►
From YouTube: Keptn Developer Meeting - Oct 20, 2022
Description
Meeting notes: https://docs.google.com/document/d/1y7a6uaN8fwFJ7IRnvtxSfgz-OGFq6u7bKN6F7NDxKPg/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
Hi
everyone
welcome
to
the
captain
developer
meeting
and
just
in
advance
sorry
for
the
background.
Noise
I
have
a
second
contributor
participating
in
the
meeting
yeah.
So
today
we
have
a
few
items
now
agenda.
A
Yeah,
so
we
have
a
proposal
to
move
Capital
life
cycle
controls
and
we
have
released
a
0.3
zero
for
that
and
also
there
is
a
graph
on
a
dashboard
for
accessibility
for
the
release.
I
also
wanted
to
discuss
Captain
1.0
timeline
quickly.
For
those
who
will
be
working
in
the
chat.
Do
we
have
any
other
topics
we
would
like
to
discuss
today.
B
A
Mac
I
thought
we
wanted
to
discuss
the
plan
for
Captain
website
bulk
updates.
A
A
C
Actually,
here,
I'm
gonna
add
one
more
thing:
that'll.
A
Okay
Germany,
would
you
like
to
take
care
of
these
topics.
B
Yes,
I
just
opened
a
few
minutes
ago,
a
proposal
to
move
the
lifecycle
controller
from
the
sandbox
organization
into
the
main
Captain
organization,
because
I
think
now
it's
time
and
by
the
comments
of
the
people.
I
think
there
are
huge
awesome
sentiment
on
doing
this
change,
because
lifecycle
controller
in
the
last
days
of
work
has
really
become
much
more
mature
and
can
start
to
be
really
used
in
your
kubernetes
cluster.
A
There
is
actually
certain
advantages
for
doing
it
now,
so,
as
I
was
grumbling
last
week,
so
there
is
no
way
to
migrate
projects
between
our
GitHub
organizations
at
the
moment,
and
it
would
lead
to
a
royal
mass
in
everything
related
to
firstly,
roadmap,
second
backlog
and
also
there
are
other
issues
like
task
lists,
which
also
do
not
get
integrated
properly.
When
you
move
between
the
repositories
and
the
start
experimenting
with
them.
A
I
already
reported
this
effect
to
the
GitHub
team
and
also,
if
we
create
a
GitHub
Pages,
for
example,
for
a
temporary
documentation,
location
Etc.
All
of
that
will
also
want
to
be
persisted
between
the
changes,
so
I
would
rather
go
forward
and,
as
I
said
in
the
chat,
I
will
create
another
discussion
to
actually
make
it
a
default
to
create
projects
inside
the
captain
unless
it's
a
previous
unboxing
experimentation
project,
so
I
think
that
this
approach
with
multiple
GitHub
ports-
it
just
doesn't
make
much
sense
at
the
moment
yeah.
A
So
we
have
a
poll.
I
will
keep
it
for
a
while,
but
unless
there
are
strong
negative
feedback,
I
will
probably
start
to
move
today,
because
I
need
some
time
to
do
that
because
of
coupon
we
have
presentations
for
Captain
life
cycle
controllers
common
cup,
so
we
will
need
all
the
presenters,
for
example,
Thomas
Sandy
Alice,
who
else
presents
something
around
this
topic
just
to
fix
all
the
links?
So
it's
better
to
start.
A
A
B
B
Yeah,
you
might
notice
here
we
have
a
huge
list
of
features.
So
now
we
fully
support
application.
Please
pre-checks
post
checks,
but
also
we
added
the
possibility
to
run
SLO
and
SLI
evaluations
as
part
of
your
deployment.
So
you
can
do
some
checks
before
each
workflow
workload
and
before
each
application
is
being
deployed,
but
also
post
evaluations.
C
B
A
A
Yeah,
so
basically
it's
totally
reasonable.
We
have
a
parallel
discussion
about
open,
SLO
adoption,
so
I'm
joining
today's
meeting
with
the
open
SLO
team
to
basically
see
what
we
can
flesh
out
with
the
standards,
because
there
are
some
uncomfortabilities
as
I,
presented
to
maybe
one
or
two
weeks
ago
at
the
community
meeting,
but
it
seems
to
be
very
feasible.
