►
From YouTube: Support Release Review - 12.8
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).
B
A
A
A
Using
this
new
feature,
we
can
install
same
applications
using
CI,
so
why
it
is,
why
do
we
need
it
and
why
it
is
important.
So
when
we
install
applications
using
single
click,
we
we
cannot
customize
anything.
So,
for
example,
if
we
install
ingress,
it
will
be
just
installed
with
some
predefined
where
values
and
we
not
manage
them.
If
we
so
toll
runner
and,
for
example,
we
want
to
use,
let's
say:
s3
object,
storage
for
cash.
It
won't
be
possible
to
modify
it
from
here.
It's
just
not
possible
to
do
it
in
right.
A
A
A
So
there
are
no
templates
for
this
project,
so
we
just
we
should
just
create
an
empty
one.
After
that
we
should
go
to
advanced
settings
of
our
cluster
and
pick
up
this
newly
created
project
here.
So
it
was
already
done
and
I
guess
and
after
that
we
should
create
some
files
in
in
the
project.
Folder
and
the
examples
of
these
files
are
given
in
the
documentation.
So
there
is
an
example
project
at
github.com
that
we
can
use
as
a
reference,
so
there
are
basically
three
main
main
times
of
types
of
files.
A
So
then
there
should
be
the
d
drop,
managed,
apps
folder,
with
configure
Tiamo
file
and
in
this
file
we
can
define
what
applications
we
would
like
to
install
in
our
cluster.
So
here
we
can
set,
for
example,
everything
to
true,
and
it
will
be
almost
the
same
as
if
we
just
install
applications
using
single
V.
But
let's
say:
let's
imagine
that
we
want
to
modify
one
of
the
applications.
C
C
I
was
I
was
because
I
was
actually
going
through
twelve
point,
six
Phillies
pretty
deeply.
This
is
like
two
months
ago
and
I
just
skip
it.
It
gets
like
okay,
sure,
I,
don't
know
why
it
was
so
special
until
you
made
you
bring
it
out
right
now
and
I'm
like
actually
yeah.
This
is
act
like
one
of
those
things
that
is.
C
A
C
Thank
you,
so
one
question
that
I
have
here
Alex,
so
the
installation
of
this
in
this
application
is
per
cluster
basis
right,
as
in
for
each
cluster
that
we
attach
to
to
get
lab,
we
can
install
this
yeah
yeah,
that's
right!
No,
because
this
is
actually
triggered
from
the
CI.
How
like,
when
do
we
need
to
trigger
it
like?
Is
it
gonna,
be
like
a
one-off
thing
like
that
how's
that
work?
A
A
C
C
A
C
C
A
A
C
C
Other
thing,
the
other
thing
I
wondering,
is:
what's
the
what's
the
best
practice
of
doing
this
again,
this
is
beyond
like
the
feature
itself
right.
Do
we
usually
want
to
put
this
cluster
description
in
the
same
project
as
the
project
that
use
the
cluster
or
you
just
have
a
separate
project
just
for
cluster
management
project?
A
B
C
B
But
it's
extra
logic
that
you
can
just
avoid
by
having
a
separate
project
and
also
by
linking
the
cluster
to
a
project.
What
you're
effectively
doing
is
that
the
project,
then
it
runs
its
CI
right.
It
has
access
to
the
cluster.
That's
really
what's
happening
because
you're
not
going
to
actually
provide
any
kind
of
clustered
information
in
the
project
itself,
except
you
just
have
the
management
commands
there.
The
actual
configuration
of
the
cluster
like
in
Alex
T's
case
it's
at
the
instance
level,
and
that
is
where
the
certificate
and
API
keys
are
defined.
B
So
from
the
project's
perspective,
when
the
CI
runs
its
get
labs
job
to
make
those
credentials
available
to
the
CI
at
the
time,
so
the
project
CI
job
can
actually
manage
that
cluster.
Using
those
credentials
right
and
in
that
sense,
I
would
be
surprised
if
you
have
the
bracken
stern
and
you
can
link
that
cluster
also
to
the
same.
C
C
So
so
what
I
saying
so
so
so
it
would
be
one-on-one
in
that
way,
but
also
you're,
saying
that
the
the
the
better
the
better
work,
what
the
the
base
the
best
practice.
The
best
practice
should
be
on
a
separate
project,
because
that
probably
yeah
that
key
and
all
that
might
be
shared
in
there
right.
You.
B
The
idea
is
that
you
can
decide
I
mean.
Maybe
there
is
some
benefit
to
having
a
thing
project,
but
then
my
I
imagine
that
if
you
do
have
a
single
project
and
you
you
merge
it
with
your
existing
CI
for
example,
then
you
have
to
put
in
some
kind
of
a
clause
in
your
CI
job
that
look
every
time
the
CI
runs.
I
don't
mean
to
manage.
The
cluster
I
only
mean
to
manage
the
cluster.
C
Yeah
yeah
I
mean
I
mean
like
we.
We
do
have
that
syntax.
If
a
particular
directory
change,
then
we
trigger
this
particular
job
only
and
only
if
this
sure
directory
change.
So
we
do
actually
have
the
feature
in
CI
but
again,
like
I,
said
I'm
I'm
not
asking
about
whether
we
can
or
cannot
do
because
I.
B
C
B
A
B
A
I
have
not
tried
so
I
have
tried
to
install
it
it
like
a
long
time
ago
and
using
this
method,
like
single
click
install
but
I
haven't
tried
installing
it
using
ci.
Okay,.
B
B
B
Here's
what
you're
going
to
do
what
I
have
done
before
the
start
of
this
thing
is
that
I
have
I,
have
set
up
a
project.
I've
set
up
a
project
on
my
machine,
which
is
a
dotnet
project.
Actually,
so
you
can
see
the
project
here
and
it's
a
it's
a
simple
project.
Actually,
you
can
see
the
program
file.
It's
just
going
to
print
hello
world.
All
right,
I
have
no
idea
what
this
weird
character
here
is,
but
it's
there.
B
This
is
the
project
that
I
have
already
pushed
okay,
so
what
we
are
simply
going
to
now
do
is
that
we
want
that
in
the
CI
of
this
project.
Every
time
the
CI
runs,
I
mean
we
build
the
package
and
then
we
kind
of
push
it
to
the
package
repository.
The
package
repository
can
be
seen
here,
so
I
can
go
to
the
package
list
and
I
have
already
published
one
package
here.
So
this
is
exactly
similar
to
all
the
other
package
repositories
that
we
have.
For
example,
we
have
for
NPM.
B
We
can
publish
docker
images
as
well,
but
that's
the
container
registry,
but
I
mean
the
concept
is
the
same
right?
Okay.
So
how
do
you
go
about
kind
of
doing
this?
Let's
have
a
quick
look
at
at
the
commands
that
are
necessary,
so
these
are
effectively
the
commands
that
we
are
talking
about
right.
So
the
first
thing
that
you
need
to
do
is
you
need
to
add
to
nugget
a
source
okay.
B
So
if
you
look
at
this
command,
here's
what
the
force
looks
like
so
you
say
nugget
source,
add,
and
then
you
give
a
name
for
source,
so
you
can
call
it
whatever
right,
like
I'm,
calling
it
in
my
app
and
then
I
give
a
source
URL,
which
is
the
which
is
a
full
URL.
The
API
URL
for
this
project
with
the
project
ID
package
is
nugget,
indexed
or
JSON
I
provide
my
username
and
I
then
also
need
to
provide
an
API
key.
Alright,
that's
what
the
source
looks
like
once
the
source
is
added
right.
B
This
isn't
this
isn't
deep
lab
specific
stuff,
but
really
this
is
the
kind
of
setup
you
need
to
do
after
that.
What
I
do
is
I
run
the
command
called
nugget
spec
running
maggots
back.
It
simply
generates
a
template
of
this
file,
which
is
the
my
add,
a
new
spec
file.
This
file
has
its
the
metadata,
it's
kind
of
like
a
package
store,
JSON
file
for
NPM
projects
and
so
on.
It
contains
like
metadata
about
my
project,
my
project
version
and
so
on.
B
B
Ok
now
you
can
think
of
this
as
like
a
package
that
you
can
just
download
from
somewhere
and
integrated
your
project
or
fully
using
nugget
or
specify
as
a
dependency
in
your
in
your
dotnet
project.
So
this
package
is
now
here
all
right
now,
of
course,
the
last
step
would
be
for
us
to
push
this
package
to
get
lab
right,
and
to
do
that,
it's
really
simple.
B
We
will
just
give
the
package
file
name
here,
so
I'm
just
going
to
copy
and
paste
it
here
and
then
the
source
is
what
we
configured
earlier
on,
so
remember
that
at
the
very
beginning,
I
ran
this
command
to
create
a
source.
In
my
nugget,
configuration
and
I
call
that
source
my
app.
That
is
the
name
that
we
are
going
to
use
here
right,
so
I'm
just
going
to
wipe
this
out
and
I'm
going
to
save
my
app
here
and
then
I
can
simply
do.
B
I
can
simply
run
this
command
and
it
has
successfully
pushed
the
package
right
now.
If
we
go
to
the
web
browser-
and
we
refresh
this
page,
we
can
see
that
this
package
has
been
published.
The
game
I
did
not
implement
demo
in
the
CI,
but
of
course,
that
demo
would
actually
have
the
same
commands.
I
would
imagine
that
there
would
be
some
kind
of
a
docker
image
that
we
can
use.
In
fact,
I
recently
also
saw
some
ticket
where
somebody
was
building
a
dotnet
application
using
our
CI
I
think
it
might
still
be
open.
B
The
other
thing
I
want
to
mention
is
that
you
see
this
warning
here
and
I'm
not
sure
what
warning
this
is
because
I
already
added
the
password
which
is
technically
the
API
key,
but
I,
don't
know
if
the
API
key
my
API
key.
They
mean
something
different.
So
here
the
password
was
specified
when
I
was
entering
the
source
it
was
specified,
but
this
password
is
really
the
API
key,
so
they
say
that
no
API
key
was
provided.
B
It's
just
a
warning
doesn't
mean
that
something
is
not
working,
but
maybe
this
is
something
that
one
could
look
at,
but
kind
of
irrelevant
to
get
love.
But
in
any
case
the
packet
is
pushed
and,
as
I
said,
it's
kind
of
available
here
and
it
can
be
used
to
let's
say,
deploy
this
application
or
whatever
yeah.
That's
it
any
questions.
C
B
B
C
B
B
C
C
In
Nathan
you
can
use
a
different
thing.
Is
you
can
use
a
CI
job
token
as
as
a
way
to
authenticate
to
to
the
package
manager?
I
guess
maven
is
been
introduced
last
year,
right
so
I
guess
it's
been
a
year
and
they
stopped
like
adding
more
feature
at
this.
At
this
point
in
time.
I
guess
the
the
new
gate
will
with
can
only
do
personal
access.
Token
yeah.
C
B
Yeah
because
I
mean
we
don't
really,
the
password
is
already
saved,
so
I'm
not
sure
what
this
is
about.
Iranian,
looking
at
the
documentation
for
today,
yeah
I
mean
it
was
kind
of
not
really
relevant
to
get
lamb,
so
I
was
like
okay.
I
can
ignore
this.
For
now,
but
yeah
I
mean
I
can
imagine
getting
the.
C
B
C
Thanks
no.
B
C
C
C
C
C
C
C
Yes,
nyanko,
that's
the
that's
one.
I
asked
I
put
the
link
on
on
the
document
as
well.
I
think
it's
part
of
it
is
like
yeah.
They
have
this
artifact
management
thing
right,
but
they
seems
to
guess.
In
our
case,
we
have
to
actually
configure
specific
technology
that
we
were
like
specific,
like
is
it
a
c-sharp?
Is
it
I,
don't
know
Java
or
whatever,
but
they
J
for
at
the
factory
seems
to
be
not
the
case
like
you
can
just
push
something
and
then
behind
the
scene
they
kind
of
figure
it
out.
B
I'm
curious
how
that
might
work,
because
all
the
commands
that
you
use
to
push
packages
like
mvn
or
NPM
or
whatever
they
are
designed
to
talk
to
their
own
type
of
package
manager.
I
mean
that's
the
protocol.
They
understand
so
yeah
that
and
even
artifactory,
unless
I
have
a
proprietary
way
of
interacting
with
it
and
not
the
usual
commands
like
NPM,
and
all
that
it
is
effectively
the
same
thing
as
what
guitar
has
should.
