►
From YouTube: [SIG-Network] Ingress NGINX meeting 20210608
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
A
A
A
So
any
any
violation
should
be
reported
to
the
sig
network
chairs
or
it
can
be
reported
directly
to
me
or
if
it's
something
that
I'd
say
and
you
want
to
complain,
you
can,
you
can
reach
code
of
conduct
on
slack
or
you
can
reach
james
as
well,
and
let
me
share
my
screen
with
you
before:
starting
do
yeah
cool
so
do
do
we
have
anyone
new
in
this
meeting
that
want
to
present
yourself.
A
Okay,
so
I
don't
think
we
have
anyone
new
or
we
do,
but
okay
cool.
So
today,
in
today's
meeting
we
have
two
actually
three
items
in
the
agenda.
A
A
So
before
going
to
the
issue
triage,
I
I
wanted
to
go
through
the
open
topics.
First,
so
as
soon
as
we
finish
the
open
topics,
we
might
have
some
some
spare
time
for
the
issue
triage.
If
this
sounds
good
for
all
of
you,
because
the
open
topics
from
my
from
my
side,
they
are
pretty
fast.
So
the
the
first
one
is
the
updates
on
on
122
build
everything
is
almost
working
fine.
I
still
need
to
figure
out
just
why
the
end-to-end
test
they
are
failing
for
the
ignition
controllers.
A
It's
mostly
a
lack
of
time
to
to
figure
out
other
than
a
real
issue,
but
I
I
wanna
just
just
add
that
pr
at
that
pr
update
the
pr.
So
if
someone
want
to
take
a
look
into
that
pr,
it's
extra
extra
large,
some
more
eyeballs
would
be
really
great
before
we
we
actually
merged
that
I
was
discussing
with
with
james.
If
you
are
driving-
and
you
want
me
to
to
to
to
take
the
notes-
don't
don't
don't
don't
hit
the
car?
Please.
A
Okay,
cool
so
yeah,
so
I
I've
been
discussing
with
carlos
and
and
other
folks,
but
I
I
think
it's
a
good
for
all
of
you
to
know
that
my
my
intention
is
actually
to
create
a
new
branch
which
is,
which
is
we
don't
have
the
name
yet,
but
probably
the
v0
dot
x,
or
something
like
that.
That
represents
the
the
pre
v1
api.
A
So
we
can
create
that
maintenance
branch
for
any
any
new
bug
fixes
related
to
that
versions
of
of
ingress
in
in
that
window
of
six
months
of
maintainings
and
create
a
new
main
branch
for
the
v1
0,
which
is
going
to
be
the
proper
master,
slash
main
branch
and
that
any
new
feature
or
any
new
bug
fixes
goes
through
that
one.
But
we
should
probably
hit
first
code
freeze
or
something
like
that,
so
we
can
stabilize
the
v0.x
for
the
next
months
like
that
card
and
other
stuffs
and
yeah.
A
So
I'm
gonna
see
with
carlos
how?
How
can
we
proceed
with
that?
But
my
idea
is
that
we
have
at
least
those
two
branches
after
we
can
get
rid
of
the
the
kiwi,
the
backlog
of
the
the
prs
that
we
that
we
have
in
our
in
the
ripple
and
then
and
then
do
a
code
freeze
into
that
and
say:
hey
yeah.
We
just
accept
now
bug
fixes.
A
So
I
would
ask
for
you,
folks,
if
you
have
some
time
to
review
the
prs
in
the
in
the
qe,
mainly
the
bug
the
bug
fixes
one
I
I
would
really
appreciate
that
so
take
a
look
into
the
into
the
prs
and
if
you
see
something
wrong
or
if
you
think
that
it
needs
some
review,
please
add
your
review
for
for
for
folks,
provide
them
feedback
and
once
you
think
it's
ready
to
be
again
reviewed
or
approved.
A
Just
being
me
on
slack,
and
I
can
put
that
on
my
my
backlog
or
maybe
elvin
as
well.
Okay,
I
have
also
added
into
our
repo
the
label
area,
slash
helm
because,
as
far
as
I
saw,
we
have
a
lot
of
pr's
from
helm,
not
not
from
the
main
go
code,
so
I
am
not
really
good
at
help.
So
if
anybody
feels
comfortable
reviewing
prs
of
helm,
please
go
ahead
and
also,
if
you
think
hey,
I
can
be
an
approver
of
those
helm.