So
I
think
that
if
we
create
full-fledged
standard,
the
best
strategy
for
us
would
be
to
implement
open
a
slope
so
that
there
is
interoperability.
A
Thanks
a
lot
I
will
try
to
do
some
promotion,
so
Julie
is
off
these
days
and
I'll
track
to
catch
up
on
social
media.
This
afternoon,.
A
And
yes
speaking
of
openness
law,
so
you
may
remember
a
presentation
and
shared
that
highlights
some
changes
needed
Etc.
So
if
you
see
some
gaps,
I
would
appreciate
if
you
communicate
them
before
6
pm
today,
so
that
I
just
can
accommodate
it.
In
the
discussion.
D
Yep,
let
me
just
share
my
screen
all
right.
You
should
see
my
browser
now
yeah,
so
basically,
this
is
a
follow-up
to
the
video
that
Thomas
trevone
and
I
recently
recorded
about
the
newly
introduced
tracing
and
metrics
capabilities
for
the
lifecycle
operator.
D
So
you
might
know
there
is
also
a
tutorial
where
we
showcase
the
capabilities
in
this
area,
and
what
I've
added
today
to
that
is
an
example,
rafana
dashboard
that
you
can
use
to
view
all
the
metrics
that
the
lifecycle
controller
generates.
So
this
lives
in
the
examples
observability
directory
of
the
lifecycle,
controller,
Repository
and
yeah.
D
Here
we
have
also
added
to
make
commands
to
make
it
as
easy
as
possible
for
users
to
follow
this
tutorial
and
two
of
them
being
the
setup
for
the
port
forward
and
then
the
import
of
the
grafana
dashboard,
together
with
some
documentation
on
how
to
make
use
of
them
and
yeah.
So
in
this
screenshot
we
will
see
how
the
how
the
resulting
dashboard
will
look
afterwards,
and
here
that's
pretty
much
it
from
my
side.
A
Thank
you.
So
the
next
topical
on
list
is
actually
Captain.
1.0
timeline,
so
I
just
wanted
to
smaller
discussions.
If
you
had
the
previous
developer
meeting
at
the
community
meetings,
so
basically
we
reached
the
consensus
so
that
0
to
20,
which
used
to
be
on
the
milestones
for
captain
man
repository,
would
become
Captain
LTS.
Unless
something
really
critical
pops
up.
It
basically
puts
us
on
the
timeline
and
to
somewhere
in
late
November.
My
understanding
for
the
release
give
or
take
so
I
wouldn't
commit
to
an
exact
deed
right
now,
but
yeah.
A
So
this
what
we
have
with
Captain
LCS
right
now,
so
yeah
we're
planning
documentation
site
with
Mac.
So
this
is
the
next
topic
about
the
website
update,
but
I.
Think
from
the
captain
core
side.
We
still
have
some
time
to
deliver
features,
maybe
some
pending
clutches
which,
if
they're
safe
enough.
So
if
you
have
a
pull
request
open
to
the
captain
core
repository,
if
you
feel
strongly
that
it
needs
to
be
landed,
probably
it's
a
time
to
bring
it
up.
A
Well
font
save
purchase
is
a
bit
different,
but
yeah,
you
know.
So
that's
the
timeline.
B
One
thing
on
the
timeline:
maybe
I,
want
to
clarify.
One
thing:
is
we
aim
towards
the
end
of
November,
but
since
we
want
to
also
to
bring
the
different
services
like
Prometheus
data
dog
and
all
the
company
ones
as
part
of
the
LTS,
this
date
might
go
a
bit
on
the
future,
because
we
want
to
make
sure
that
the
compatibility
of
the
services
with
Captain
works
out
of
the
box.
A
Yes,
so
currently
we
are
working
with
Suresh
on
data
doc,
service
I'm,
also
working
with
York
and
a
few
other
people
on
Dyna
Trace
service,
which
we
agreed
to
include
in
the
LCS
park
because
of
the
directive
maintenance,
and
there
are
basically
three
remaining
Services.
We
have
on
the
list.
It's
well
a
web
hook
service,
it's
a
job,
executor
service
and
its
primitive
service.
So
I
guess
the
biggest
problem
for
us
right
now
is
primitive
service
right.
Exactly.
