The idea is to run each test twice, once with ovn-northd, once
with ovn-northd-ddlog. To do that, we add a macro OVN_FOR_EACH_NORTHD
and bracket each test that uses ovn-northd in it.
Signed-off-by: Ben Pfaff
Acked-by: Dumitru Ceara
---
tests/ovn-ic.at | 11 +-
tests/ovn-macros.at | 96
t commit 36a8745de859 ("Add support for DHCP options 28 (Broadcast
Address)")
- Add patch to improve "ARP/ND request broadcast limiting" test.
v4:
- Fix dependencies for parallel build.
- Fix spelling error in documentation.
- Use --wait=sb, not --wait=hv, when no chassis
;ports' is a set of weak references) but if you try to delete
the last port that the group contained, the database server will reject
the whole transaction. That's really weird. By allowing a multicast
group with no ports, we avoid this special case.
Signed-off-by: Ben Pfaff
---
Otherwise this test is racy because it assumes that northd finishes
its updates between the ovn-nbctl changes and the subsequent checks.
Also, simplify some series of "check_column" into single
"check_row_count" calls (that then become "wait_row_count").
Signed-off
not do
log rate limiting.
Signed-off-by: Leonid Ryzhyk
Signed-off-by: Ben Pfaff
Acked-by: Dumitru Ceara
---
TODO.rst | 6 ++
lib/ovn-util.c | 15 +++
lib/ovn-util.h | 5 +
3 files changed, 26 insertions(+)
diff --git a/TODO.rst b/TODO.rst
index c15815539f4f
The code here checked whether 'bfd_e' was null even though that wasn't
necessary given the control flow just above it.
Signed-off-by: Ben Pfaff
---
northd/ovn-northd.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/northd/ovn-northd.c b/nort
On Fri, Dec 11, 2020 at 03:43:22PM +0100, Dumitru Ceara wrote:
> > diff --git a/tests/ovn.at b/tests/ovn.at
> > index 6e396895826f..2e4eaf7f12f7 100644
> > --- a/tests/ovn.at
> > +++ b/tests/ovn.at
> > @@ -16982,6 +16982,10 @@ AT_CLEANUP
> >
> > OVN_FOR_EACH_NORTHD([
> > AT_SETUP([ovn -- IGMP s
On Fri, Dec 11, 2020 at 03:43:22PM +0100, Dumitru Ceara wrote:
> There are a few tests that fail (also northd-C tests) and I suspect it's
> because the OVS IDL split patches [0] this commit depends on don't
> include some of the recent fixes that went into the OVS IDL code, at
> least the following
On Fri, Feb 12, 2021 at 11:15:46AM -0800, William Tu wrote:
> I'm confused with what l2_pad_size is.
> I thought it's between L2 and L3 header, there is a 2-byte padding to
> make it 16-byte alignment. But it doesn't look like that.
Ethernet has a 64-byte minimum packet length. l2_pad_size is the
On Wed, Feb 10, 2021 at 08:51:22PM +0100, Ilya Maximets wrote:
> On 2/10/21 8:38 PM, Ben Pfaff wrote:
> > On Wed, Feb 10, 2021 at 03:39:46PM +0100, Ilya Maximets wrote:
> >> From: Flavio Leitner
> >>
> >> Although not required, padding can be optionally added
On Wed, Feb 10, 2021 at 03:39:46PM +0100, Ilya Maximets wrote:
> From: Flavio Leitner
>
> Although not required, padding can be optionally added until
> the packet length is MTU bytes. A packet with extra padding
> currently fails sanity checks.
I think that, as much as anything else, this vulne
On Mon, Feb 08, 2021 at 11:49:03AM +0100, Lorenzo Bianconi wrote:
> Signed-off-by: Lorenzo Bianconi
Thanks!
Acked-by: Ben Pfaff
I suggest folding in the following as well to make it clearer the roles
of the different groups of columns.
diff --git a/ovn-nb.xml b/ovn-nb.xml
index 58083f101
On Fri, Feb 05, 2021 at 11:50:21PM +0100, Dumitru Ceara wrote:
> On 2/5/21 8:50 PM, Ben Pfaff wrote:
> > The existing code here seemed like an expensive way to find every ACL.
>
> It is :)
>
> > This simplifies it.