A
Shards
changes
just
let
us
know,
because
we
we
need
someone
taking
care
of
those
of
the
the
helm,
the
helm,
shards
prs
as
well.
A
I
think
this
is
the
main
topic,
the
other
one
is
we
have
this
issue
with
the
cordon
after
the
release
0.43,
we
have
released
a
new
nginx
ingress
controller
in
two
days
ago,
with
a
new
enginex
version,
and,
and
probably
there
is
a
new-
must
mostly
alpine
wesley
library
updated
as
well,
but
I
think
we
need
some
testing
right
now
to
see
if
those
card
update
still
happens
on
the
ve120
or
not-
and
I
wanna
hear
about
you
folks,
I'm
thinking
if
we
should
keep
alpine
as
the
main
as
the
main
distribution
that
we
use
in
ingress
in
gynex
or
if
we,
if
we
need
to
to
think
on
the
v10
about
having
like
two
the
same
way
they
do
like
with
python
and
the
other
containers.
A
If
we
should
have
like
a
an
alpine
version,
an
adapted
version
and
build
both
of
them,
so
folks
can
decide
if
they
want
to
have
alpine
or
debian
or
if
we
should
just
go
to
deviant,
but
just
be
aware
that
this
is
gonna.
Add
for
us,
some
more
needs
of
tests
and
some
more
effort
to
keep
two
at
least
like
two
new
container
images
and
james
saw
that
this
is
pretty
painful.
To
maintain
graduates
to
create
the
base
image.
B
Yeah,
the
basic
hitch,
I
mean
the
first
one
we
were
troubleshooting.
It
took
like
three
hours
before
we
realized
it's
built
was
stalled
out
and
then
on
average
it
takes
about
an
hour
and
a
half
so
moving
to
a
new
distro
yeah,
it's
gonna
take
a
lot
more
testing
as
well.
So
it's
definitely
a
discussion
point
if
the
old
reasons
for
being
on
alpine
are
still
relevant.
Well,
if
we
can
address
those,
it's
definitely
a
conversation
that
needs
to
be
had
yeah
yeah
go
ahead.
I
think
it's
maybe
like.
A
Nope
yeah,
we
we
but
yeah.
We
we
we
for
sure
we
need
to.
We
need
to
discuss
about
that
and
what
what
would
make
us
go
to
debian
or
not
like
if,
if
the
image
size
is
just
the
the
point
of
decision
between
alpine
and
debian
and
the
effort
doesn't
count
like
we
have
this,
like
the
difference
of
of
size
is
like
10
megabytes,
we
should
probably
go
to
debian
and
keep
this
this
pain
point
away
from
us
right.
A
We
got
some
some
metallic
yeah,
but
I
I
I
could
I
could
get
it
like,
but
yeah
probably
we
should.
We
should
at
least
move
with
the
v122
thing
and
as
soon
as
we
we
got
this
stabilization,
we
can.
We
can
fake
another
new
thing
to
get
this
stabilization
on
that
and
stabilization
again
right
otherwise
be
fun.
A
Yeah,
okay,
so
the
the
third
item
in
the
agenda
is
the
support
for
s39390x,
which
is
I
just
wanted
to
hear
from
the
community.
If
we
have
people
using
that
other
than
ib
ibm.
A
A
I
think
that
from
a
community
perspective,
they
want
us
to
keep
that
because
kubernetes
supports
the
s390x
architecture,
but
this
is
this
was
mostly
raised
because
on
our
last
release
curve,
it
was
pretty
painful
to
wait
like
out
those
times
that
the
es390x
failed
and
rerun
everything
again
and
then
wait
and
and
see
that
yeah,
it's
just
a
matter
of
re-running
things.
A
So
do
you
want
to
review?
Do
you
want
to
make
some
issues
react
today,
or
do
you
want
to
make
some
pull
requests?
Three
years
today?
Folks,
it's
a
like
yeah,
this
pull
request.
Should
we
should
bring
some
attention
to
this?
I
I
am
up
to
doing
that
right
now
as
well,
and
maybe
a
review
one
or
two
pr's.
A
So
anyone
wants
you
saying
I
I
see
that
you
you've
removed
yourself
from
the
mute.
C
A
Okay,
cool
so
take
a
look
into
the
the
ones
with
with
the
area,
slash
helm,
labels
and
while
we
don't
have
the
like,
the
the
approval
thing,
just
keeping
me
on
on
slack
and
say:
hey
I've
reviewed
this
one.
