dpdk 17.05 package from rawhide.
It build and it works (tested with dpdkvhostuserclient) so, unless you 
need a specific fix from 17.05.1, I suggest to change the BuildRequires 
to 17.05 instead of 17.05.1 or the package cannot be built on Fedora, 
unless you are using an additional repository (copr).

___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH v2] netdev: check for NULL fields in netdev_get_addrs

[Timothy M. Redaelli]

On 07/18/2017 05:25 PM, Daniel Alvarez Sanchez wrote:
> When the interfaces list is retrieved through getiffaddrs(), there
> might be elements with iface_name set to NULL.
> 
> This patch checks ifa_name to be not NULL before comparing it to the
> actual device name in the loop that calculates how many interfaces
> exist with that same name.
> 
> Also, this patch checks that ifa_netmask is not NULL for coherence
> with the existing code so that it doesn't allocate more memory
> than needed if this field is NULL.
> 
> Note, that these checks are already being done later in the function
> so it should be done in both places.
> 
> Signed-off-by: Daniel Alvarez 
> ---

Acked-by: Timothy Redaelli 
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] tests: Disable no-format-truncation warnings

[Timothy M. Redaelli]

On 07/13/2017 07:21 PM, Ben Pfaff wrote:
> On Thu, Jul 13, 2017 at 04:29:33PM +0200, Timothy Redaelli wrote:
>> test_snprintf function (tests/test-util.c) tests snprintf with shorter 
>> length,
>> but this emit a warning on GCC 7.0 or later.
>>
>> This commit disables that warning on tests only.
>>
>> Signed-off-by: Timothy Redaelli 
> 
> How about disabling it just for those lines of code?

Good idea, it's surely better to limit the portion of disabled warning
in order to maximize the effectiveness of the warning itself.

LGTM
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] acinclude: netdev-dpdk: use pkg-config of libdpdk

[Timothy M. Redaelli]

On 07/05/2017 11:19 AM, Christian Ehrhardt wrote:
> If available use dpdk pkg-config info of libdpdk to set the right
> include paths.
> That for example, allows packagers to provide non default include
> paths in a common way (pkg-config).

I like the idea, but I think it's better to wait for upstream (DPDK) to
add pkg-config support before pushing something that actually can only
be used in debian-based distributions (libdpdk-dev package).
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH v2 4/4] redhat: allow dpdk to also run as non-root user

[Timothy M. Redaelli]

Hi Aaron,

On 07/05/2017 07:56 PM, Aaron Conole wrote:
> After this commit, users may start a dpdk-enabled ovs setup as a
> non-root user.  This is accomplished by exporting the $HOME directory,
> which dpdk uses to fill in it's semi-persistent RTE configuration.
> 
> This change may be a bit controversial since it modifies /dev/hugepages
> as part of starting the ovs-vswitchd to set a hugetlbfs group
> ownership.  This is used to enable writing to /dev/hugepages so that the
> dpdk_init will successfully complete.  There is an alternate way of
> accomplishing this - namely to initialize DPDK before dropping
> privileges.  However, this would mean that if DPDK ever grows an uninit
> / reinit function, non-root ovs likely could never use it.
> 
> This does not change OvS+DPDK's SELinux requirements.  It still must be
> disabled.
> 
> Signed-off-by: Aaron Conole 
> ---
>  rhel/.gitignore |  1 +
>  rhel/automake.mk|  3 ++-
>  rhel/openvswitch-fedora.spec.in | 13 
> +
>  ...rvice => usr_lib_systemd_system_ovs-vswitchd.service.in} |  4 
>  4 files changed, 20 insertions(+), 1 deletion(-)
>  rename rhel/{usr_lib_systemd_system_ovs-vswitchd.service => 
> usr_lib_systemd_system_ovs-vswitchd.service.in} (87%)
> 
> diff --git a/rhel/.gitignore b/rhel/.gitignore
> index 164bb66..e584a1e 100644
> --- a/rhel/.gitignore
> +++ b/rhel/.gitignore
> @@ -4,3 +4,4 @@ openvswitch-kmod-rhel6.spec
>  openvswitch-kmod-fedora.spec
>  openvswitch.spec
>  openvswitch-fedora.spec
> +usr_lib_systemd_system_ovs-vswitchd.service
> diff --git a/rhel/automake.mk b/rhel/automake.mk
> index 2d9443f..439af26 100644
> --- a/rhel/automake.mk
> +++ b/rhel/automake.mk
> @@ -29,6 +29,7 @@ EXTRA_DIST += \
>   rhel/usr_lib_systemd_system_openvswitch.service \
>   rhel/usr_lib_systemd_system_ovsdb-server.service \
>   rhel/usr_lib_systemd_system_ovs-vswitchd.service \
> + rhel/usr_lib_systemd_system_ovs-vswitchd.service.in \
>   rhel/usr_lib_systemd_system_ovn-controller.service \
>   rhel/usr_lib_systemd_system_ovn-controller-vtep.service \
>   rhel/usr_lib_systemd_system_ovn-northd.service \
> @@ -59,7 +60,7 @@ RPMBUILD_TOP := $(abs_top_builddir)/rpm/rpmbuild
>  RPMBUILD_OPT ?= --without check
>  
>  # Build user-space RPMs
> -rpm-fedora: dist $(srcdir)/rhel/openvswitch-fedora.spec
> +rpm-fedora: dist $(srcdir)/rhel/openvswitch-fedora.spec 
> $(srcdir)/rhel/usr_lib_systemd_system_ovs-vswitchd.service
>   ${MKDIR_P} ${RPMBUILD_TOP}/SOURCES
>   cp ${DIST_ARCHIVES} ${RPMBUILD_TOP}/SOURCES
>   rpmbuild ${RPMBUILD_OPT} \
> diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in
> index 7c805b2..83cfb18 100644
> --- a/rhel/openvswitch-fedora.spec.in
> +++ b/rhel/openvswitch-fedora.spec.in
> @@ -95,6 +95,10 @@ Requires: openssl hostname iproute module-init-tools
>  Requires(post): /usr/bin/getent
>  Requires(post): /usr/sbin/useradd
>  Requires(post): /usr/bin/sed
> +%if %{with dpdk}
> +Requires(post): /usr/sbin/usermod
> +Requires(post): /usr/sbin/groupadd
> +%endif
>  Requires(post): systemd-units
>  Requires(preun): systemd-units
>  Requires(postun): systemd-units
> @@ -366,6 +370,15 @@ if [ $1 -eq 1 ]; then
>  
>  sed -i 's:^#OVS_USER_ID=:OVS_USER_ID=:' /etc/sysconfig/openvswitch
>  
> +%if %{with dpdk}
> +getent group hugetlbfs >/dev/null || \
> +groupadd hugetlbfs
> +usermod -a -G hugetlbfs openvswitch
> +sed -i \
> +
> 's@OVS_USER_ID="openvswitch:openvswitch"@OVS_USER_ID="openvswitch:hugetlbfs":'\

