►
From YouTube: Velero Community Meeting/Open Discussion - June 2, 2020
Description
June 2, 2020
Status Updates
michael
Discuss the community roles in Velero Governance
nrb
Added a draft PR for automating our release process
Reviewed some community PRs, specifically reporting progress on volume snapshotting
Working on some internal Velero-related projects
On community support this week
ashish
Moving all our plugin repos to use go modules
Decommission Travis and switch to using github actions for our CI and release
Have you tried the new CSI volume snapshotting feature in 1.4. This blog post should help you get started.
Reviewed some PRs on the restore hook and the volume snapshot progress reporting proposal added comments.
stephanie
discuss up-vote and down-vote on feature requests, and announcement of ‘go-live’ for that capability later today
carlisia
community support and catching up on PRs last week
Addressing code reviews for https://github.com/vmware-tanzu/velero/pull/2561 (Convert manifests + BSL api client to kubebuilder)
jonas
Introducing Abbie, Tony, Duffie, Matyas, and Mike
Discussion Topics
michael
Discuss the community roles in Velero Governance
Action Item - Carlisia to prep a PR for a reviewer role. Let her know if you’re interested in being a reviewer!
carlisia
Modernize the Velero code base/tests
abbie
Documentation style guide
kevin leffew
Kubernetes backup to the decentralized cloud via Velero and StorJ
https://twitter.com/kleffew94/status/1262382842086141953
https://forum.storj.io/t/kubernetes-backup-via-velero/6879
A
Hello
and
welcome
to
this
week's
episode
of
the
Valero
community
meeting,
slash
open
discussion
today
is
June
2
to
2nd
as
how
you
pronounce
it
June,
2nd,
2020
and
yeah
we're
going
to
bunch
people
here
on
the
call,
fantastic
and
I
will
introduce
some
of
these
lovely
folks
later
on
in
this
meeting,
but
for
now,
let's
get
through
our
status
updates.
If
you
have
anything
that
you
would
like
to
add
to
the
meeting,
please
do
so
in
the
meeting
notes.
A
B
Yeah
absolutely
Thank,
You
Jonas,
so
hello,
everybody.
So
if
you've
not
noticed,
we
actually
created
a
little
governance
document
that
describes
the
high-level
view
of
how
we
run
the
community
for
Valero
I.
Think
of
this
as
the
Charter
for
our
project.
It
includes
their
different
repositories.
I
do
have
under
management.
B
B
And
how
can
we
help
users
that
are
interested
in
contributing
to
Valero
have
a
good
path
towards
that
and
I
was
hoping
to
have
a
discussion
on
that.
I'll
offer
my
thoughts.
Hopefully
Kieran
and
others
do
view
this
offline,
and
maybe
that's
something
that
you
know
the
if
they
want
to
bring
it
up
to
another
future
community
meeting.
We
can
do
that
as
well.
B
My
views
are
following
when
we,
when
we
have
users,
are
heavily
contributing
to
a
project
like
well
error
in
this
case,
and
they
start
becoming
owners
of
a
functional
area
and
what
I
mean
by
functional
area
is
not
necessarily,
it
doesn't
have
to
be
an
area
where
you
contribute
code.
It
could
be
test
and
release,
for
example,
it
could
be
the
home
chart,
well,
that's
actually
code
or
it
could
be
something
along
the
lines
of
security.
B
We
want
to
recognize
those
individuals
and
very
likely
you
can
make
them
the
technical
leaders
of
that
area.
We
didn't
call
it
out
as
a
specific,
separate
community
role
in
this
case,
because
I
don't
feel
like
you
know
that
that's
necessary
as
individuals
and
contributors
are
kind
of
progressing
and
they
become
this
feature
leads
or
technical
leads
in
a
specific
area.
B
I
think
that
that
sets
them
up
very
well
on
a
path
towards
maintainer
ship
and,
as
they
consistently
maintain
the
level
of
contribution
to
the
project,
they'll
get
to
maintain
errs
in
no
time
so
that's
kind
of
the
thinking
behind
it.
So
first
you
demonstrate
that
technical
leadership
in
an
area
or
in
a
role
and
then
once
you've
consistently
shown
that
over
time
you
become
a
maintainer,
that's
kind
of
our
path,
so
we
didn't
have
to
call
out
technical
lead,
a
role
just
yet
because
we're
also
a
small
project.
B
C
Just
day,
I
largely
agree
with
you.
There
Michael,
my
I,
don't
think
we're
quite
ready
for
it.
I
think
the
next
logical
step
like
when
we
would
be
ready
for
it.
The
the
dividing
the
the
easy
dividing
line
would
be
along
repo
lines
like
plug-in,
repos
and
home
charge,
and
things
like
that
like
that.
That
makes
a
logical
dividing
line
for
me,
but
that
doesn't
mean
there
aren't
areas
within
core
Valero
that,
like
it
totally
makes
sense
that
there
would
be
specializations
within
the
widow
to
you.
So
absolutely.
B
And
and
we're
not
closing
the
door
here,
you
know,
none
of
these
decisions
are
final.
This
is
a
project
that
will
continuously
is
like
almost
like.
Oh,
we
don't
necessarily
have
project
reviews
and
and
and
in
coming
back
and
and
kind
of
thinking
about
what
they
did
in
every
release.
I'm
not
gonna
use
the
word
that
Jonas
likes
because
under
the
huge
fellow
of
those,
but
our
goal
here
is
to
continuously
make
minor
improvements
and
and
make
the
project
by
the
right.
B
So
when
the
time
comes
and
the
right
discussion
to
have
will
have
it
and
will
do
the
right
thing.
Jonas,
if
you
don't
mind,
go
to
the
maintainer
link
on
the
maintenance
really
quickly.
It's
an
under
super
majority.
There's
a
link
there
for
maintainer.
So
I
want
to
point
out
one
thing.
So,
for
example,
we
have
so
the
home
charred
area.
B
Sorry
I
didn't
know
what
that
was
linking
I
thought.
It
would
take
you
through
maintainer,
so
we
also
have
their
maintainer
z--,
which
are
the
leaders
of
Valero
today
they're
all
from
VMware,
but
it
doesn't
have
to
stay
that
way.
We
want
to
encourage
other
contributors
to
to
continue
get
involved
in
the
project,
and
you
know
when
the
community
recognizes
you
you,
your
name
will
be
up
here.
You
know
one
opponent.
One
last
thing
here
is
that
under
deployment,
Carlos
has
been
taking
a
lot
of
initiatives
around
the
home.
B
You
know
this
he's
not
the
maintainer,
but
he's
the
lead
for
deployment
so
kind
of
started
already
adopting
that
model.
That
know
my
course
I'm
gonna
lead,
you
know,
he's
the
technical
expert
for
hum
charge
and
and
deployment
and
packaging
for
for
Valera,
along
with
Kali
Sheila's
doing
a
lot
of
work.
Around
UX,
install
configuration,
CLI
and
all
the
time
comes.
Individuals
are
in
these
areas
that
may
not
be
maintained.
Errs,
you
know,
don't
make
the
jump
if,
if
they're,
if
it
makes
sense,
that's
it
for
me.
Thank
you.
E
You
would
I
think
that
was
very
valuable
for
the
ones
who
re,
but
so
the
maintainence
are
able
to
merge
to
have
like
a
pre
sort
of
like
a
you
know,
it's
an
editor.
If
there's
a
first
pass
and
then
will
come
and
take
that
into
account
and
because
I
think
I'm
maintainer
is
they
need
to
both
be
able
to
contribute
contribute
any
whatever.
It
is,
but
also
to
PR
reviews,
and
that
will
allow
us
to
assess
that
side
of
things
too.
E
In
addition
to
also
being
a
valued
contribution
to
have
that
first
pass
or
an
additional
pass
on
the
pompiers
I,
don't
know
how
how
that
would
work
in
practice
like
because
we
don't
use
BOTS
I,
don't
know
how
in
other
words,
but
I
would
love
to
see
people
doing,
helping
us
review
PRS,
and
maybe
we
can
promote
them
to
like
okay,
this
these
contributors
are
PR
reviewers
with
we
start
building
trust
that
they
understand
the
codebase.
They
understand
the
logic
and.
C
E
B
Yeah,
so
so
the
you
know,
obviously
some
companies
like
kubernetes
have
a
very
formalized
way
to
track
reviewers
versus
approvers,
and
you
know
that
that
makes
sense,
I
think
in
our
case
we
could
actually
do
this
without
formalizing.
The
roles
like,
for
example,
and
Dylan
I'm.
Sorry,
I'm
gonna
pick
on
you.
If
Dylan's
is
a
PR
and
he
loves
it,
he
can
actually
go
and
say:
hey.
You
know,
I
think
I
approve
this
changes,
I
looked
it
over
extensively
and
and
I
agree
with
it
and
that
reduces
the
burden.