> >
> > There is one behavioral change he
t was why
the code was written the way it was.
Signed-off-by: Ben Pfaff
---
northd/ovn-northd.c | 27 ++-
1 file changed, 6 insertions(+), 21 deletions(-)
diff --git a/northd/ovn-northd.c b/northd/ovn-northd.c
index 63966ef22a21..d6e921bbb2ac 100644
--- a/northd/ovn-nor
On Fri, Jan 29, 2021 at 03:03:05PM +0100, Ilya Maximets wrote:
> Some manpages are generated from rST, but these are not included
> in 'dist-docs' make target.
>
> Fixes: fd0837a76f4c ("doc: Convert ovs-vlan-test to rST")
> Signed-off-by: Ilya Maximets
ou didn't hit this on your machines, but please let me know what
> additional info you would require.
I was able to fix this with the following:
-8<--cut here-->8--
From: Ben Pfaff
Date: Mon, 21 Dec 2020 18:17:57 -0800
Subject: [PATCH ovn
On Fri, Dec 11, 2020 at 01:36:31PM +0100, Dumitru Ceara wrote:
> On 12/2/20 7:50 AM, Ben Pfaff wrote:
> > These changes should make it easier to debug various tests.
> >
> > Signed-off-by: Ben Pfaff
> > ---
>
> Looks good to me, thanks!
>
> Acked-by: Du
On Fri, Dec 11, 2020 at 01:36:36PM +0100, Dumitru Ceara wrote:
> On 12/2/20 7:50 AM, Ben Pfaff wrote:
> > Signed-off-by: Ben Pfaff
> > ---
>
> Looks good to me, thanks!
>
> Acked-by: Dumitru Ceara
Thanks, applied to master.
___
On Tue, Feb 02, 2021 at 05:27:08PM +0100, Ilya Maximets wrote:
> On 11/12/20 12:22 AM, Ben Pfaff wrote:
> > I would have found these useful for the OVN tests. The {in} operator
> > is the same as {<=}, but it's still useful to have the alternate syntax
> > because mos
On Tue, Feb 02, 2021 at 05:11:09PM +0100, Ilya Maximets wrote:
> On 1/21/21 11:33 PM, Ben Pfaff wrote:
> > The calculation in tcp_reader_run() failed to account for L2 padding.
> > This fixes the problem, by moving the existing function
> > tcp_payload_length() from a conntrac
On Tue, Feb 02, 2021 at 05:15:20PM +0100, Ilya Maximets wrote:
> On 1/21/21 11:49 PM, Ben Pfaff wrote:
> > Signed-off-by: Ben Pfaff
> > ---
> > lib/ofp-table.c | 9 ++---
> > 1 file changed, 6 insertions(+), 3 deletions(-)
> >
> > diff --git a/lib/
The test "ovn -- nested containers" adds, removes, and then re-adds a
container port bar2. It appears to me that ovn-controller never brings
the re-added port up. I don't know why.
To see the problem, apply the following patch then run that test
(currently test 88):
diff --git a/tests/ovn.at b/
On Fri, Dec 11, 2020 at 01:36:25PM +0100, Dumitru Ceara wrote:
> On 12/2/20 7:50 AM, Ben Pfaff wrote:
> > Many of these could be replaced by "ovn-nbctl sync". Some weren't
> > really needed at all because they were adjacent to something that itself
> > called
On Fri, Dec 11, 2020 at 01:36:13PM +0100, Dumitru Ceara wrote:
> On 12/2/20 7:50 AM, Ben Pfaff wrote:
> > Otherwise when something hangs or goes wrong, it's not obvious what
> > happened.
> >
> > Signed-off-by: Ben Pfaff
> > ---
>
> Looks good
On Mon, Feb 01, 2021 at 08:15:32PM +0800, Tao YunXiang wrote:
> This patch populate faq about datapath-id.
>
> Author: Tao YunXiang
There's no "Author:" tag. We use Git to track commit authorship. I
deleted this.