You
can
approve
this
one.
I
have
you
this
one
you
can.
Can
I
prove
this
one,
because
I
guess
that
from
our
kiwi
I
we
have
at
least
area
slash.
A
D
A
A
B
One
for
you,
too
72-14
it's
a
documentation,
update,
72-14,
yeah
72-14.
B
This
is
trying
to
make
it
a
little
easier
so
that
people
can
see
at
the
high
level
what
versions
of
nginx
alpine-
and
I
think,
a
couple
other
things
in
there
I
was
thinking.
Maybe
we
could
map
it
to
the
helm
charts
as
well,
but
I
didn't
really
have
to
find
an
easy
way
to
map
that
out
so
but
yeah,
just
just
an
update
to
the
readme
in
the
document.
A
I
I
have
one
of
that
that
they
are
bringing
me
about
mod
security.
I
just
can't
remember
which
one
is
this,
but
it
seems
pretty
important.
D
D
A
A
And
it's
my
my
my
safari
screen,
then,
can
you
see
my
screen?
Oh
yeah.
A
A
Soon
and
we'll
review
this
this
weekend,
so
if
someone
is
aware
of
the
behavior
of
mod
security
and
wanna
review
this
one,
please
let
me
know
otherwise,
I'm
I'm
gonna
review
this
one,
probably
during
this
this
afternoon
or
during
the
weekend,
yeah.
Okay,
any
other
prs
that
you
want
to
take
a
look
right
now.
Folks,.
A
We
have
this
one,
which
is
a
helmet
as
well.
I
think
so,
yeah,
okay,
so
moving
forward
to
the
issue
triage.
A
We
have.
We
really
need
that
needs
triage
both,
but
we
have
11
pages
of
issues
so
james.
How
have
you
been
doing
the
those
issue
triage
here
book?
First.
B
The
two
two
two
meetings,
guys
just
I
was
looking
at
just
going
through
the
older
ones.
We
can
probably
do
a
lot
of
that
offline,
but
I
think
with
the
the
segmentation
we
should
at
least
be
asking
people
to
upgrade
to
43
and
test
and
see
if
it's
still
relevant.
We
just
need
to
keep
going
through.
A
lot
of
these
tickets
are
two
three
years
old
and
they're
using
much
older
versions.
B
A
D
B
So
I
got
a
question
about
the
there's
one
that
was
recently
opened
and
I
think
it's
72
20..
B
B
So
I
don't
know
if
they
should
just
close
that
one
out
or
not
trying
to
find
where
that
number
was
it's
just
little
things
like
that.
It's
like!
Well,
I
think
it's,
I
think
it's
responding
appropriately
because
you're
sending
a
lowercase
git
and
that's
not
what
you
should
be
sending.
I
I
don't
know.
A
Yeah,
I
think
that
so
so
in
this
case,
when,
when
we
have
like
support
ones
and
those
those
ones,
we
can
simply
maybe
from
the
priority
or
like
don't
put
a
priority
on
them.
A
A
B
B
I
I
don't
think,
there's
anything
for
us
to
fix
in
this
issue,
so
I
guess
just
yeah,
like
he
said
going
through
all
of
these
issues
and
figuring
out
juan.
Are
they
still
relevant?
Are
they
using
older
versions
and
two?
Are
these
support
requests
versus
actual?
Like
looking
for
the
triage,
I
think
that's
the
biggest
ask
that
we
just
need
everyone's
help
from
is
going
through
and
just
helping
triage
of
these.
D
B
D
B
B
A
And
the
other
one
like,
if
you
send
an
accept
and
it's
it's
gonna
answer
like
the
the
page,
regardless
of
the
accept
either
that
it
gives
you
sending
right.
So
it's
a
yeah,
okay,
so,
okay,
we
have
30
more
minutes.
Let's
take
another
look
into
the
other
ones,
can't
use
ingredients
with
only
that
type
exact.
A
Okay,
so
long
is
working
on
this
one,
I'm
gonna
do
what
we
have
accepted.
A
A
But
yeah
this
seems
like
this
seems
like
an
issue
actually
with
the
controller,
with
the
admission
like
hook
right.
It's
not
it's
not
an
issue
with
with
the
nginx
controller,
but
it
is
an
issue
with
with
the
admission
like
hook
so
and
he's
and
the
user
is
using.