There is a typo on this sed, you need to terminate it by "@" and not ":"
or you will have:
"sed: -e expression #1, char 76: unterminated `s' command" error

> +/etc/sysconfig/openvswitch
> +%endif
> +
>  # In the case of upgrade, this is not needed.
>  chown -R openvswitch:openvswitch /etc/openvswitch
>  fi
> diff --git a/rhel/usr_lib_systemd_system_ovs-vswitchd.service 
> b/rhel/usr_lib_systemd_system_ovs-vswitchd.service.in
> similarity index 87%
> rename from rhel/usr_lib_systemd_system_ovs-vswitchd.service
> rename to rhel/usr_lib_systemd_system_ovs-vswitchd.service.in
> index 48231b3..2d2e9e6 100644
> --- a/rhel/usr_lib_systemd_system_ovs-vswitchd.service
> +++ b/rhel/usr_lib_systemd_system_ovs-vswitchd.service.in
> @@ -10,8 +10,12 @@ PartOf=openvswitch.service
>  [Service]
>  Type=forking
>  Restart=on-failure
> +Environment="HOME=/var/run/openvswitch"

Usually in systemd service file you don't need to put quotes, unless you
need to assign a value containing spaces to a variable.

>  EnvironmentFile=/etc/openvswitch/default.conf
>  EnvironmentFile=-/etc/sysconfig/openvswitch
> +@begin_dpdk@
> +ExecStartPre="/usr/sbin/chown :hugetlbfs /dev/hugepages"

quotes are not needed in ExecStartPre too and I suggest to remove them
for coherency with the other service files.
chown is in "/usr/bin" 

Re: [ovs-dev] [PATCH 2/2] rhel: delete transient ports on boot when starting ovsdb-server

[Timothy M. Redaelli]

On 05/05/2017 08:41 PM, Aaron Conole wrote:
[...]
>> @@ -534,7 +538,7 @@ fi
>>  %doc COPYING NOTICE README.rst NEWS rhel/README.RHEL.rst
>>  /var/lib/openvswitch
>>  /var/log/openvswitch
>> -%ghost %attr(755,root,root) %{_rundir}/openvswitch
>> +%dir %{_rundir}/openvswitch
> 
> Doesn't this still need to be flagged as %ghost?

If you use %ghost the directory is not created by rpm and it's only
created after a reboot, since tmpfiles.d scripts are only executed by
systemd on boot.