> Signed-off-by: Tao YunXiang
I edited the answer to be better suited for the
On Fri, Jan 22, 2021 at 12:50:38AM +0100, Ilya Maximets wrote:
> On 1/22/21 12:42 AM, Ben Pfaff wrote:
> > On Wed, Jan 13, 2021 at 05:32:52PM +0100, Dumitru Ceara wrote:
> >> On 1/13/21 2:56 AM, Ilya Maximets wrote:
> >>> On the side note: don't we need
On Wed, Jan 13, 2021 at 05:32:52PM +0100, Dumitru Ceara wrote:
> On 1/13/21 2:56 AM, Ilya Maximets wrote:
> > On the side note: don't we need to increase 'cond_seqno' here since
> > we're practically dropping one of the condition requests? User might
> > wait for this sequence number, but 'req_con
On Wed, Jan 13, 2021 at 02:56:45AM +0100, Ilya Maximets wrote:
> On 1/8/21 2:59 PM, Dumitru Ceara wrote:
> > On 12/19/20 3:44 AM, Ben Pfaff wrote:
> >> +/* If there was no "unsent" condition but instead a
> >> + * monitor_cond_change re
On Fri, Jan 08, 2021 at 02:59:24PM +0100, Dumitru Ceara wrote:
> On 12/19/20 3:44 AM, Ben Pfaff wrote:
> > This change breaks the IDL into two layers: the IDL proper, whose
> > interface to its client is unchanged, and a low-level library called
> > the OVSDB "client syn
On Wed, Jan 13, 2021 at 02:20:00AM +0100, Ilya Maximets wrote:
> On 12/19/20 3:44 AM, Ben Pfaff wrote:
> > Until now, "test-ovsdb idl" has printed the steps that it goes through
> > to stdout. This commit also makes it log the same information. This
> > makes it ea
Signed-off-by: Ben Pfaff
---
lib/ofp-table.c | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/lib/ofp-table.c b/lib/ofp-table.c
index 6ed6c5e3b274..a956754f2d56 100644
--- a/lib/ofp-table.c
+++ b/lib/ofp-table.c
@@ -523,9 +523,12 @@ ofputil_decode_table_features
The calculation in tcp_reader_run() failed to account for L2 padding.
This fixes the problem, by moving the existing function
tcp_payload_length() from a conntrack private header file into
dp-packet.h and renaming it to suit the dp_packet style.
Signed-off-by: Ben Pfaff
---
lib/conntrack
On Tue, Dec 22, 2020 at 10:38:50AM +0100, Dumitru Ceara wrote:
> On 12/22/20 12:03 AM, Ben Pfaff wrote:
> > On Fri, Dec 18, 2020 at 12:47:12PM +0100, Dumitru Ceara wrote:
> >> On 12/17/20 7:55 PM, Ben Pfaff wrote:
> >>> On Thu, Dec 17, 2020 at 08:54:56AM
On Mon, Dec 28, 2020 at 07:02:55PM -0300, Flavio Leitner wrote:
> The initial website page is difficult to read because of
> the large amount of links from different parts of the whole
> documentation. Most of all those links come from their
> index page referenced in the section 'Contents' on the
I am not a fan of the term "out of tree" because it assumes that the
reader knows what tree you're talking about. I would guess that people
reading OVS documentation tend to be thinking about OVS, not Linux, so
if they're thinking about any tree at all, it's the OVS tree.
So:
> +Beginning wi
This series still applies, and needs a final review.
On Fri, Dec 18, 2020 at 06:44:21PM -0800, Ben Pfaff wrote:
> This series breaks the IDL into two layers: the IDL proper, whose
> interface to its client is unchanged, and a low-level library called
> the OVSDB "client synchr
On Tue, Jan 05, 2021 at 09:57:46PM +0100, Ilya Maximets wrote:
> On 12/22/20 1:21 AM, Ben Pfaff wrote:
> > This follows up on commit 4241d652e465 ("jsonrpc: Avoid disconnecting
> > prematurely due to long poll intervals."), which implemented the same
> > thing in
On Mon, Dec 21, 2020 at 12:58:14PM +0100, Ilya Maximets wrote:
> On 12/19/20 3:11 AM, Ben Pfaff wrote:
> > On Fri, Dec 18, 2020 at 10:11:15PM +0100, Ilya Maximets wrote:
> >> On 12/18/20 6:31 AM, Ben Pfaff wrote:
> >>> Open vSwitch has a few different jsonrpc-
This follows up on commit 4241d652e465 ("jsonrpc: Avoid disconnecting
prematurely due to long poll intervals."), which implemented the same
thing in C.
Requested-by: Ilya Maximets
Signed-off-by: Ben Pfaff
---
python/ovs/jsonrpc.py | 5 -
python/ovs/reconne
On Fri, Dec 18, 2020 at 12:47:12PM +0100, Dumitru Ceara wrote:
> On 12/17/20 7:55 PM, Ben Pfaff wrote:
> > On Thu, Dec 17, 2020 at 08:54:56AM -0800, Girish Moodalbail wrote:
> >> Hello all,
> >>
> >> Say, ovn-nbctl is started in daemon mode with options set fo
This new module has a single direct user now. In the future, it
will also be used by OVN.
Signed-off-by: Ben Pfaff
---
lib/automake.mk | 2 +
lib/ovsdb-cs.c | 326 +++
lib/ovsdb-cs.h | 70
lib/ovsdb-idl.c | 442
Until now, "test-ovsdb idl" has printed the steps that it goes through
to stdout. This commit also makes it log the same information. This
makes it easier to match up the steps with the rest of the log (in
particular with the jsonrpc logging).
Signed-off-by: Ben Pfaff
---
tests/te
g.
v3->v4:
- Apply (and drop) several reviewed patches.
- Log more and fix minor nits in "test-ovsdb: Log steps in idl test.".
- Fix memory leak in ovsdb_cs_db_init() and in ovsdb_cs_send_transaction()
in "ovsdb-idl: Break into two layers."
Ben Pfaff (3):
test-ovs
On Fri, Dec 18, 2020 at 10:31:19AM +0100, Dumitru Ceara wrote:
> On 12/18/20 6:31 AM, Ben Pfaff wrote:
> > This change breaks the IDL into two layers: the IDL proper, whose
> > interface to its client is unchanged, and a low-level library called
> > the OVSDB "client syn
On Fri, Dec 18, 2020 at 11:16:50PM +0100, Ilya Maximets wrote:
> On 12/18/20 6:31 AM, Ben Pfaff wrote:
> > Until now, "test-ovsdb idl" has printed the steps that it goes through
> > to stdout. This commit also makes it log the same information. This
> > makes it ea
On Fri, Dec 18, 2020 at 10:11:15PM +0100, Ilya Maximets wrote:
> On 12/18/20 6:31 AM, Ben Pfaff wrote:
> > Open vSwitch has a few different jsonrpc-based protocols that depend on
> > jsonrpc_session to make sure that the connection is up and working.
> > In turn, jso
Signed-off-by: Ben Pfaff
---
lib/ovsdb-idl.c | 8
1 file changed, 8 insertions(+)
diff --git a/lib/ovsdb-idl.c b/lib/ovsdb-idl.c
index 02a49b32454b..c6b6075d51e3 100644
--- a/lib/ovsdb-idl.c
+++ b/lib/ovsdb-idl.c
@@ -617,6 +617,14 @@ ovsdb_idl_destroy(struct ovsdb_idl *idl
This new module has a single direct user now. In the future, it
will also be used by OVN.
Signed-off-by: Ben Pfaff
---
lib/automake.mk | 2 +
lib/ovsdb-cs.c | 326 +++
lib/ovsdb-cs.h | 70
lib/ovsdb-idl.c | 442
Signed-off-by: Ben Pfaff
---
lib/ovsdb-idl.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/lib/ovsdb-idl.h b/lib/ovsdb-idl.h
index a1a577664e8e..789337df9f33 100644
--- a/lib/ovsdb-idl.h
+++ b/lib/ovsdb-idl.h
@@ -76,7 +76,6 @@ void ovsdb_idl_set_lock(struct ovsdb_idl *, const char
Until now, "test-ovsdb idl" has printed the steps that it goes through
to stdout. This commit also makes it log the same information. This
makes it easier to match up the steps with the rest of the log (in
particular with the jsonrpc logging).
Signed-off-by: Ben Pfaff
---
tests/te
Adding parameter names makes these prototypes clearer.