B
Open
stuff,
like
basically
it
understands
holder
for
like
when
you
run
the
command
it
automatically
kind
of
put
it
into
that
protocol
like
oh,
it's
an
NPM
thing
coming
in.
So
let
me
act
as
an
NPM
repo
now
or
something
like
that,
but
I
mean
I.
Think
the
difference
is
very
trivial.
The
most
I
can
imagine
is
that
okay,
the
URL
will
stay
the
same.
Perhaps
in
our
case
that
URL
is
a
bit
different
right.
D
B
B
B
It's
it's
pretty
incredible
what
we
are
doing,
a
collab,
so
I
guess
that's
a
good
thing
and.
C
C
C
C
C
B
C
I
actually
got
through
the
documentation,
but
it
took
me
a
while
to
go
through
the
whole
documentation
because
I
didn't
like
highlighting
the
feature
one
by
one
kind
of
thing.
But
let
me
again,
this
is
like
a
brief
overview
thing
that,
like
like
kind
of
what
I
did
before
I
haven't
finished
everything
to
be
fair,
but
I
just
have
I
just
put
down
whatever
I
have
so
far
sure
so
I
just
put
it
on
the
bottom
of
the
whole
thing.
C
B
B
B
Yeah,
so
cool
I
mean
we
have
I
mean
next
time.
Whoever
is
doing
this
I
mean
we
can
always
add
a
note
section.
I
think
the
format
of
the
table
might
also
evolve,
so
feel
free
to
do
whatever
you
think,
makes
more
sense
and
I
think
we
will
eventually
reach
some
kind
of
equilibrium
in
terms
of
how
to
be
record.
B
B
B
B
B
B
C
C
C
C
A
C
C
A
And
another
actually
sure,
maybe
it's
also
shown
for
clusters
and
group
level,
but
I
saw
it
only
for
project
so
on
instance,
white
cluster.
It's
not
sure,
and
if
I
go
to
my
management
project
it
will
just
it
will
not
show
anything
and
I
guess
that's
because
of
the
just
because
of
the
fact
that
this
project
is
specific.
A
C
C
Aggregation
thing
be
very
difficult
to
figure
out
what
happened
so
I
think
this
is
kind
of
one
of
those
things
that
going
to
be
more
important
in
the
future
in
terms
of
like
debugging
on
what's
happening,
we
don't
know
yet
how
this
is
implemented
and
how
to
enable
it
and
things
like
that,
but
I
yeah
I
wanted
to
highlight
that
part
just
because
I
think
at
some
point,
if
sooner
or
later
we
probably
gonna
spend
a
lot
more
time.
Looking
into
these
things,
the
more
customers
looking
into
manage
apps
and
all
that
yeah
yeah.
C
There
might
be
some
issues
with
terms
because
if
you
look
into
compliance,
we
have
license
compliance,
that's
talking
about
when
they
use
some
open
source
project,
and
then
you
wanted
to
know
what
licenses
are
being
used.
However,
this
compliance
dashboard.
It's
got
nothing
to
do
with
that,
so
I
don't
even
know
like.
Maybe
the
naming,
it's
not
quite
good
enough
from
from
our
perspective
or
something
and
you
see
license
compliance.
C
Again,
I'm
not
exactly
sure
how
this
is
gonna
evolve,
but
yeah
there's
also
license
compliance,
which
is
part
of
the
CI
CD
things
so
yeah
there
could
be
I,
don't
know
if
we
dealing
with
ticket
at
some
point
in
the
future
and
they
say
compliance,
then
you
need
to
sort
of
look
into
more
detail
on
the
context
of
what
are
you
where
they're
saying?
Is
it
a
the
merge
request
compliance
or
is
this
like
a
yeah,
the
license
compliance
I,
just
copy
paste.
C
Yeah,
anyway,
just
something
to
be
aware
of
the
next
one
is
the
network
policy
for
container
networks.
Security
yeah
I
have
another
too
much
details
into
this,
but
this
is
something
this
looks
pretty
interesting,
yeah
we're
using
psyllium
if
anyone
played
with
it
before
I've
never
played
with
it
before.
C
C
So
so
the
the
fourth
one
here,
the
network
policy
contained
in
Metro
security.
So
we
introduced
these
network
policies
for
container
network
security
powered
by
psyllium
I.