Using %dir is the suggested way by Fedora Wiki:
https://fedoraproject.org/wiki/Packaging:Tmpfiles.d
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] tests/ofproto.at: Workaround some races

[Timothy M. Redaelli]

On 05/03/2017 10:28 PM, Ben Pfaff wrote:
> On Thu, Apr 27, 2017 at 05:24:22PM +0200, Timothy Redaelli wrote:
>> Port commit a6d1a2997db4:
>> ofproto.at: Workaround a race
>>
>> While a barrier serializes requests from the same connection,
>> it doesn't wait for requests from other connections to the switch.
>> Replace the barrier with infamous "sleep 1" to workaround the problem.
>>
>> to the following tests:
>> "ofproto - asynchronous message control (OpenFlow 1.0)",
>> "ofproto - asynchronous message control (OpenFlow 1.3)",
>> "ofproto - asynchronous message control (OpenFlow 1.4)" and
>> "ofproto - asynchronous message control (OpenFlow 1.5)"
>>
>> Sometimes one of these tests fails because the OFPT_BARRIER_REPLY is
>> printed before the other message we expect to have.
>>
>> Suggested-by: Lance Richardson 
>> Signed-off-by: Timothy Redaelli 
> 
> Thanks for working to make the tests more reliable.
> 
> Adding "sleep 1" makes tests slower, especially when it's in a loop as
> it is here.
> 
> What if we instead make the test wait until monitor.log grows large
> enough?  We already have OVS_WAIT_UNTIL, which waits only a short time
> when it can.
> 
> What do you think of this?

The problem is not that the logfile is not large enough, the problem is
that we have "reversed" output:

--- expout  2017-05-04 06:22:03.352044646 -0400
+++ /root/ovs/tests/testsuite.dir/at-groups/944/stdout  2017-05-04
06:22:03.386044787 -0400
@@ -5,5 +5,5 @@
  state:  LIVE
  speed: 0 Mbps now, 0 Mbps max
 OFPT_FLOW_REMOVED (OF1.4):  reason=delete table_id=0
-OFPT_FLOW_REMOVED (OF1.4):  reason=group_delete table_id=0
 OFPT_BARRIER_REPLY (OF1.4):
+OFPT_FLOW_REMOVED (OF1.4):  reason=group_delete table_id=0

Complete logs at
https://gist.github.com/drizzt/dd5a3c4b2bea0d83e756f26c88fda418
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH 4/8] ovsdb-idl: Fix memory leak

[Timothy M. Redaelli]

On 04/24/2017 06:30 PM, Ben Pfaff wrote:
> On Fri, Apr 07, 2017 at 02:43:42PM -0700, Yi-Hung Wei wrote:
>> In testcase "simple idl, conditional, multiple clauses in condition - C",
>> valgrind reports a memory leak with the following call stack.
>> xmalloc (util.c:112)
>> resize (hmap.c:100)
>> ovsdb_idl_condition_clone (ovsdb-idl.c:1075)
>> ovsdb_idl_set_condition (ovsdb-idl.c:1095)
>> update_conditions (test-ovsdb.c:2299)
>> do_idl (test-ovsdb.c:2388)
>> ovs_cmdl_run_command__ (command-line.c:115)
>> main (test-ovsdb.c:73)
>>
>> Signed-off-by: Yi-Hung Wei 
> 
> Thanks, I applied this to master, branch-2.7, and branch-2.6.

Hi,
unlucky this commit breaks build on branch-2.6 since the function
"ovsdb_idl_condition_destroy" was added in commit
0164e367f5d8 ("ovsdb-idl: Change interface to conditional monitoring.")
and it's not available on branch 2.6.
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] ofproto: Report only un-deleted groups in group stats replies.

[Timothy M. Redaelli]

On 04/19/2017 08:29 PM, Ben Pfaff wrote:
> Deleted groups hang around in the group table until the next grace period,
> so it's important for the group stats code to pretend that they're gone
> until they really get deleted.

[...]
Thank you for the patch, I tested it in all the scenarios and it works.

Tested-by: Timothy Redaelli 
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] m4: Add hard requirements for python in "configure"

[Timothy M. Redaelli]

On 04/06/2017 12:16 AM, Ben Pfaff wrote:
> On Wed, Mar 29, 2017 at 01:49:42PM +0200, Timothy Redaelli wrote:
>> Since Python 2.7 and python-six are needed to build Open vSwitch,
>> ./configure should return an error if they are missing or if they are too old
>>
>> Signed-off-by: Timothy Redaelli 
> 
> Thanks for the patch.
> 
> I think that if we do it this way, then eventually we'll have to debug
> the problem.  I suggest that we use the following instead, which does a
> better job of dumping problems into config.log for troubleshooting.
> 
> What do you think?

LGTM

Tested cases:
 - missing python
 - python version < 2.7
 - python version >= 2.7, without python-six
 - python version >= 2.7, with old python-six

 - python version >= 2.7, with correct python-six

Thank you

___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev