►
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
So,
hey
everyone
welcome
to
the
harvard
community.
The
agenda
for
today
is
I'll
talk
a
little
bit
about
some
of
the
releases
and
then
we'll
download
the
proxy
cache
feature.
So
before
we
go
into
the
demo,
a
few
announcements
we
released,
the
1104
as
well
as
the
200.1,
the
the
2.0.1,
was
quite
a
few
fixes
related
to
contribute
scanner
image
replication.
A
A
Image
application
tree
scanner,
we
also
removed
the
ability
to
ping
the
web
endpoint
out
of
security
concerns
and
then
we're
starting
to
look
at
the
harbor
2.2
release.
So
if
we
look
at
the
project
board
here,
the
2.1
is
on
track
to
ga
towards
the
end
of
august,
and
so
we
have
a
2.2
product
suggestions.
Swindling
here.
Some
features
listed
here,
aren't
necessarily
locked
down
yet
just
some
of
the
things
that
were
given
priority
to,
but
they
can
still
change
that.
A
There's
something
you'd
like
to
see
from
the
backlog
or
something
from
here
that
hasn't
hasn't
been
captured.
Please,
you
know,
raise
a
ticket
to
us
or
just
bring
it
up
on
the
committee
meeting
here
so
like
a
few
things
that
we're
considering
the
robot
account
being
shared
across
multiple
hardware
projects.
We
talked
about
that.
A
There
is
this
ticket
here,
which
is
fine
green
access
control.
This
is
referring
to
adding
additional
granularity
in
hardware
multi-tenancy
beyond
what
we
have
right
now
with
just
projects.
A
A
Prometheus
integration
to
expose
metrics
to
prometheus,
as
you
can
see,
this
is
a
pretty
old
issue.
We've
been
planning
for
this
for
about
two
years
now,
and
that
makes
sense
very,
I
think,
especially
when
deploys
a
kubernetes
cluster.
There
are
already
existing
solutions
for
atheists.
That's
very
robust
and
expose
all
your
end
points
your
pause,
your
ingress
to
prometheus
as
service
discovery,
definitely
something
where
we
can
think
about
for
2.2
release.
A
A
A
B
Yep:
okay,
okay,
okay,
let's
start
before
we
can
use
the
process
cache.
We
should
install
the
latest
field
and
television
of
the
hubble.
B
We
are
going
to
release
the
classic
cache
release
feature
in
2.1
and
we
have
already
merged
this
on
most
of
the
appears,
and
you
can
try
it
in
ladies
rebuild
after
install
the
latest
build
and
then
we
can
chart
these
features.
B
B
And
is
at
least
our
two
pre-conditions
for
create
our
project
catch
project,
and
now
we
can
create
our
project
cache
project,
alcohol,
proxy-
and
here,
when
you
are
one,
the
common
user
is
a
system
user.
Then
there
is
a
project
cache
options
here.
When
you
create
our
object
and
then
you
can
select
the
registries
to
be
proxied,
we
select
dot
hub.
B
Okay.
Then
there
is
a
process,
cache
type
project
created
after
credit.
Then
we
can
use
the
process
cache
income
in
dollar
command.
B
B
B
The
commander
to
pour
images
from
the
proxy
project
docker
or.
B
The
dog
poor
commander,
we
have
changed
it
to
needed
to
add
as
a
perfect
prefix
of
harvard
proxy,
and
this
is
the.
B
B
For
example,
the
lift
moy
is
already
catched
in
crossy
catches,
the
prostate
cage
and
the
weapon
name
is
a
prefix
duck
dog
harbor
proxy.
This
is
a
little
different
with
the
normal
project
for
other
feature.
Previous
features
such
as
members,
are
back
on
label
and
scan
and
the
retention
policy.
B
B
B
For
any
docker
images
that
you
didn't
want
to
retain
in
a
disk,
you
can
set
up
a
target
retention
policy.