Signed-off-by: Ben Pfaff
---
lib/ovsdb-idl.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/ovsdb-idl.h b/lib/ovsdb-idl.h
index 789337df9f33..05bb48d66c3f 100644
--- a/lib/ovsdb-idl.h
+++ b/lib/ovsdb-idl.h
When there's no open session, we still have to free the messages that
we make but cannot send.
I'm not confident that these fix actual bugs, because it seems possible
that these code paths can only be hit when the session is nonnull.
Signed-off-by: Ben Pfaff
---
lib/ovsdb-
ovsdb_idl_db_parse_monitor_reply() clears the IDL and parses the
received data. There's no need to do it again afterward.
Signed-off-by: Ben Pfaff
Fixes: 1b1d2e6daa56 ("ovsdb: Introduce experimental support for clustered
databases.")
---
lib/ovsdb-idl.c | 3 ---
1 file chang
7;m pretty sure that I've seen it a number of times in testing.
Signed-off-by: Ben Pfaff
---
lib/jsonrpc.c | 5 -
lib/reconnect.c| 25 +++--
lib/reconnect.h| 1 +
tests/test-reconnect.c | 1 +
4 files changed, 29 insertions(+), 3 deletions(-)
havior reproduce
the same behavior as the existing idl. This fixes a test failure
I was seeing after (but not before) rebase.
- I don't see any failures in upstream ovn with this applied.
- New patch "test-ovsdb: Log steps in idl test." that I found useful
in debug
On Thu, Dec 17, 2020 at 08:54:56AM -0800, Girish Moodalbail wrote:
> Hello all,
>
> Say, ovn-nbctl is started in daemon mode with options set for certs, and
> those certs do not exist on the file system. For example, in the following
> invocation assume that `/ovn-cert` folder is empty
>
> ovn-nb
On Fri, Dec 11, 2020 at 09:54:47PM +0100, Ilya Maximets wrote:
> Currently, ovsdb-server stores complete value for the column in a database
> file and in a raft log in case this column changed. This means that
> transaction that adds, for example, one new acl to a port group creates
> a log entry
true, argv[1], &in6addr_any);
> +
> unixctl_command_reply(conn, "OK");
> } else {
> unixctl_command_reply_error(conn, error);
>
> (I believe the code should also remove the local entries on IP removed.)
>
> Is there
These changes should make it easier to debug various tests.
Signed-off-by: Ben Pfaff
---
tests/ovn-controller-vtep.at | 17 +-
tests/ovn-northd.at | 144 +---
tests/ovn.at | 413 ---
tests/ovs-macros.at | 20 +-
4 files
wait for what they were waiting for, or because I didn't understand
the circumstances deeply enough.
Signed-off-by: Ben Pfaff
---
tests/ovn-macros.at | 15 +++
tests/ovn-northd.at | 11 +-
tests/ovn.at| 259 ++--
3 files changed, 155 insertions(+
not do
log rate limiting.
Signed-off-by: Leonid Ryzhyk
Signed-off-by: Ben Pfaff
---
TODO.rst | 6 ++
lib/ovn-util.c | 15 +++
lib/ovn-util.h | 5 +
3 files changed, 26 insertions(+)
diff --git a/TODO.rst b/TODO.rst
index c15815539f4f..ecfe62870fb0 100644
--- a/TODO.rst
P and MLD tests by porting commit 9d2e8d32fb98
("ofctrl.c: Fix duplicated flow handling in I-P while merging
opposite changes."), which I had missed before.
Ben Pfaff (5):
ovs-sandbox: Print and check results of *ctl commands.
tests: Prepare for multiple northd types.
tes
The idea is to run each test twice, once with ovn-northd, once
with ovn-northd-ddlog. To do that, we add a macro OVN_FOR_EACH_NORTHD
and bracket each test that uses ovn-northd in it.
Signed-off-by: Ben Pfaff
Acked-by: Dumitru Ceara
---
tests/ovn-ic.at | 11 +-
tests/ovn-macros.at | 96
Signed-off-by: Ben Pfaff
---
tests/ovn.at | 32 ++--
1 file changed, 14 insertions(+), 18 deletions(-)
diff --git a/tests/ovn.at b/tests/ovn.at
index 9b419ec3aa7a..6e396895826f 100644
--- a/tests/ovn.at
+++ b/tests/ovn.at
@@ -1838,42 +1838,38 @@ ovn_start
# Turn on
Otherwise when something hangs or goes wrong, it's not obvious what
happened.
