►
From YouTube: Kyma Core SIG community meeting 20190213
Description
Meeting notes: https://docs.google.com/document/d/1vWleTon7sJIk0teee4SoVezS4mR3K8TlkvXkgLJwRD8/edit#heading=h.p2ukjuifn4qw
A
A
Bachynski
will
say
some
words
about
the
release
about
seven,
then
I
will
go
with
short,
how
to
where
to
find
documentation
and
how
to
install
kima
on
the
aks.
Then
again,
toboggan
ski
with
ODOT
age
scope
with
some
highlights,
then
first
we
will
do
that.
We
will
do
first
time
the
board
preview.
A
That
means
we'll
go
to
the
Zen
hub
board
when
we
will
look
into
the
issue:
smart
with
SiC
core
label
and
let's
see
how
it
will
look
like
then
in
the
community
slot
I
met,
wanted
to
show
us
new
quality
labels,
but,
as
I
know,
is
sick,
then
probably
we
will
omit
that
and
then
we'll
have
more
time
for
questions
and
feedback.
Then
that's
does
the
plan
for
today
PB
already.
C
B
Then
yeah
so,
unfortunately,
I
cannot
show
you
the
release,
but
we
have
a
release
candidate
and
we
have
released
notes
already
prepared
and
you
can
expect
the
release
today,
even
right
after
this
call
or
tomorrow
morning,
but
yeah.
We
have
bad
timing,
so
we
miss
the
opportunity
to
present
fresh
really
is
on
during
this
meeting,
but
nothing
will
change
from
from
what
we
have
in
a
risk
candidate
one.
So
it's
it's
working
and
this
is
just
a
technical
process
to
to
take
it
out.
B
So
what
are
the
highlights
highlights
you
can
see
also
in
the
comeup
project
website.
There
is
a
pull
request
with
the
release
notes,
so
it's
not
published
yet,
but
you
can
preview,
of
course,
and
this
this
stuff,
so
I
am
showing
what
you
can
access
already
later
on.
It
will
be
visible
in
our
official,
came
a
project
I/o
page
as
a
blog
post.
So
what
are
the
highlights
of
0.7
Rudi's?
First
of
all,
we
have
assets
stores,
so
this
is
some
news
or
a
solution
for
handling
different
resources
we
use.
B
So
we
use
that,
for
example,
or
documentation
pages
health
pages.
We
use
that
for
API
documentation,
so
the
swagger
files
or
data
files,
I,
think
API,
a
descriptions
and
diagrams
etc.
So
this
is
the
content
and
we
we
have
in
keema
itself.
We
have
new
solution
to
do
that
in
the
Q&A.
This
native
way,
the
asset
store
is
ready,
the
beta
beckoned
site
and
a
spoiler
in
the
next
series.
We
will
have
also
some
UI
for
that,
so
so
we
will
see
this
new,
our
documentation
rendered
with
the
asset
store
applied.
B
Next,
we
have.
We
continued,
of
course,
work
on
with
creative,
so
our
contribution
to
creative
with
not
streaming
is
accepted,
and
we
worked
right
now
on
a
better
integration
with
with
kima's
still,
this
is
enabled
with
feature
toggles.
So
if
you
want
to
try
kima
with
clean
it,
if
you
need
to
use
fissure
toggle
but
publishing
events
are
there
in
the
next
release,
we
will
continue
that
and
probably
we
can
have
creative
enabled
by
default.
But
this
is
not
the
promise
we
will
see
if
it
will
handle
that.
But
this
is
something
that
will
happen.
B
B
We
will
have
a
long-running
cluster
with
with
tests
running
in
the
loop,
so
in
case
of
any
memory
leaks
instabilities.
We
can
discover
it
quickly
and
before
our
customers
know
that,
and-
and
the
next
topic
came
on
a
case-
I
will
skip
it
because,
okay,
we
will
talk
about
that
and
a
couple
of
smaller
improvements,
so
we
have
better
eight
auto-scaling,
API
aprox
tweaked
a
little
bit
our
resource
quotas
and
and
our
startup
scripts
are
really
simpler.
B
So
some
small
refactoring
happens,
but
you
will
see
that
improvements
in
everyday
work
with
kima
and
last
big
point
here
is
UI
api
layer
security.
So
this
is
really
something
great.
So
we
have
this
graph
QL
layer
on
top
of
kubernetes
api,
and
now
we
can
apply
all
these
arbok
rules
in
this
layer.
So
mutation
and
queries
are
checked
against,
coordinates
are
bug,
and
this
is
some
I
think
that
work
really
do
not
hit
the
ice
with
all
these
calls
from
from
UI,
and
this
is
really
nice
stuff.
B
A
The
install
kima
on
the
AKS
cluster
and
install
him
on
the
AKS
cluster
with
will
will
called
DNS
and
overall
you
can
see
that
we
also
reflect
or
the
documentation
about
the
whole
installation
that
you
can
find
really
nice
way
how
we
can
install
locally
on
the
clusters
and
with
other
things
like
update,
kima
or
reinstall
kima
documents
also
selective,
also
about
the
modular
res
motorisation,
a
lot
of
things
and
feel
free
to
to
look
into
that.
Then,
let's
go
to
the
AKS
cluster
installation.
A
We
added
some,
of
course,
prerequisites
like
we
already
need
to
have
a
domain
of
course,
then
the
some
CLI
to
the
other
aks
access
and
small
other
things,
and
more
and
more
or
less
it's
about
setting
up
some
environment
variables
at
the
beginning.
Unlike
the
mother
torii
variables
like
the
custom
name,
which
was
group
region,
then
in
that
first
face,
you
can
already
decide
if
you
want
to
install
keema
from
the
release
or
you
decide
to
go
with
even
given
commit
from
the
from
the
keema
sources.
A
But
if
you
will
set
up
that
environment
variable
all
of
the
other
steps
will
be
already
prepared
for
you
for
the
release.
Installation
and,
of
course
you
need
to
also
decide
about
the
said,
the
DNS
domain
and
sub
domain,
which
is
explained
and
more
or
less
there
are
not
big
changes
regarding
to
the
GK
installed,
GK
installation,
it's
more
or
less
some
slight
differences.
That
needs
to
be
done.
A
Then
there
is
the
whole
Dennis
setup
explained
how
to
get
TLS
certificate
and,
as
the
difference
with
the
GK
is
that
the
Adina's
still
do
not
support
the
server.
But
it's
not
supported
by
itself,
but
then
we
need
to
do
that
that
verification
manually.
It's
explained
how
to
do
that
with
that
takes
T
records
and
so
on.
And
then
it's
read
a
KS
installation.
The
cluster
itself
manage
one
then
how
to
get
cube.
A
Then
we
need
to
prepare
some
installation
configuration
files
and
same
installation
for
the
release,
installation
for
the
own
image
from
the
commit,
and
then
it's
just
deploying
a
kheema,
regardless
of
the
setup,
its
release
or
commit,
and
by
the
end
we
we
were.
We
managed
to
install
that
in
the
case
looks
quite
stable.
Everything
is
available
all
of
our
test
paths
with
that
and
the
next
documentation
is
with
the
wheeled
cart
DNS.
It's
about
xip,
I/o
usage.
A
It's
not
the
production,
let's
say
setup,
but
you
can
use
it
for
your
first
tries
with
your
aks
to
make
it
a
bit
faster,
and
you
can
also
find
how
to
you
can
allocate
required,
eyepiece,
dynamically
or
manually
and
and
so
on.
Then
you
can
just
go
through
that
and
have
it
installed
and
some
spoiler
for
the
next
release.
We
plan
to
extend
our
CI
pipelines
with
some
nightly
and
check
schema
with
on
Ike
is
with
some
tests
in
the
loop.
It's
called
stability
checker
in
in
the
kima
to
just
really
check
and
be
up-to-date.
A
B
So
after
dublin
0.7
we
have
0.8
Edmonton
and
the
the
big
things
that
we
want
to
deliver
in
the
next
release
in
four
weeks
is
moral.
Yes,
we
want
to
finish
this
clean
native
eventing
integration,
so
I
hope
that
we
will
have
it
really
without
the
teacher
toggle
available,
so
you
can
use
even
external
message
queues
instead
of
in
building
nuts.
This
is
the
work
we
will
focus
on
then
big
topic
for
us
is
for
the
next
release
is
backup
and
restore.
B
So
oh,
we
started
some
time
and
go
with
Ark
to
make
up
our
own
resources
that
kima
installs,
but
also
work
from
customers
so
from
from
the
namespaces
that
users
will
create
and
similarly
to
the
upgrades
tests,
we
want
to
have
such
test
for
for
backup,
so
we
will
really
actively
test
in
our
pipeline
if
we
can
backup
and
restore
kima
with
workload
without
the
issues.
So
this
is
the
big
focus
for
the
next
release.
B
We
continue
it
upgradability.
So
in
previous
release,
we
deliver
this
CI
plans
and
we
will
have
0.8
version
upgradeable
from
0.7.
That's
the
goal,
so
you
will
not
have
to
install
a
new
cluster
with
kima
with
new
release.
You
can
upgrade
existing
one
to
the
newest
version.
Then
we
do
a
couple
of
smaller
things.
Pk
mentioned
this
xip
iot
is
dynamic
dns
we
want
to
make
it
default,
so
this
will
simplify.
This
tri
came
up
first
time,
so
we
didn't
as
you've
seen
when
PK
presented
that
the
manual
for
for
an
installation.
B
B
Most
time-consuming
task
and
manual
tasks
and
we
want
to
get
rid
of
that
and
with
XIT
io
you
can
just
install
kima
on
on
the
cluster
and
that's
it
so
there's,
DNS
and
and
certificates
will
be
prepared
out
of
the
box.
So
this
is
one
of
the
improvements.
One
thing
more
I
forgot
to
mention
before
when
I
set
about
this
0.7,
we
migrated
to
mini
cube
0.33,
so
we
have
and
we
migrated
to
New
Era
Burnett
is
version.
We
support
right
now,
111
version
and
in
next
series.
B
B
B
Adding
better
TLS
handling
in
Hema,
we
will
also
decrease
ID
token
expiration
time
to
increase
security,
and
we
will
also
work
on
a
couple
of
security
issues
in
our
CI
in
prowl,
but
it
is
yeah,
so
a
lot
of
work
with
security,
but
with
treat
it's
really
seriously
and
we
want
to
have
really
secure
version
of
kima
in
four
weeks
yeah.
So
these
are
I,
think
big
things.
Oh,
maybe
one
more.
We
want
to
replace,
also
okay,
look
in
Hema
with
locking,
which
is
the
project
from
Ravana.
C
B
Is
really
promising
and
looks
better
than
okay,
a
lock,
which
was
which
is
an
archived
project
and
will
have
different
logging
in
the
next
race
and
also
integrated
with
UI?
So
that's
that's
important,
and
for
people
who
are
waiting
for
Oh
data
support,
we
will
have
rendering
crane
service
cataloging
console
UI
for
for
all
data
specification.
B
A
You
baby,
of
course,
about
the
blog
posts.
Always
they
are
put
into
the
Akuma
projectile
when
you
can
find
all
of
them
about
their
leases
about
the
highlights
for
the
next
releases,
and
you
can
see
exactly
what
what
PB
said
today.
Ok,
then,
let's
go
with
the
next
point,
which
we'll
do
first
time
and
that's
about
our
board
review.
Then
let
me
go.
A
Ok
and
let's,
let's
see
how
many
things
are
marked
really
with
the
Sakura
label
and
on
that
board
we
see
the
issues
and
pull
requests
and
maybe
for
today
we
will
just
go
with
the
review
and
progress
things
more
or
less.
They
are
proposals
and
where
you
can
look
into
and
if
they
are
in
the
review.
There
is
a
time
also
to
leave
the
comments
or
some
other
ideas,
and
but
they
are
not
only
proposals,
but
it's
the
majority
of
them
and
the
first
one.
A
It's
about
the
restrictor
ization
of
the
documentation
that
we
are
going
further
and
service
brokers,
view
will
be,
will
be
gone
in
the
documentation
and
we'll
have
a
nice
separation
that
the
brokers
like
application
broker
will
be
a
real
aunt
already
in
the
application.
Connector
and
her
broker
will
have
the
separate
route
and
we
decided
to
move
the
brokers
in
the
right
components
where
it
really
belongs
to,
but
about
the
overall
service
brokers.
A
A
A
In
kima
we
have
the
mono
repo,
where
we
have
a
code
of
the
components
and
as
well
the
charts
modules
where
we
are
using
images
from
those
components
and
right
now
we
need
to
create
two
pull
requests:
one
for
merging
the
component
code
to
the
master
and
then
bump
the
version
in
the
chart,
and
that
proposal
from
Christian
is
more
or
less
about
improving.
That
procedure
really
make
it
easier
the
release
process
by
really
splitting
the
repository.
You
can
look
into
that
idea
and
leave
your
comment.
A
On
the
other
hand,
if
the
in
the
to-do
list,
there
is
a
second
proposal
about
proper,
providing
proposal,
increment
PUC,
that
we
are
able
to
bump
the
version
in
a
1/2
request
and
also
then
it
will
be
more
or
less
the
work
will
be
like
we
will
do
the
release
every
day.
That
means
all
of
the
changes
in
the
master
of
the
components
they
will
be
already
reflected
in
the
charts,
and
you
can
already.
A
The
proposal
has
not
started
its
planned
for
the
let's
say
next
upcoming
weeks
to
work
on
the
proposal,
but
you
can
early
read
the
description
and
you
can
compare
those
two
proposals
how
we
can
work
about,
and
then
we
have
some
proposals
about
consolidation
conventions
for
the
test,
folder
for
the
component
folder.
If
you
go
in,
you
can
see
that
there
is
the
proposal
itself
and
then
some
issues
with
exact.
A
Like
there
was
already
the
analyzes
done
about
all
of
our
components
and
how
we
we
want
to
consolidate
really
the
readme
is
how
the
things
are
done
in
those
components
about
the
structure
and
so
on.
Then
you
can
also
look
at
least,
and
you
can
even
have
a
nice
overview,
how
many
and
where
they
are
placed,
how
many
commands
we
have
and
where
the
power,
whether
they
are
placed
in
the
chemo
and
all
of
those
things,
are
done
for
really
simplify
and
unified.
A
C
A
For
example,
one
reason
is
to
have
some
selective
testing
in
and
other
capabilities
like
running
tests
in
parallel
and
so
on.
Witch
hunt
just
didn't
support
because
right
now
it
supports
parallel
testing,
but
it's
still
quite
basic
parallel
functionality.
We
want
to
go
even
further
with
that.
On
the
other
side,
you
we
also
want
to,
as
I
said,
that
proposal
about
that
one
to
request.
A
We
decided
to
skip
it
for
the
next
release,
but
we
already
want
to
implement
the
job
that
we
will
see
in
the
pro
that,
on
the
master,
some
components
are
not
there.
Version
are
versions,
are
not
used
in
the
charts
and
we
will
see
that
and
with
that
ticket.
Also,
there
is
some
things
about
providing
some
conventions
about
where
we
are
braiding,
those
verses
and
so
on
that
it
will
help
help
us
to
also
improve
the
release
process
that
it
will
be
automatic
sooner
or
later
without
developer.
A
You
know
developer,
we
don't
need
to
bump
the
versions,
it
will
be
done
automatically
by
the
job
itself
and
so
on
then
I
guess,
that's
it
from
from
I
said
right
now
we
are
doing
that
first
time.
Let's
see
how
it
looked
like
next
time,
but
it's
a
nice
place
we
to
see
about
the
more
general
concepts
we
want
to
implement
and
provide
into
the
keema
exactly
issues
and
requests
marked
with
the
Sakura
label.
B
I
have
a
question
about
this
two
different
proposal:
how
to
how
to
work
with
with
kima.
So
do
we
want
to
have
some
due
date
for
that?
We
have
written
really
in
review
this
proposal
from
Christian
and
we
have
new
proposal,
as
you
said,
somehow,
to
maybe
create
some
due
date
and
have
discussion
on
slack
or
dedicated
meeting
for
that.
Yes,.
A
A
Probably
right
like
like
that
and
I
am
in
contact
with
them
and
also
I
think
that
I
will
try
to
invite
people
assigned
to
those
things
in
the
review
or
to
do
from
those
tickets
to
the
C
core
that
we
can
even
discuss
some
of
those
topics
during
that
meeting.
That
will
have
better
overview
about
that.
A
A
B
B
A
A
A
And
we
see
that
we
have
three
of
them
right
now
and
as
they
are
named
right,
the
observability
it's
more
or
less
about
I,
guess,
monitoring
and
see.
What's
going
on
in
the
in
the
system,
the
productivity,
it's
the
quality
about
how
to
make
easier
to
me
to
develop
in
in
kimbap
and
to
reduce
the
time
of
the
tests
and
other
aspects
to
make
it
easier
and
and
funny
and
have
some
fun.
Maybe.
B
A
A
A
Okay,
maybe
Chris,
because
I've
seen
some
communication
because
fukushi-kun
is
key,
is
absent
this
week,
but
he
sent
to
me
some
information
that
you
are
interested
to
play
with
kima
and
also
contribute.
As
you
said,
and
mukesh
mentioned
me,
that
may
be
the
first
good
thing
is
to
work
on
examples,
and
how
do
you
see
that
already
do.
C
You
have
some
yes,
so
at
least
at
least
we
know
that
we
want
to
work
on
examples
and
also
some
how
applicable
examples,
and
so
in
a
way
that
they
can
be
consumed
or
insult
or
whatever
right,
but
we
don't
have
any
concrete
plans
how
to
do
that.
I
would
if
we
I
mean
I,
don't
to
be
honest,
I,
don't
really
know
we
actually
start
right.
So
should
I
just
get
gather
some
information
and
create
a
proposal
or
that's.
A
What
I
can
suggest
yeah
we
have
that
examples
repository
in
under
the
keema
my
project
and
right
now
we
see
that's
the
structure
of
them
could
be
better
and
because
they
are
granulate
granulated
in
two
different
ways
that
one
of
them
are
really
big.
The
second
one
is
very
small
and
they
are
not
connected
to
each
other
at
all.
Maybe
that
would
be
also
the
proposal
about
that
how
to
organize
those
examples
what
they
should
represent.
A
Maybe
there
should
be
even
some
order
of
the
examples
that
we
can
treat
it
at
them
as
a
tutorial
tutorial.
Somehow
not
the
separated
examples
not
connected
at
all,
and
the
second
thing
is
about.
You
can
also
investigate
the
bundles
concept,
where
some
of
the
examples
or
bigger
showcases,
for
example,
that
you
want
to
install
into
the
cluster
without
going
to
some
github
checking
out
and
so
on,
but
just
register
the
repository
with
bundles
to
the
kima.
Then
you
see
them
as
a
service
classes,
and
then
you
can
install
them.
A
That's
let's
say
this:
the
another
way
that
let's
say
that
could
be
used.
I
see
it
more
for
the
release
showcases
as
a
bundles
to
provide
the
solutions
and
then
about
where
to
put
such
a
things.
Please
contact
me
directly
because
right
now
we
have
abundance
repository
in
the
keema
I'm
thinking
about
doing
a
fork
in
the
keema
incubator,
with
the
bundles
that
we
can
put
some
bundles
that
let's
say
right
now.
We
don't
want
to
have
them
in
the
release,
but
you
can.
C
That,
okay,
that
takes
total
sense
to
me
so
I,
will,
as
first
step
take,
for
example,
depository
try
to
make
it
more
accessible
or
more
and
more
consumable
and
also
check
the
bundles,
because
I
mean
we
are
interested
in
the
Roxtec
bundle.
I
think
that
would
be
to
something
that
could
be
done
as
a
first
step
and.
A
Let's
say
how
you
see
that
how
they
could
be
a
restrictor
and
if
you
add
the
new
ones,
that
let's
say
we
need
to
keep
the
rules
of
adding
the
examples
and
to
have
such
a
proposal
will
be
also
nice
right,
yeah,
even
for
you
that
it
will
get
familiar
more
with
how
it
works,
then
we
should
try
it
right.
Yeah,
yeah,
I,
agree.
Okay,
so
I
get
in
touch
with
you.
It
Justin's
like
yes
and
then
I
already
already
familiar
with
the
channels
right.
We
have
come
up
with.