For
example,
you
can
add
a
lure
that,
like
only
you,
can
recently
pour
that
and
for
example,
last
one
week,
then
it
will
remove
other,
not
use
any.
Any
images
is
not
not
used
in
recent
weeks.
B
B
Currently,
we
cannot
switch
it
back
to
our
normal
projector.
We
just
create
and.
C
Yeah
yeah,
and
could
you
click
the
new
project
again
yep?
So
we
cannot
create
public
proxy
cache.
B
A
Okay
and
the
the
permission
of
the
project
is
determined
by
the
permission
of
the
the
user
of
the
docker
hub
account
right.
A
Okay,
I
need
logs
under
for
that
project.
B
Launched
currently,
we
have
some
bugs
in
the
course
cache.
B
B
Yeah,
actually,
we
did
not
change
the
log
actually
in
proxy
for
proxy
project,
because
it
is
a
push
command
initiative
by
the
dot
hub
by
the
proxy
service.
C
D
B
C
C
D
But
but
I
so
it's
ideally
daniel,
it
should
be
three
things
right.
One
is
the
refresh
of
the
image
from
docker
hub
to
harbor
and
that's
the
and-
and
I
believe,
that's
what
it
says
create
here
right,
that's
maybe
that
one
is
and
then
the
second
one
is
the
pull
command
from
the
client
to
harbor.
E
D
Right
so
there's
two
things
that
happen
so,
for
example,
if
now
your
envoy
latest
was
updated
upstream
in
docker
hub.
If
you
do
another
pull,
you
should
show
again
two
commands
one,
the
refresh
from
docker
hub
to
hardware
and
one
from
hardware
to
the
client,
and
if
the
image
is
not
updated
upstream,
it
should
only
be
one
entry
which
is
client
to
hardware.
A
B
A
D
Maybe
there's
a
filter
click
on
advanced,
let's
see
if
you
have
a
filter
on
click
on
operations,
operations.
C
D
C
A
Deleted
the
he
keep
the
docker
monitoring,
he
deleted
the
image.
C
D
D
Anyway,
by
the
way,
let's,
this
is
amazing
work.
This
is
awesome
to
see.
I
love
that
we
make
so
much
progress
on
proxy
cache.
These
are
neither
little
little
details,
let's
there's
iron
iron
them
out,
but
this
is
amazing,
work
folks,
great
to
see.
C
By
the
way,
I
have
another
common
document,
I
think
when
you
trigger
the
pull
at
the
first
time
there
is
a
relatively
long
lag
between
the
you
know.
Data
transfer
actually
begins
what
was
happening
there.
C
Okay
yeah,
but
if
this
that
slow,
I
think
we
need
some
improvement.
C
A
C
C
C
B
C
B
B
E
A
I
thought
that's
all
we
have
to
find
for
today.
Does
anyone
else
have
any
questions
or
comments?
Any
announcements,
michael.
D
No,
no
nothing
new
on
on
our
end,
there
was
quite
a
bit
of
press
on
harbor
so
and
as
part
of
the
graduation
announcement,
so
so
that
was
good.
That
was
great
for
our
project.
Thank
you
to
all.
Thank
you
to
our
community.
D
I
think
the
the
next
step
right
now
is
everybody's
preparing
for
cubicon,
shanghai,
the
virtual
event
and
kubicon
europe
also
another
virtual
event
in
both
of
them
happening
in
the
next
month
and
a
little
bit
over
a
month.
So
if
you
have
sessions,
thank
you
for
your
contributions
to
hardboard.
If
you
want
feedback
and
advice
feel
free
to
utilize
everybody
in
the
community,
and
we
look
forward
to
all
of
you
attending
and
and
participating
in
those
events.
D
And
alex
when
you
know,
when
the
office
hours,
the
harbor
we
have
at
the
cncf
answer
bar
we'll
make
sure
we
publicize
it
to
the
community
in
case
anybody
wants
to
attend
and
and
visit
and
chat
with
us
yep
we'll
do.