Signed-off-by: Ben Pfaff
---
tutorial/ovs-sandbox | 32
1 file changed, 16 insertions(+), 16 deletions(-)
diff --git a/tutorial/ovs-sandbox b/tutorial/ovs-sandbox
index 1841776
On Thu, Nov 26, 2020 at 06:35:44PM +0530, Numan Siddique wrote:
> On Thu, Nov 26, 2020 at 11:30 AM Numan Siddique wrote:
> >
> > On Thu, Nov 26, 2020 at 10:54 AM Ben Pfaff wrote:
> > >
> > > On Wed, Nov 25, 2020 at 01:13:22PM +0530, Numan Siddique wrote:
> &
This new module has a single direct user now. In the future, it
will also be used by OVN.
Signed-off-by: Ben Pfaff
---
lib/automake.mk | 2 +
lib/ovsdb-cs.c | 326 +++
lib/ovsdb-cs.h | 70
lib/ovsdb-idl.c | 442
Signed-off-by: Ben Pfaff
---
lib/ovsdb-idl.c | 8
1 file changed, 8 insertions(+)
diff --git a/lib/ovsdb-idl.c b/lib/ovsdb-idl.c
index 5502fdab8644..e1013da86c56 100644
--- a/lib/ovsdb-idl.c
+++ b/lib/ovsdb-idl.c
@@ -617,6 +617,14 @@ ovsdb_idl_destroy(struct ovsdb_idl *idl
Adding parameter names makes these prototypes clearer.
Signed-off-by: Ben Pfaff
---
lib/ovsdb-idl.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/ovsdb-idl.h b/lib/ovsdb-idl.h
index 789337df9f33..05bb48d66c3f 100644
--- a/lib/ovsdb-idl.h
+++ b/lib/ovsdb-idl.h
7;m pretty sure that I've seen it a number of times in testing.
Signed-off-by: Ben Pfaff
---
lib/jsonrpc.c | 5 -
lib/reconnect.c| 25 +++--
lib/reconnect.h| 1 +
tests/test-reconnect.c | 1 +
4 files changed, 29 insertions(+), 3 deletions(-)
Signed-off-by: Ben Pfaff
---
lib/ovsdb-idl.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/lib/ovsdb-idl.h b/lib/ovsdb-idl.h
index a1a577664e8e..789337df9f33 100644
--- a/lib/ovsdb-idl.h
+++ b/lib/ovsdb-idl.h
@@ -76,7 +76,6 @@ void ovsdb_idl_set_lock(struct ovsdb_idl *, const char
When there's no open session, we still have to free the messages that
we make but cannot send.
I'm not confident that these fix actual bugs, because it seems possible
that these code paths can only be hit when the session is nonnull.
Signed-off-by: Ben Pfaff
---
lib/ovsdb-
ovsdb_idl_db_parse_monitor_reply() clears the IDL and parses the
received data. There's no need to do it again afterward.
Signed-off-by: Ben Pfaff
Fixes: 1b1d2e6daa56 ("ovsdb: Introduce experimental support for clustered
databases.")
---
lib/ovsdb-idl.c | 3 ---
1 file chang
tracking, but these patches will not make it better or worse.
There are lots of idl patches flying around on the mailing list
currently; maybe one of those fixes the real problem.
Ben Pfaff (8):
jsonrpc: Avoid disconnecting prematurely due to long poll intervals.
ovsdb-idl: Av
On Thu, Nov 26, 2020 at 06:35:44PM +0530, Numan Siddique wrote:
> On Thu, Nov 26, 2020 at 11:30 AM Numan Siddique wrote:
> >
> > On Thu, Nov 26, 2020 at 10:54 AM Ben Pfaff wrote:
> > >
> > > On Wed, Nov 25, 2020 at 01:13:22PM +0530, Numan Siddique wrote:
> &
These changes should make it easier to debug various tests.
Signed-off-by: Ben Pfaff
---
tests/ovn-controller-vtep.at | 17 +-
tests/ovn-northd.at | 144 +---
tests/ovn.at | 413 ---
tests/ovs-macros.at | 20 +-
4 files
The idea is to run each test twice, once with ovn-northd, once
with ovn-northd-ddlog. To do that, we add a macro OVN_FOR_EACH_NORTHD
and bracket each test that uses ovn-northd in it.