B
For
example,
pick
on
someone
else
here,
Nolan
to
come
in
and
say:
hey
you
know,
I
mean
Dylan
has
been
reviewing
our
PR
she's
been
doing
great
job
later
he's
an
expert
in
this
area.
He
understands
this,
you
know
I,
don't
need
to
give
give
it.
You
know
the
you
know:
I
can
give
you
the
white
glove
treatment
for
this
for
this
PR
and
just
to
a
cursory
review
and
merge
it
in
because
Dylan
did
most
of
the
heavy
lifting.
We
could
do
that
Dylan.
B
You
know,
given
that
Dylan
doesn't
have
the
ability
to
merge
the
PR
he's
acting
almost
as
an
approver
in
a
in
an
unofficial
role
when
Dylan
does
it
on
five
ten
PRS,
a
hundred
PRS,
you
know
whatever
the
number
is.
Maybe
that's
the
jumping
point
that
we
say
you
know
Dylan.
You
know
he
knows
his
stuff.
He
knows
what
he's
doing
now,
it's
time
to
recognize
him
and
give
him
also
authority
on
top
of
basically
influence
yeah.
E
But
I
think
what
I'm
talking
about
I
take
part
until
he's
talking
about
is
making
explicit.
Why
not
make
a
role
for
it?
So
so
we
are
making
explicit
people
who
are
breeding.
That
document
understand
that
there
is
such
a
thing,
because
otherwise
that
is
sort
of
obvious
anybody
can
go
in
any
open
source
project
and
keep
their
inputs
on
PRS,
but
making
explicit
that
we
we
are
looking
for
that,
that
that
is
a
stepping
stone
to
be
a
maintainer
I.
Think
it's
valuable,
but
maybe
it's
just
a
hassle.
B
I'm
open
to
that,
if
someone
was
to
go
into
github
and
up-to-date
or
not
review
role
and
basically
added
into
the
owners,
thought
owners
filed
and
if
everybody
else
agree
said
I,
don't
know
what
team
or
or
you
know,
ashish
or
steve,
a
I
don't
have
any
concerns.
If
we
want
to
do
that
and
formalize
it
I'm
all
for
it,
but
I
think
also
the
thing
the
key
driver
is
at
the
point
in
time
that
we
are
these
reviewers
that
will
have
two
or
three
folks
that
are
ready
to
fill
that
role.
F
F
Guess
I,
don't
have
a
strong
opinion.
Whether
we
need
to
make
it
formal
at
this
time
like
write
it
down
in
technology
or
not,
so
it
either
way,
I
guess.
On
the
plus
side,
it's
it
shows
it
gives
potential
maintainer
x',
more
iterative
steps
in
their
path
to
become
maintainer,
so
I
think
that's
a
that's
a
big
benefit.
B
E
B
B
E
B
B
F
Just
going
to
say
the
other,
the
really
big
benefit
is
for
Valero.
The
the
core
maintainer
team
spends
a
pretty
significant
amount
of
their
time
doing
reviews,
and
so
please
consider
this
to
call
refer.
Reviewers
everyone,
that's
in
the
that's
in
the
call.
We
would
love
your
help,
we
would
love
to
have
you
be
more
involved,
it's
it's!
It's
fun,
but
everyone's
doing
a
lot
of
cool
stuff
and
so
takes
a
lot
of
time
to
look
at
it
and
we
love
to
open
the
tent.
A
C
So
the
last
PR
are
the
last
release
we
did
took
us
to
this
a
little
while
I
took
that
over
from
Steve
and
I
saw
some
places
where
we
could
just
like
put
all
these
steps
in
a
script.
So
I
worked
on
that
some
and
there's
there's
some
other
things.
I
want
to
talk
to
Steve
about
and
put
those
put
more
of
these
steps
into
scripts.
But
this
is
this
may
not
be
exactly
how
e
we
end
up
doing
it
and
I
know.
C
Ashish
was
talking
about
maybe
taking
the
output
of
this
and
putting
it
in
all
the
plug-in
repos.
So
we
can
just
do
or
putting
it
in
a
separate
repo.
That
just
does
this
for
all
the
all
the
repos,
so
we
could
run
the
same
process
for
all
of
our
repos,
so
we
can
release
all
of
these
things
with
minimal
effort.