C
We
have
no
request
what
about
a
specific
download
page
when
we
go
into
LDS
right
now,
I
looked
I've
I've
got
I
ran
into
it
as
a
link
for
where
you
go
for
Sia
for
the
CLI
to
install
the
CLI
and
I
ran
into
it's
beautifully,
written
like
release,
notes
page,
which
I
finally
dug
down
far
enough
and
found
assets
which
gave
me
links
to
Captain,
but
not
to
the
captain.
Cli
yeah
I.
A
Think
we
will
go
more
or
less
use,
Captain
LCS
by
default.
So
when
we
have
links
installation
guidelines,
I
think
that
we
should
default
to
LCS,
because
it
should
provide
more
stable
experience
for
users,
and
we
know
that
for
fact
that
in
previous
releases
it
was
a
problem
for
early
adopters
to
keep
everything
in
place.
So
the
tutorials
will
always
outdated
and
probably
they
will
be
also.
A
There
will
be
a
bunch
of
integration
issues,
so
my
default
approach
would
be
to
recommend
LCS
everywhere,
except
for
the
features
where
we
explicitly
recommend
to
try
out
new
version,
because
there
is
something
available,
maybe
even
behind
the
feature
flag.
Yeah.
That's
why.
A
A
Download
so
I
guess
what
it
implies
that,
for
example,
for
these
links
Etc.
So,
basically,
if
you
follow
the
guidelines,
you
would
be
installing
calcius
the
consequences
that
for
latest
releases
we
will
need
to
create
something
else
to
have
all
these
download
infrastructure.
So
I
need
to
talk
to
more
to
understand
how
much
of
the
effort
it
would
take,
but
yeah
I
think
that
by
default
we
targeted
that
when
we
release
0.20
or
1.0,
basically
all
the
links
remain
the
same.
A
So
basically
the
current
release
by
plan
for
what
we
have
in
Main
Range,
it
becomes
abilities,
release
Pipeline
and
for
further
releases
of
latest
and
greatest
in
Jenkins
terms,
we
create
a
new
release,
Pipeline
and
probably
a
new
Repository.
A
Yep,
but
it's
actually
a
very
good
point:
Mike,
because
I
need
to
recommend
the
thing
in
the
cap.
I
haven't.
C
D
Would
suggest
for.
A
A
C
We
have
to
quickly
upgrade
for
a
security
release
and
we
can't
get
all
of
the
Integrations
done
that
quickly.
So,
if
they're
really
tied
to
an.
A
And
maybe
a
since
we
are
speaking
about
other
Branch
full
latest.
Maybe
they
could
just
have
continuous
delivery
for
now,
because
again,
if,
if
it's
latest
and
greatest,
if
you,
if
you
have
charts
Etc
I,
know
that
you
have
or
experience
with
doing
quite
quickly
cycles
and
Captain
when
it
was
needed
so
yeah
as
we
discussed
in
the
cap,
we
are
not
doing
this
monthly
releases
anyway
until
it's
strictly
needed,
but
we
can
actually
go
for
even
forward
and
do
continuous
delivery
or
automatic
weekly
releases.
C
Sure
I
think
we
discussed
last
week
we're
going
to
restructure
the
documentation
so
that
the
quick
start
and
the
concepts
and
installation
are
at
the
same
level
as
operate
and
manage
in
the
reference
Pages
to
implement
this
I
I
have
a
feeling.
I
can
do
in
a
day.
I'm
almost
positive
I
can
do
it
in
two
days,
but
there's
always
the
risk
of
surprises.
C
So
I
was
looking
for
a
time
when
we
could
shut
down
the
repo
and
I
was
told
that
everybody
in
Europe
has
November
1st
off,
which
means
almost
everybody's,
going
to
take
October,
31st
off
and
since
I'm
in
the
United
States
I'm
like
fine,
that's
my
time
so
Giovanni
is
it
possible.
We
need
to
like
at
any
given
point,
let's
close
as
many
PR's
as
we
can
and
then
create
the
1.0.x
source
tree
or
whatever
it's
going
to
do
and
do
that
standard.
Put
everything
over
there.
A
My
recommendation
would
be
to
create
one.x
Branch,
well,
not
Branch,
folder
as
soon
as
possible,
because
it
would
also
allow
to
document
all
these
download
guidelines
Etc,
so
we
can
build
for
it
and
yeah.