Signed-off-by: Ben Pfaff
Acked-by: Dumitru Ceara
---
tests/ovn-ic.at | 11 +-
tests/ovn-macros.at | 96
Signed-off-by: Ben Pfaff
---
tests/ovn.at | 32 ++--
1 file changed, 14 insertions(+), 18 deletions(-)
diff --git a/tests/ovn.at b/tests/ovn.at
index a6f2bf8ad2b5..438545693a6a 100644
--- a/tests/ovn.at
+++ b/tests/ovn.at
@@ -1838,42 +1838,38 @@ ovn_start
# Turn on
wait for what they were waiting for, or because I didn't understand
the circumstances deeply enough.
Signed-off-by: Ben Pfaff
---
tests/ovn-macros.at | 15 +++
tests/ovn-northd.at | 11 +-
tests/ovn.at| 259 ++--
3 files changed, 155 insertions(+
and AS_BOX(...) to announce parts of a test)
- Modernize some of the ddlog code a bit, e.g. by using the method-like
calls such as foo.get("key") in place of map_get(foo, "key").
Ben Pfaff (5):
ovs-sandbox: Print and check results of *ctl commands.
tests: Prepare fo
not do
log rate limiting.
Signed-off-by: Leonid Ryzhyk
Signed-off-by: Ben Pfaff
---
TODO.rst | 6 ++
lib/ovn-util.c | 15 +++
lib/ovn-util.h | 5 +
3 files changed, 26 insertions(+)
diff --git a/TODO.rst b/TODO.rst
index c15815539f4f..ecfe62870fb0 100644
--- a/TODO.rst
Otherwise when something hangs or goes wrong, it's not obvious what
happened.
Signed-off-by: Ben Pfaff
---
tutorial/ovs-sandbox | 32
1 file changed, 16 insertions(+), 16 deletions(-)
diff --git a/tutorial/ovs-sandbox b/tutorial/ovs-sandbox
index 1841776
On Thu, Nov 19, 2020 at 12:53:45PM +0530, Numan Siddique wrote:
> I have few comments w.r.t to scaling.
>
> 1. In the last OVN weekly irc meeting, I had mentioned about
> ovn-northd-ddlog taking a lot of time in the loop. I also shared the
> scale test results with ddlog
> (https://imgur.com/JJpsM
On Wed, Nov 25, 2020 at 01:13:22PM +0530, Numan Siddique wrote:
> On Wed, Nov 25, 2020 at 4:21 AM Ben Pfaff wrote:
> >
> > The tests "superseding ACLs with conjunction" and "ARP replies for SNAT
> > external ips" trigger bugs in the ovn-controller incr
s,
since I don't know anything about the feature.
Ben Pfaff (1):
tests: Work around ovn-controller incremental processing bugs.
tests/ovn.at | 13 +
1 file changed, 13 insertions(+)
--
2.26.2
___
dev mailing list
d...@openvs
The tests "superseding ACLs with conjunction" and "ARP replies for SNAT
external ips" trigger bugs in the ovn-controller incremental processing
logic. This works around those bugs.
Signed-off-by: Ben Pfaff
---
tests/ovn.at | 13 +
1 file changed, 13 inserti
When there's no open session, we still have to free the messages that
we make but cannot send.
I'm not confident that these fix actual bugs, because it seems possible
that these code paths can only be hit when the session is nonnull.
Signed-off-by: Ben Pfaff
---
lib/ovsdb-
7;m pretty sure that I've seen it a number of times in testing.
Signed-off-by: Ben Pfaff
---
lib/jsonrpc.c | 5 -
lib/reconnect.c| 25 +++--
lib/reconnect.h| 1 +
tests/test-reconnect.c | 1 +
4 files changed, 29 insertions(+), 3 deletions(-)
ovsdb_idl_db_parse_monitor_reply() clears the IDL and parses the
received data. There's no need to do it again afterward.
Signed-off-by: Ben Pfaff
Fixes: 1b1d2e6daa56 ("ovsdb: Introduce experimental support for clustered
databases.")