Think
the
point
is
the
reason
why
this
is
important
is
because
my
understanding
on
how
kubernetes
works.
If
you
have
multiple
parts
within
kubernetes,
there
is
no
actual
like
they
can
just
talk
to
each
other
right.
There
is
no
there's.
No
networking
in
that
way
is
that
is
that
correct?
It's
my
understanding,
correct!
You
know.
A
C
So
it
might
be,
it
might
be.
The
networking
is
on
the
pot
level,
but
then
might
not
be
on
the
it
may
be
on
the
application
level,
but
not
in
the
infrastructure
level
or
something
along
that
line
again.
This
is
something
that
I've
probably
need
to
read
more
about,
but
yeah
I
I.
Just
remember
at
some
point.
I
was
reading
about
this
and
I'm,
like
all
the
parts
can
talk
to
each
other,
that
sounds
iffy
and
when
I
read
this,
like
oh
yeah,
okay,
of
course,
make
sense.
C
We
have
this
now,
but
yeah
again
take
everything
with
a
grain
of
salt
I,
don't
know
for
sure,
but
I
think
that
was
gonna
pass.
The
whole
issue
with
kubernetes
and
internal
network
security
anyway
move
on
blocking
issues.
So
this
blocking
issues
I.
Think
if
you
look
in
issues
what
we're
looking
to
open
issues
all
the
time,
I
guess
so
we
know
this
is-
has
been
implemented
in
Geelong
for
quite
a
while.
So
it's
just
another
type
of
related
issues.
It's
really
nice
for
issue
organization
as
well
as
there
is
a
new
icon.
C
If
you
look
in
the
issue
board,
if
the
issue
is
blocked
issues
yeah,
it's
it's
not
really
a
super
massive
feature
or
anything.
It's
just
more.
It's
just
a
nice
addition.
C
C
Yeah
so
basically
can
just
set
a
work-in-progress
limit
and
it
just
kind
of
change
color
and
it's
not
gonna
block
any
it
shouldn't
block
anything
from
from
from
from
our
work
perspective
perspective
like
I,
don't
he's
not
gonna,
stop
people
adding
things
to
the
to
the
issue
board,
but
yeah,
it's
just
good
to
know,
environment
out
of
stock.
This
is
coming
from
us.
You
know
sysadmin
managing
servers
and
cost
blow-up
happen.
C
It's
kind
of
funny
said
can
be
manually
overridden
in
the
UI,
not
really
sure
how
that
works,
probably
from
the
editing
button,
and
they
haven't
tried
this
myself
but
yeah.
It's
kind
of
cool
yeah.
It
is
a
simple
addition
from
from
the
customer
perspective.
They
just
need
to
add
that
line,
but
yeah.
It's
we
have
this
now,
so
it's
cool
next
one's
just
another
improvement
is
the
instant
level
security
dashboard.
C
Sophisticated
dashboard
has
been
around
for
quite
a
while.
It's
now
half
the
instance
level
instead
of
the
group
level
well,
on
top
of
the
group
level
on
the
project
level,
so
yeah
pretty
cool,
so
the
admin
can
sort
of
see
everything
between
their
good
live
instance
and
the
last
one
is
about
accessibility
of
scanning
accessibility
scanning
for
review
apps.
C
C
C
All
the
release
above
the
the
columns,
if
that
makes
sense,
if
you
look
into
the
the
release,
I
haven't
gone
through
the
bottom
ones,
because
so
yeah
it'd
take
me
a
while
to
go
through
one
by
one
to
just
sort
of
like
we're
trying
to
understand
what
what
it's
all
about
but
yeah.
Hopefully,
that's
helpful
yeah.
C
Worries
was
useful,
yeah
it
it's
kind
of
I
personally,
find
it
useful
when
you
start
thinking
about
what
kind
of
implication
is
this
gonna
happen
for
the
users
for
us
dealing
with
with
tickets,
but
I
kind
of
like
your
details,
clinician
as
well
Alex,
because
again
I
saw
that
two
months
ago.
I
didn't
really
think
too
much
of
it
was
like
okay
sure,
but
you
know
when
you
start
saying
that
you
can
customize
it
I'm
like.
Oh,
actually,
that's
really
big,
so
yeah
I
think
I.