C
The
next
item-
I
reviewed
some
community
PRS
I,
specifically
went
into
the
progress
on
backup/restore
operations
by
volume,
snapchatter
plugins
from
this
you
and
the
it's
really
good
stuff
in
there
I
had
I
had
some
questions
and
requests
for
clarification
on
this.
I
didn't
see
this.
You
want
to
call
earlier,
but
this
looks
pretty
good
to
me
so
at
this
point,
I'm
just
waiting
for
some
clarification.
D
So
last
week,
I
spent
moving
all
all
of
our
plugin
repos
to
go
modules.
We
were
using
depth
to
manage
dependencies
before,
but
now
they're
all
using
go
modules
and
then
the
other
thing
that
I
did
was
decommissioned
Travis
from
all
of
our
repos
for
both
CI
and
these
we
are
now
using
github
actions
to
do
tsehai
entities.
D
D
So
please
do
take
a
look
at
that
and
then
I
also
did
some
peer
reviews
like
last
evening
and
earlier
this
morning
and
I
added
some
feedback
on
the
restored
hook
proposal
and
also
the
volume
snapshot
from
this
report
proposal
that
was
out
there.
So
that's
pretty
much
everything
that
I
had
as
status
updates.
G
You
I'm
mute
sorry
about
that
hi
everyone.
So
last
week
we
talked
a
little
bit
about
the
upload
down
about
capability
that
we'd
like
to
add
into
the
PRS
for
bolero.
This
isn't
an
effort,
just
a
recap:
to
make
more
data-driven
decisions
about
what's
important
to
the
community
by
putting
the
capability
in
the
hands
of
the
community
to
vote
on
which
issues
you
feel
are
most
important.
G
So
this
week
we
plan
to
make
that
feature,
go
live
and
have
an
announcement,
I'll
coordinate
with
Jonas
to
get
that
up
on
Twitter
and
on
slack
so
that
everyone
knows
it's
available
and
how
to
use
it,
and
we're
really
looking
forward
to
seeing
the
higher
level
of
community
engagement
through
that
capability.
If
you
have
any
feedback,
any
questions
or
anything
that
you
feel
would
be
valuable
in
that
sense,
please,
just
let
us
know,
that's
it
for
me,
Jonas
thank.
A
A
E
Pretty
sure
I
was
on
community
support
last
week
and
did
a
catching
up
on
PR.
So
I
did
a
lot
of
peer
reviews
in
Friday.
It
was
pretty
much
a
personal
day
for
me
and
this
week,
I'm
working
on
addressing
the
reveals
on
the
cube
builder,
new
backup,
start
location,
resource
build
with
computer
and
controller
runtime
library,
and
that
is
as
a
part
of
the
discussion
that
we
heard
so
yeah
we
will
have
to
when
we
get
to
it.
But
that's
that's
what
I'm
doing
it's
a
it
would
be
the
beginning
of
involved
work.
E
Think
if
anything,
the
community
could
look
at
the
issue
that
I
added
to
the
discussion
list.
That's
very
tight
up
with
this
in
it's
more
like
a
gives,
a
higher
vision
of
why
this
even
is
even
there.
So,
let's
talk
about
it
during
the
discussion,
but
as
far
as
review
I
think
if
anybody
reviews
are
always
welcome,
please
but
yeah.
A
A
We
will
I've
added
a
topic
for
her
today
in
the
discussion
topics
there
I'm
going
to
talk
about
the
documentation
style
guide,
so
welcome
Abby
to
the
of
the
community
call
here,
and
then
we
have
a
bunch
of
other
folks
as
well.
So
we
have
Tony
batard
who
couldn't
join
us
today,
he's
prepping
for
the
C
CK,
a
exam
that
is
taking
in
just
a
few
hours,
so
he
will
be
doing
that.
A
We
have
my
critter
bow,
who
couldn't
join
either
today,
but
we
do
have
two
other
people
here:
Duffy
Cooley
we
have
mukesh
enter
so
welcome
and
these
four
lovely
people
will
help
us
with
a
few
specific
things.
So
the
first
task
at
hand
is
to
run
a
few
security
linters
and
things
like
that,
like
cossack
and
and
stuff
like
that
to
find
potential
issues
in
the
source
code
of
Valera
and
if
they
find
something
we'll
make
sure
that
we
make
issues
for
them
and
try
to
fix
them
before
they've
become
real
issues.
A
So
that's
the
the
first
part
of
this,
and
the
second
part
is.
They
will
also
help
us
upgrading
the
other
website
to
Jekyll,
4
or
potentially
also
making
a
you
go
template
of
the
website.