---
lib/ovsdb-idl.c | 3 ---
1 file chang
ntroller's implementation of
incremental processing. I'll send out a separate OVN patch that
partly addresses that.
Ben Pfaff (4):
jsonrpc: Avoid disconnecting prematurely due to long poll intervals.
ovsdb-idl: Avoid redundant clearing and parsing of received data.
ovsdb-idl: Fix memory l
On Tue, Nov 24, 2020 at 01:20:03PM +0100, Dumitru Ceara wrote:
> On 11/24/20 1:16 PM, Dumitru Ceara wrote:
> > This series fixes 3 different issues:
> > - Patch1: memory leak of tracked old datum of reinserted orphan flows.
> > - Patch2: memory leak of tracked deleted "pure" orphan rows.
> > - Patc
On Mon, Nov 23, 2020 at 08:03:08PM +0100, Ilya Maximets wrote:
> TL;DR OVS can not use Travis CI anymore. So, we need a new CI provider.
> GitHub Actions looks like a sane replacement.
Ilya, thank you for doing all of the research here and presenting all of
it so thoroughly and clearly. It sound
On Fri, Nov 20, 2020 at 02:23:04PM -0500, Ihar Hrachyshka wrote:
> On Fri, Nov 20, 2020 at 1:16 PM Ben Pfaff wrote:
> >
> > On Fri, Nov 20, 2020 at 12:44:30PM -0500, Ihar Hrachyshka wrote:
> > > On Thu, Nov 19, 2020 at 12:19 PM Ben Pfaff wrote:
> > > >
> &
On Fri, Nov 20, 2020 at 12:44:30PM -0500, Ihar Hrachyshka wrote:
> On Thu, Nov 19, 2020 at 12:19 PM Ben Pfaff wrote:
> >
> > On Thu, Nov 19, 2020 at 08:31:44AM -0500, Ihar Hrachyshka wrote:
> > > On Thu, Nov 19, 2020 at 12:51 AM Ben Pfaff wrote:
> > > >
> &
On Fri, Nov 20, 2020 at 12:35:18PM +0530, Numan Siddique wrote:
> On Fri, Nov 20, 2020 at 8:04 AM Mark Michelson wrote:
> >
> > For the series:
> >
> > Acked-by: Mark Michelson
>
> Thanks Dumitru and Mark for the reviews. I applied this series to master.
>
> @Ben - There are changes in ovn-nort
On Fri, Nov 20, 2020 at 05:12:46AM -0800, Toms Atteka wrote:
> This optimization caused FLOW_TNL_F_UDPIF flag not to be used in
> hash calculation for geneve tunnel when revalidating flows which
> resulted in different cache hash values and incorrect behaviour.
>
> Reported-at: https://github.com/
On Thu, Nov 19, 2020 at 09:59:04AM -0800, Ben Pfaff wrote:
> On Thu, Nov 19, 2020 at 11:50:57AM +0100, Dumitru Ceara wrote:
> > On 11/19/20 5:15 AM, Ben Pfaff wrote:
> > > On Tue, Nov 17, 2020 at 05:50:20PM +0100, Dumitru Ceara wrote:
> > >> It is not correct for ovn-
On Thu, Nov 19, 2020 at 11:50:57AM +0100, Dumitru Ceara wrote:
> On 11/19/20 5:15 AM, Ben Pfaff wrote:
> > On Tue, Nov 17, 2020 at 05:50:20PM +0100, Dumitru Ceara wrote:
> >> It is not correct for ovn-controller to pass the current SB_Global.nb_cfg
> >> value to ofctrl
On Thu, Nov 19, 2020 at 02:31:04PM +0530, Numan Siddique wrote:
> Thanks for v5. This patch doesn't apply cleanly on the latest master.
>
> I'd suggest to post the patch 1 was an independent patch rebased on on
> top of latest master
>
> I think you can submit the patch 2 once the patch 1 is acce
On Thu, Nov 19, 2020 at 09:11:38AM +0100, Dumitru Ceara wrote:
> On 11/19/20 6:13 AM, Ben Pfaff wrote:
> > From: Leonid Ryzhyk
> >
> > This implementation is incremental, meaning that it only recalculates
> > what is needed for the southbound database when northboun
501 - 600 of 7535 matches
Mail list logo
