►
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).
A
A
B
By
the
way,
if
anybody
wants
to
discuss
something,
please
add
it
to
the
agenda.
So
since
you
have
the
agenda
for
this
meeting,
if
there's
anything
else,
you
wanna
talk
or
bring
up
something.
Please
add
it
I'll
I'll
put
the
agenda
link
to
the
meeting
as
well
to
the
notes.
A
A
30
user
stories
we've
mapped
out
for
proxy
cache.
It's
looking
at
the
cache.
You
know
we're
trying
to
solve
the
issue
of
when
docker
clients
can't
connect
to
an
upstream
registry.
You
want
to
pull
something
from
a
registry
that
you
have
no
emphasis
to.
So.
If
you
have
harvard
deployed
you
can
start
use
harvard
as
that
middleman
hit
the
upstream
properly
with
walking
upstream
and
then
push
it
down
to
the
clients.
A
A
A
A
Review,
okay,
so
the
first
thing
is,
you
know:
how
do
you
create
a
proxy
cache?
What
does
it
mean
to
create
a
proxy
cache,
and
so
the
design
we
sort
of
gathered
around
is
imagine
you're,
trying
to
connect
you
up
to
a
repository
on
an
upstream
that
you
have
no
connectivity
to
so
imagine,
creating
a
project
or
a
proxy
project
very
similar
to
an
existing
hardware
project
right
now
that
allows
you
to
map
to
that
repository.
A
So
you
create
the
proxy
project
and
specify
the
employing
the
credentials
choose
the
name
and
aliases
projects
and
that
property
project
now
becomes
your
your
your
front
end.
Basically,
so
if
you're
trying
to
hit
the
repository
on
the
remote,
you
can
just
docker
people
from
the
and
you
can
do
that
for
any
number
of
target
repositories
and
they
could
be
target
apostrophes
from
the
same
carbons
on
the
same
registry
instance
or
they
could
be
across
different
target
registries.
A
A
So
again,
you
know,
there's
there's
a
role
based
access
control.
It's
mostly
like
it's
most
likely
going
to
be
read
only
so
you
know
this
ties
into
the
third
story
here
as
well.
So
cash
is
boiling.
You
can't
really
push
to
the
cash.
A
The
cash
is
populated
from
pulling
from
the
republican
repository
via
some
sort
of
an
internal
mechanism,
but
in
terms
of
user
attack
interaction,
the
user
only
pulls
from
it.
The
user
doesn't
really
push
from
it.
You
can
configure
credentials
where
the
cache
can
be
public
or
private
and
the
experience
we
want
to
deliver
to
the
client
around
pulling
from
the
cache
is
you
know
we
don't
want
any
changes
made
to
the
poll
command.
A
A
I
don't.
I
want
to
be
able
to
hit
my
remote
repository
without
having
to
edit
the
box
button.
So
imagine
you're
trying
to
hit
gcr
dot
io
so
through
the
use
of
a
mutating
web
hook
or
something
else
that
can
be
translated
to
a
hardware
proxy
project.
That's
been
set
up
for
it
or
if
you
specify
the
project
proxy
in
the
prospect
directly,
then
they'll
just
pull
from
the
cache,
but
you
know
we
don't
want
to
have
to.
A
A
If
it's
in
the
cache
and
there's
a
newer
copy
in
the
upstream
it'll
always
pull
the
latest
copy
from
the
upstream
before
serving
it
to
the
client
so
because
you're
pulling
by
tags
with
it,
so
imagine
that
tag
is
now
mapped
to
a
different
digest
on
the
upstream
it'll
pull
the
latest
digest
and
then
serve
that
decline
and
if
the
cache,
if
the
upstream
doesn't
have
the
image
or
the
image
has
been
removed,
then
it
will
basically
describe
the
request.
I
think
the
reasoning
here
is
mostly
driven
by
intent.
A
A
A
So
you
can
say
something
like
you
can
specify
the
retention
time
during
during
the
creation,
which
means
how
long
the
artifacts
will
live
in
the
cache
before
it's
removed
or
you
can
configure
something
like.
I
only
want
to
retain
I'll,
only
only
retain
the
most
recently
five
artifacts
or
something
like
that,
so
negative,
one
for
infinite.
A
Yeah
as
a
stretch
goal,
I
think
this
is
referring
to
you.
Have
the
ability
to
to
constantly
pulse
the
upstream,
and
the
cache
itself
will
always
update
itself
with
the
dispersion
from
the
stream,
because
we
had
talked
about
you
know
the
behavior
would
be
to
always
pull
things
upstream,
refreshes
copy
anyway,
so
this
is
just
doing
it
in
the
background
to
minimize
the
likelihood
of
actually
doing
it
while
you're
when
you're,
when
there's
an
actual
pull
from
the
proxy.
A
B
B
B
A
But
yeah,
so
that's
referring
to
for
those
who
aren't
attending
meetings,
but
it's
basically
for
certain
type
of
commercial
scanners.
There
needs
to
be
a
special
robot
account
that
has
the
ability
to
pull
images
across
from
harvard,
because
your
harvard
has
various
security
deployment
policies.
You
can
say
something
like
you
know.
A
Any
image
with
a
certain
vulnerability
of
a
certain
severity
cannot
be
pulled
from
so
there
needs
to
be
right
because
the
the
security
scanners
are
using
the
robot
accounts
to
pull
the
images
across
before
they
can
scan,
so
they
need
to
get
around
them
somehow
yeah.
Hopefully
we'll
have
more
on
that
for
the
next
meeting
I'll
work
with
you
on
that.
A
So
yeah
that
that's
all
I
really
wanted
to
share
2.1
is
targeted
for
the
end
of
august,
so
we
can
look
at
some
of
the
features
that
are
being
planned.
A
So
if
you
guys
have
any
questions
on,
you
know
things
that
are
being
worked
out
for
the
next
release.
Please
look
at
this
particular
swimlane
2.1
project
suggestions,
which
will
turn
those
into
commitments
at
some
point.
A
A
Yeah
stephen:
do
you
want
to
share
something
about
the
2.0
release.
B
Rc2
is
on
github.
So
if
you
go
under
the
releases
that's
available
and
the
targeted
ta
for
the
full
release,
general
availability
is
one
week
from
today,
so
on
the
13th,
and
at
that
time
we'll
have
the
blog
post
out
talking
about
all
of
the
different
features
of
to
the
daw.
That
includes
oci
ssl
between
the
different
components:
slack
integration
for
web
hooks
individually,
callable,
workbooks
and.
A
B
We're
not
discussing
the
operator
in
the
blog
quest,
because
we
already
had
a
specific
target,
the
blog
post
for
the
operator,
but
everything
else
will
be
in
there
and
next
wednesday
have
everything
wrapped
up
and
if
everything
goes
well
documentation
and
everything
else
will
be
in
there,
be
aware
that
at
the
end
of
the
month
on
the
28th,
we
also
will
have
a
hardware
webinar,
where
we're
going
to
basically
talk
about
carbotudo
and
everything
else.
That's
happening
within
the
general.