►
From YouTube: Keptn Refinement Meeting - Jun 20, 2023
Description
Meeting notes: https://docs.google.com/document/d/10Fig1eYFZ9iQFSYWkz0c4eTwzgJiPtQI5IsczbvLsuE/edit
Learn more: https://keptn.sh
Get started with tutorials: https://tutorials.keptn.sh
Join us in Slack: https://slack.keptn.sh
Star us on Github: https://github.com/keptn/keptn
Follow us on Twitter: https://twitter.com/keptnProject
Sign up to our newsletter: https://bit.ly/KeptnNews
A
Everything:
okay
now
the
recording
should
have
been
started.
So,
let's
kick
it
off.
We
have
two
tickets
only
to
refine.
Let's
start
from
this
one,
since
the
Community
member
already
did
an
amazing
job
honestly
kudos
to
Batman,
he
researched
into
all
the
possible
different
place,
images
that
we
could
use
for
kltail.
A
A
B
We
are
kind
of
free
to
choose
whatever
we
want.
If
we
want
to
change.
B
The
only
thing
is
that
UV
is
the
only
class
of
images
in
this
case
that
kind
of
has
some
experience
so
to
say
with
modern
abilities,
because
it's
based
on
Red
Hat
Linux.
So
it
gets
the
same
treatment
basically
in
terms
of
security
updates
and,
of
course,
slas
behind
Red,
Hat,
Linux
and
stuff.
The
kind
of
safeguard
that
for
a
UB
as
well
and
then
chain
guard
kinda
also
has
slas.
But
there's
no
no
info
available,
slash
less
background
info
or
experience.
C
B
Yeah
KO
also
uses
strangled
images
just
with
the
chaos
UI
yeah,
the
Community
member
kind
of
recommended
to
stay
on
what
we
are
right
now,
but
of
course
we
can
discuss
that
foreign.
A
C
B
D
A
I
need
to
log
in
so
I.
Don't
have
any
image
locally
I
just
clean
up
my
cache,
but
should
be
around.
C
B
A
A
Not
image.
E
Hey,
obviously,
is
there
what
we're
using?
Is
there
a
reason
that
we'd
like
something
else.
A
I
couldn't
hear
the
connection
was
interrupting.
Interrupt
I
couldn't
follow
what
you
said.
Sorry
Mike.
Could
you
please
repeat.
E
Sorry
and
just
got
like
yeah
I,
just
don't
know
around
this
and
why
we're
thinking
about
switches
problem
with
this
winner
or
how
does
what
we're
using
I.
D
A
That
guarantees
security,
vulnerability
fixes
so.
A
C
A
C
B
It
basically,
you
get
a
14
day,
critical,
cve
fixed
time
or,
if
you
have,
if
you
pay
more,
you
get
seven
days.
A
A
A
So
the
next
kit,
the
ticket,
is
about
some
migration
guide,
a
bit
of
context
to
bring
everyone
up
to
speed
so
captain.
Now
we
have
two
different
versions:
Captain
B1
and
klt,
and
the
captain
announcement
proposal
98.
If
I,
don't
recall
the
wrong.
A
The
ID
is
to
use
klt
as
a
image
of
Captain
and
not
anymore,
Captain
V1,
because
KD
is
more
kubernetes
native,
but
we
need
to
bring
everyone
from
Captain
1
to
klt,
eventually,
because
this
is
the
main,
no
facial
captain,
and
for
that
we
need
to
provide
a
way
for
users
to
migrate.
The
current
use
cases
that
they're
running
with
captivity
one
how
they
can
run
the
same
exact
use
cases
with
klt
and
I.
E
It
will
grow
over
time
but
we'll
put
down.
Well
what
would
you
do?
First,
if
you
were
starting
to
do
that
and
I
think
you
know,
you
know
it's
like
okay,
get
yourself
a
kubernetes
environment,
that's
deploying
your
software
with
Argo
or
flux,
or
something
and
then,
but
do
you
do
tasks
first
or
evaluations
first
or
and
if
you've
got
40
tasks
you
throw
them
all
in
at
once
or
you
do
them
one
at
a
time.
E
You
know
this
sort
of
sort
of
Soft
Stuff
be
sort
of
a
sort
of
a
checklist
of
roughly
in
order
and
reasons
why
you
know
you
might
want
to
do
one
thing
before
you
did
the
other
and
links
to
where
there's
more
information
about
them.
It
would
be
one
of
those
things
that
a
lot
of
people
wouldn't
read,
but
some
people
would
okay.
A
Just
more
more
in
depth,
step
by
step
right.
E
A
We
should
focus
on
the
continuous
delivery
use
case
because
if
you
look
at
two,
the
captains
like
this
is
still
quite
easiest
and
popular
having
multi-stage
delivery
of
your
application,
Dev
hardening
ordering
two
hardening
uis,
hardening,
Central,
Europe
and
finally,
production
to
everyone
the
same
but
with
klt,
then
quality
Gates.
The
other
use
case
of
Captain
is
to
prevent
bad
software
to
go
into
production
and
how
we
can
migrate.
A
Then,
the
quality
gate,
solo
files
with
klt
that
we
now
called
evaluations
and
then
another
use
case
quite
popular-
is
to
get
another
tool
into
your
pipeline
thanks
to
Jess.
So
many
people
wants
to
run
their
own
different
tools
at
any
stages
of
the
delivery
and
they
are
using
that
they're
doing
that.
Sorry
with
job
executor
service,
so
a
way
to
replace
this,
and
this
can
be
done
thanks
to
the
klt
tasks,
especially
now
that
we
have
the
container
runtime.
A
This
migration
should
be
straightforward
and
then
the
last
point,
which
is
not
widely
using
Captain,
but
still
some
people
have
experimented
with
that
is
continuous
remediation
or
day
two
operations
also
called
where
we
do.
Maybe
we
can
refer
to
the
captain,
metric
and
watch
for
them
to
react
to
changes
and
provide
something.
A
E
A
Try
that
automatically
and
everyone
failed
because
of
a
state
in
kubernetes
it's
hard,
and
you
cannot
roll
back
or
automate
that
there
is
always
the
need
of
a
human
to
really
look
into
so
this
we
might
even
not
want
to
provide
any
migration
guide.
E
C
E
I
think
part
of
it
as
I.
Look
at
the
duck.
You
know
I'm
still
figuring
all
the
stuff,
reading
the
documentation
for
remediations
it
was
a
little
complicated
and-
and
there
certainly
is
value
in
having
it
out
there
and
saying
hey
this,
and
this
is
going
wrong
and
notifying
me
I-
don't
have
to
get
up
every
two
hours
and
read
through
all
the
logs
myself.
E
It's
I
mean
yeah
for
that's
a
math
Advantage,
we'll
put
more
energy
into
converting
your
Jes
stuff
into
exactly.
A
C
E
A
You
have
a
lot
of
your.
E
E
Okay,
going
to
chat.
A
Also
all
of
these
might
be
very
good,
first
good,
first
issues,
since
this
means
you
need
to
play
with
V1
and
klt,
so
did
understand
the
differences
and
support
other
people
understanding
these
differences.
A
There
is
a
prerequisite
before
we
can
go
ahead
and
work
on
this
ticket,
which
is
the
release
of
0.8,
because
we
need
0.8.
All
tickets
have
been
implemented.
A
Now
we
need
just
to
polish
it
in
the
documentation
phase.
There
are
still
one
PR
that
we
need
to
merge
and
make
I
will
have
a
look
after
this
meeting.
Then
we
can
go
over
the
last
PR's
merged
make
sure
that
the
titles
are
correct,
otherwise
change
them
and
polish,
the
change
log
and
finally
triggered
official
release.
This
will
happen
tomorrow,
right
before
our
community
meeting.
We
should
have.
C
A
A
Your
connection
is
still
dropping,
we
don't
have
experience
migrating,
but
I'm
thinking,
we
start
putting
information
together
and
we
will
grow
over
time.
Yes,
I
agree,
let's
start
somewhere
and
then
learn
from
that.
A
For
the
last
point
that
you
said
that
this
one
yes
I
change
it,
because
maybe
I
miscon
spoke
with
you,
it
should
be
not
really
my
Great
Captain
B1
Helm
charts
to
kill
T,
but
is
how
people
can
deploy
their
home
chart
for
Captain,
but
instead
of
doing
that
with
Captain
V1
have
multi-stage
delivery
or
what
we
call.
The
continuous
delivery
with
klt
I
will
adopt
this
here.
I.
D
A
A
So
we
can
really
migrate
all
the
users
promising
the
same
feature
sets
needs
to
be
simultaneously
working
on
how
to
implement
klt.
We
have
a
start,
and
much
of
the
info
is
the
same.
Migration
guide
should
reference
basic
how
to
implement
klt?
A
E
A
E
E
A
A
Otherwise,
I
know
that
you
Joffrey
prepare
up
the
r.
A
Where
is
it,
we
are
together
with
worksheet
working
on
blocking
that
until
we
do
the
release,
so
we
can
unlock
you
tomorrow.
Basically,.
D
So,
based
on
what
you
are
saying,
I'm
currently
working
on
that
issue,
so
should
I
wait
on
the
issue
until
am
I
being
blocked
on
the
issue.
I
just
want
to
go.
We
made
something
clear.
I
know
I'm,
not
too
clear
what
you
just
said
now.
D
I
this
was
the
injection
of
the
decoder,
and
we
agree.
This
is
something
which
is
transparent
to
users,
so
that
one
was
merged.
A
Already,
okay,
perfect
true,
true
I
forgot
true
true,
true,
the
only
follow
up
the
the
bump
of
the
content
of
the
oh,
it's.
E
A
Or
bring
everyone
up
to
speed
about
why
we
are
doing
that
is
because
having
the
controller
runtime
bumped
here.
Basically,
this
dependency
requires
a
lot
of
breaking
changes,
it's
transparent
to
the
end
users
beside
the
fact
of
which
kubernetes
version
we
do
support
pumping
this
runtime
means
that
we
need
to
drop
support
for
version
23.
B
B
A
E
A
But
your
connection
bad,
but
we
need
to
adapt
also
yeah.