Basically,
when
on
this
All
Saints
update,
we
just
focus
on
moving
the
content
from
the
route
to
folders.
C
I'm
going
to
avoid
there,
it
just
I'm
right
now,
I'm
doing
I'm
doing
dark
improvements,
yeah
and
I'm.
Whenever
we
do
it,
let's
close
any
PR,
as
we
can
I'm
doing
doc,
improvements
right
now
that
do
benefit
0.19,
but
yeah,
that's
it,
but
we
quickly
need
to
get
in.
This
whole
thing
is
going
to
need
to
be
scrubbed
for
LTS,
okay,.
D
C
When
your
people
are
ready
to
start
dealing
with
that
too
I,
don't
I,
don't
know
that
we're
having
big
content,
new
things
to
document
or
any
I'm,
not
sure,
but
but
there's
probably
a
lot
of
little
cleanup,
so
we'll
coordinate.
That's
why
I
just
said
by
by
the
Friday
before
everybody
takes
off
for
All
Saints
weekend
I.
C
Repo
and
sooner
is
fine
and
then
and
is
any,
is
there
anybody
that's
going
to
be
inconvenienced
by
not
being
able
to
merge
stuff
into
captain.github.io
on
October,
30th,
31st
and
November
1st,
okay,
good?
Nobody
wants
to
do
that
anyhow.
Also,
just
I
gave
you
a
link,
just
a
heads
up
the
tutorial.
The
old
tutorials
are
a
mess
we
keep
talking
about
it.
We
finally
decided
we
have
created
an
actual
GitHub
project
for
the
10
tutorials
that
show
up
on
the
tutorials
page.
Each
one
is
an
issue.
C
Most
of
them
have
owners
who
so
far
haven't
done
anything
but
I'm
optimistic
who
are
supposed
to
go
in
and
evaluate
and
say
either.
This
isn't
where
you
know
this
is
so
broken
and
we
don't
need
it
or
we've
got
the
one
that's
been
active,
went
in
and
they
want
it
and
they're
already
making
changes
to
update
it.
But
as
as
Oleg
says,
if
they're
broken,
we
want
them
off
the
web
page.
C
So
so,
if
any
of
you
are
curious,
there's
a
link
where
you
can
see
any
action
or
if
anyone
that
you
care
about
any
of
the
tutorials
feel
free
to
jump
in
on
them.
Nobody
cares
if
two
people
make.
A
Announcements,
firstly,
there
is
quick
start
which
is
referenced
on
the
capitalization
page,
and
there
are
also
tutorials
on
tutorials
that
captain.sh,
which
we've
been
unable
to
maintain
I,
think
that
for
them
plan
is
just
the
basically
archive
what
we
have
and
see
whether
we
can
rebuild
particular
beats
with
killer
Coda
and
with
new
stuff.
C
Tutorials
are
the
most
expensive
way
we
have
of
transmitting
information
both
to
do
it
at
first
and
maintenance.
We
shouldn't
be
doing
tutorials
gratuitously.
We
do
them
if
we
need
them.
As
for
quick
start,
we
already
agreed
I'm
going
to
be
driving
a
stake
into
heart
of
quick
start,
probably
tomorrow
and.
B
C
A
But
yeah
thanks
a
lot
for
this
heads
up
and
yeah
I
think
the
three
will
be
running
in
the
communication
and
it
will
help
everyone.
So
any
other
topics
to
discuss
today.
C
A
Maybe
just
a
quick
heads
up
for
you
that
basically,
the
plan
for
Capital
CS
is
that
we
keep
the
current
release,
infrastructure
and
pipelines.
So
basically,
what
is
current
to
become
just
become
Celsius
and
for
latest
and
greatest.
We
can
probably
just
build
continuous
delivery
for
now,
or
maybe
some
lightweight
release
process,
because
we
might
need
another
chart,
repository
for
that,
etc,
etc.
So
maybe
we
can
just
kick
it
down
the
road
a
bit
until
there
is
real
use
case
to
make
a
major
release
of
LCS.
A
A
You
so
much
and
the
hell
is,
and
everyone
else
Johannes
who
is
else
is
going
from
Captain
arzal
is
going
if
I
understood
correctly
so
yeah.
It
will
be
quite
a
party
there.
Bye.