C
A
A
recent
client
server
version
of
of
kubernetes
and
also
you
know.
B
A
B
A
Yeah
no,
but
you
know
in
this
case
specifically
because
it's
a
invalid
memory
address
it.
It's
a
probably
a
pointer,
that's
being
referenced
without
existing,
so
it's
it's
probably
trying
to
use
a
pointer.
It's
not
it's!
Not
it's!
Not!
That
coordinated
like
like
we've
been
seen
or
like
a
memory,
a
lot
of
memory
or
something
like
that,
but
it
seems
it
really
seems
like
a
like
a
something
some
bad
validation
into
pointers
here.
D
A
Go
so
the
error
is
in
here:
it
is
in
controller
controller,
dot,
com,
transmission,
internal,
let's
see.
A
A
Yeah
he's
seriously
over
here
from
the
from
the
server,
but
it's
it's
already
happening
this
new
pointer
here,
and
so
I
need
to
to
figure
out
why
wha,
what
what
is?
Actually
this
try
to.
D
A
Yeah,
so
here
it
is
so
it's
it's
it's
it's
iterated
through
this
server.locations.
So
this
is
a
point
there.
So
if
this
this
pointer
is
new-
and
it's
not
very
fine
here,
so
we
have
this-
we
are
receiving
the
server,
but
if
there
is
no
location,
so
the
pointer
is
is
no
and
it
panics
here.
So
we
need
to
figure
out
what
exactly
is
a
and
it's
it's
his
passing
and
if,
if
you
take
a
look
into
this
like
there
is
the.
A
Yeah,
so
it
might
be
not
adding
this
as
a
because
this
this
server,
that
location
is
an
internal
structure
from
ingress
and
not
from
not
from
from
it's
from
from
ingress
in
giants,
are
not
from
from
ingredients
objects.
So
in
some
moment
this
might
be
being
like
converted
or
checked,
and
it's
a
it's.
It's
not
adding.
This
is
a
point
there,
so
it's
panicking.
So
we
can.
We
can
take
a
look
into
this,
but
yeah.
D
A
A
B
A
really
good
point
from
a
meeting
perspective
should
we
probably
look
at
maybe
alternating
these
from
a
time
perspective.
B
I
don't
know
what
time
would
be
a
little
bit
more.
We've
got
two
or
three
developers
in
asia,
pacific
time
zone.
So
I
don't
know
if
we
want
to
alternate
these
because
I
think
some
of
the
other
I
was
looking
at
some
of
the
other
meetings.
They
have
two
separate
meetings,
one
that's
more
asia,
pacific
friendly
ema
versus
us
friendly.
So
I
don't
know
if
there's
a
if
we
should,
you
know
to
get
more
like
I
know
long
hasn't
been
able
to
join
these,
and
then
I
was
saying
it's
pretty
late
for
him.
Yeah.
B
A
It's
I
am
I'm
okay
with
that,
like
I
have
this,
the
same.
The
same
thing
happens
at
my
my
my
work.
Also,
so
it's
I'm.
I
am
really
okay
with
that
we
can.
We
can
probably
raise
this
on
slack
and
see
hey
folks
if
we
do
some
asia
pacific,
more
user-friendly
who,
who
is
gonna,
join
and
what's
the
best
time
like
if
we
might
have
this
on
tuesday,
but
it's
unlike
it's
8
a.m:
asia,
pacific
time
which
is
10
pm
for
me
or
maybe
earlier,
but
earlier.
B
A
A
Yeah,
I
don't
have
anything
else,
okay,
so
I
guess
we
can
then
ain't
for
today
and
then
go
go
to
sleep.
Please
we
don't
want
you
to
get
fired
because
of
us
and
see
you
folks.
Thank
you
for
joining
us
and
james.
If
let
me,
let
me
know
if
you
remember
anything
else,
otherwise
I
can.
I
can
also
ask
for
folks
in
slack
if
they
prefer
some,
if
they,
if
we
do
some
more
more
asia-pacific
friendly,
if
more
of
them
would
join
us.
Yeah.
B
Yeah
yeah,
the
only
other
thing
is
that
can
you
I
don't
know
who
we
need
to
ping
to
get
our
blog
posts
reviewed
I'll,
put
it
in
the
notes,
but
I
think
we
need
to
get
that
out
there
so
that
people
can
start
seeing
that.
A
Yeah
go
to
seek
dogs
and
ping
team
team
ballister.