Yes, from following trace, DRBD and OVS block each other and enter into
deadlock.
It seems someone sent a patch years ago to refine genl global lock as family
granularity. So change the DRBD or OVS's genl family to different ones may fix
this problem?
Thanks,
Tianpeng
Set primary
On Fri, Mar 29, 2013 at 8:38 PM, Tianpeng Zhang (Gmail)
wrote:
> Hi All,
>
> I met an issue when running DRBD in Xenserver with ovs-1.7.1. DRBD works
> fine when creating and sync data. But when trying to down DRBD resource,
> ovs-vswitchd hangs for about 20 minutes, then all network connections
>
Hi All,
I met an issue when running DRBD in Xenserver with ovs-1.7.1. DRBD works fine
when creating and sync data. But when trying to down DRBD resource,
ovs-vswitchd hangs for about 20 minutes, then all network connections broken.
I add some debug trace, ovs-vswitchd finally stopped at sendmsg
On Mar 29, 2013, at 4:16 PM, Ben Pfaff wrote:
> On Wed, Mar 27, 2013 at 11:02:21PM -0700, Neil Mckee wrote:
>> This patch adds an sFlow test to the test suite (in branch 1.10). To make
>> that work properly I added netdev_dummy_get_ifindex() so that a dummy netdev
>> can return a dummy ifindex
On Wed, Mar 27, 2013 at 10:44 PM, Simon Horman wrote:
> On Thu, Mar 21, 2013 at 04:50:03PM +0900, Simon Horman wrote:
>> Allow datapath to recognize and extract MPLS labels into flow keys
>> and execute actions which push, pop, and set labels on packets.
>>
>> Based heavily on work by Leo Alterman
On Fri, Mar 29, 2013 at 6:46 AM, Thomas Graf wrote:
> A couple of small Open vSwitch fixes and cleanups that accumulated
> while working on larger pieces.
>
> The biggest change is the exposure of to user
> space to make the Netlink interface available without requiring every
> appliation to copy
On Fri, Mar 29, 2013 at 12:52 PM, David Miller wrote:
> From: Thomas Graf
> Date: Fri, 29 Mar 2013 14:46:46 +0100
>
> > A couple of small Open vSwitch fixes and cleanups that accumulated
> > while working on larger pieces.
> >
> > The biggest change is the exposure of to user
> > space to make
On Wed, Mar 27, 2013 at 11:02:21PM -0700, Neil Mckee wrote:
> This patch adds an sFlow test to the test suite (in branch 1.10). To make
> that work properly I added netdev_dummy_get_ifindex() so that a dummy netdev
> can return a dummy ifindex when asked. Is there anywhere in OVS that
> assum
On Fri, Mar 29, 2013 at 6:14 AM, 黄登辉 wrote:
> My understanding of how internal type device work.
>Internal type device is special. when a packet sent to internal
> type device from outside, it will immediately forward to openvswitch module
> to decide where to go?
> when a packet sent t
From: Thomas Graf
Date: Fri, 29 Mar 2013 14:46:46 +0100
> A couple of small Open vSwitch fixes and cleanups that accumulated
> while working on larger pieces.
>
> The biggest change is the exposure of to user
> space to make the Netlink interface available without requiring every
> appliation t
On Fri, Mar 29, 2013 at 12:08:32PM +0800, Yeming Zhao wrote:
> I spent last couple days writing a tool interacting with OVSDB, using the
> ovs python package in OVS distribution. It generally works.
The Python bindings are generally not as mature as the C bindings.
They have some rough edges.
> B
Specifying the minimal length in the policy makes it reuseable
and documents the interface.
Signed-off-by: Thomas Graf
---
net/openvswitch/datapath.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 8759265..5b
Signed-off-by: Thomas Graf
---
include/linux/openvswitch.h | 13 +++--
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/include/linux/openvswitch.h b/include/linux/openvswitch.h
index 67d6c7b..8b9d721 100644
--- a/include/linux/openvswitch.h
+++ b/include/linux/openvswitch.h
It contains the public netlink interface bits required by userspace to
make use of the interface.
Signed-off-by: Thomas Graf
---
include/linux/openvswitch.h | 433 +
include/uapi/linux/Kbuild| 1 +
include/uapi/linux/openvswitch.h | 456
Kills the FLOW_BUFSIZE constant which needs to be calculated manually
and replaces it with key_attr_size() based on nla_total_size().
Calculates the size of datapath messages instead of relying on
NLMSG_DEFAULT_SIZE and moves the existing message size calculations
into own functions for clarity.
S
Signed-off-by: Thomas Graf
---
net/openvswitch/datapath.c | 36
1 file changed, 16 insertions(+), 20 deletions(-)
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 49ee37b..d406503 100644
--- a/net/openvswitch/datapath.c
+++ b/net/ope
Less error prone as it takes into account the length of both the
destination buffer and the source attribute and documents when
data is copied from an attribute.
Signed-off-by: Thomas Graf
---
net/openvswitch/datapath.c | 2 +-
net/openvswitch/flow.c | 2 +-
2 files changed, 2 insertions(+),
A couple of small Open vSwitch fixes and cleanups that accumulated
while working on larger pieces.
The biggest change is the exposure of to user
space to make the Netlink interface available without requiring every
appliation to copy the header file from a source tree.
Thomas Graf (6):
openvsw
My understanding of how internal type device work.
Internal type device is special. when a packet sent to internal type
device from outside, it will immediately forward to openvswitch module to
decide where to go?
when a packet sent to internal type device from openvswitch module, it will
19 matches
Mail list logo
