►
From YouTube: Keptn Community & Developer Meeting - July 5, 2023
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
B
A
Let's
make
this
bigger
okay,
so
in
today's
agenda
we
have
three
things
now
24
hours
too.
The
first
is
about
a
few
bugs
that
Florian
and
Giovanni
found
in
klt,
and
we
decided
that
this
is
enough
to
justify
release
fix
the
bugs
are
listed
here
and
I
have
seen.
Most
of
them
have
already
solution,
PR
merged
or
work
in
progress.
That's
blue!
A
The
first
is
about
multiple
Captain
tasks,
and
then
we
have
some
problem
with
the
hand
chart
validation,
Web
book
was
not
having
the
injection
of
our
certificates.
A
B
This
one
exactly
so
there
we
had
a
problem
that
the
captain
task
controller
was
registered
as
the
owner
for
kubernetes
jobs,
which
means
that
the
reconciliation
look
for
kept
tasks
was
triggered
every
time
there
was
an
update
of
a
job
resource,
and
that
means
that
basically,
in
the
reconciliation,
Loop,
the
task
controller
checks.
B
Whether
a
job
already
exists
for
that
task,
if
not
creates
it,
and
that
then
triggered
the
reconciliation
loop
again
before
the
task
controller
could
update
the
status
of
the
task
resource,
and
this
meant
that
it
thought
there
was
no
job
yet
and
basically,
it
kept
creating
new
jobs.
But
that's
luckily
resolved
now.
A
B
Exactly
this
was
related
to
the
captain,
app
version
or
Captain
workload.
Instance
controller,
and
this
was
affected
by
the
scenario
where,
for
example,
you
have
multiple
tasks
defined
for
say
the
pre-deployment
phase,
and
one
and
for
one
of
the
task
definitions
is
not
available.
So
then,
during
the
loop,
where
all
tasks
were
created,
the
task
Handler
returned
an
error
which
caused
the
status
not
to
be
updated
so
similar
to
the
other
problem.
No.
A
B
Okay
yeah.
Basically,
as
we
announced
in
the
last
couple
of
meetings,
we
were
busy
implementing
the
functions,
Runner
runtime
for
Python
and
containers,
and
now
I
would
like
to
show
you
a
really
quick
demo
for
that.
So,
basically
we
are
in
the
sample
app
directory
within
our
repository
and
there
we
have
added
a
python
task.
So
let's
first
look
at
that.
So
here
we
have
the
captain
task
definition,
and
this
is
pretty
similar
to
the
to
the
previous
JavaScript
definitions
that
we
also
had
with
the
difference
that
we
now
have.
B
This
python
object
where
you
have
the
basically
the
same
capabilities
you
had
in
as
in
the
JavaScript
runtime,
so
you
can
either
directly
insert
the
code
for
your
python
script
or,
as
we
did
in
this
case
here
we
are
referencing
a
URL,
so
let's
have
a
short
deck.
So
this
is
a
really
really
powerful
script
that
basically
reads
the
data
environment
and
very
and
very
environment,
variable
kind
of
talk
today
and,
for
instance,
and
yeah,
and
in
this
case,
just
for
the
sake
of
demoing
this
that
we
can
also
do
this.
B
This
the
secret
here,
where
we
have
some
secure
data
and
this
will
be
executed
in
the
pre-deployment
phase,
I
believe
of
the
catnip
yeah
exactly
so
here
in
the
cap
nap
in
version
3,
for
example,
we
have
two
pre-deployment
tasks.
One
is
the
python
secret
task
that
we
want
to
execute
and
the
second
one
is
a
container
sleep
task.
So
let's
get
really
quick.
Look
at
that
one.
So
this
is
the
newly
introduced
container
runtime.
B
So,
first
of
all,
the
parts
for
the
application
has
started,
and
hopefully
soon
we
should
see
the
ports
for
the
tasks
here.
They
are
and
yeah.
Let's
have
a
look
at
the
lock.
So
here
we
have
our
super
secure
token
printed
out
by
the
python
function
that
we
wrote
and
likewise
in
the
container
function,
we
print
out
this
beautiful
Hello
message
and
yeah.
A
Yeah
I
believe
we
don't
have
much
more
to
discuss
about.
Maybe
we
can
look
again
at
the
current
Milestone
I
think
last
week,
master
was
the
same,
so
maybe
we
won't
spend
long
on
this.
As
far
as
I
know,
we
are
working
on
this
on
klt,
which
is
the
scheduler
Readiness
Gates,
and
currently
we
had
some
contribution
from
Joffrey.
A
He
already
removed
the
decoder
injector
and
he's
working
on
the
support
for
the
controller
runtime
about
that.
This
is
a
bit
complicated
for
a
contributor,
because
basically
we
have
dependencies
among
each
of
the
operators
that
needs
to
be
updated
and
to
do
that
in
a
fork
means
that
you
need
to
change
and
refer
to
commits
of
your
own
Fork
inside
of
those
of
our
main
Rebel
or
you
need
to
merge
a
PR
which
is
not
passing
classes
so
by
Friday.
A
We
will
try
to
figure
out
this,
but
anyhow
thanks
Jeffrey
for
volunteering
and
helping
up
to
now
for
this
other
task.
I
think
we
discussed
it
in
multiple
times
so
I
guess
there
is
no
need
for
more.
Does
anybody
have
any
other
issue
that
I
forgot
to
mention
in
the
agenda.