So
we
can
move
over
to
you
go
if
we
wanted
to.
So.
Thank
you
all
for
for
joining
today.
Super
super
happy
to
have
you.
A
E
E
Maybe
we
should
do
the
things
with
a
cube
builder
way,
which
I
wasn't
even
for
me
that
I
just
ran
into
the
Ames
test
library
and
they,
let
me
talk
you
builder
and
then
I
was
investigating
and
then
I
found
out
that
I
sheesh
had
suggested
that
we
converted
our
code
to
the
Pew
Builder
way.
I'm
saying
this
very,
like
hand-wavy
is
he'll,
be
other
ways
sort
of
like
a
handle
for
this
migration,
which
involves
there
are
multiple
parts
to
it.
E
One
part
is
running
things
which
could
be
generating
things
with
the
other,
in
the
other
part,
is
using
the
same
libraries.
There
are
offshoots
of
the
queue
builder
project
like
the
control
one
time
like
the
test,
libraries,
so
just
to
give
an
idea,
and
so
I
kept
checking
in
with
the
team,
and
everybody
says
yeah.
Let's
try
to
do
this,
so
I
did
convert
all
of
the
the
new
controller
was
going
to
be.
E
To
make
that
work
to
hook
up
to
the
server
and
make
out
that
will
work
in
the
big
question.
For
me
at
the
time
was:
are
we
going
to
be
able
to
run
these
two
things
in
parallel,
so
turns
out
that,
yes,
that
works?
The
APR
is
the
proof
of
concepts,
and
so
we
had
a
meeting
yesterday
and
we
were
evaluating
if
we
wanted
to
move
on
with
this
project,
because
that
PR
would
only
make
sense
to
go
in
if
we
wanted
to
eventually
convert
all
the
other
resources
and
controllers.
E
So,
while
we
do
this,
this
work
we
can
keep
both
AP
is
both
packages
side
by
side,
so
you
wouldn't
break
anything
for
anyone
as
we
convert
resource
per
resource
and
when
I
resource
and
controller
you
know
coupling.
So
we
would
do
one
by
one
and
when
took
2.0
2.0
comes
around
then
we'll
get
rid
of
the
package
that
we
have
now
and
use
the
new
API
package.
E
Now
that's
one
thing
and
not
so
so.
The
motivation
for
that
is,
it
has
a
lot
to
do
with
how
we
can
write
better
tests
and
how,
in
the
process
of
doing
this,
we
would
also
remove.
We
have
a
lot
of
logic
in
our
controllers.
You
know
controllers,
which
also
makes
the
test
harder
to
write
in
maintain
so
yeah.
We
would
do
the
work
of
removing
the
control
the
logic
out
of
the
controller
you
Steve
actually
started
doing
that
with
the
backup
logic.
Is
we
just
have
a
merge
that
yet?
E
E
So
the
next,
so
this
is
so
yeah.
So
this
is
what
it
is
a
large
undertaking.
It
is
something
that
we
need
to
plan
and
the
next
step
is
to
make
sure
the
products
that
Stephanie
and
Michael
understand
what
this
is
and
I
don't
board
on
it
with
it.
So
we
can't
so
we
can
do
the
work
and
the
next
step
will
be
to
get
a
PR
merged
and
then
to
work
on
the
the
Sushma
degree.
E
So
that
suggestion,
which
was
to
start
with
the
smallest
resource,
we
have
resource
slash
controller,
which
is
the
the
server
request,
and
in
that,
because
the
PR
that
I
have
now
it
sort
of
gives
me
an
idea,
but
I
was
also
learning
how
to
put
piece
it
all
together.
So
for
me
to
do
the
convert
after
that
PR
or
working
on
off
of
that
PR,
do
the
conversion
of
just
this
one
good
set
of
resource
and
controller
and
get
an
estimate
of
how
long
it
would
take
for
each
one.
We
have
ten.
E
Yeah,
so
we
have
ten
resources,
I
already
converted
the
backup
storage
location,
so
that
leaves
nine
so
we'll
do
that,
get
an
idea
and
then
playing
how
we
would
space
things
out
and
leaving
the
backup
and
restore
resources
for
less,
because
they
have
they're
very
involved.
They
are
the
like
comparatively
way
more
involved
in
the
others.
Yeah.
C
And
the
the
the
plan,
with
wit
back
of
the
store,
would
be
to
stever
steve's
already
got
a
PR
out
for
this,
but
it's
very
old
and
stale.
So
I
need
to
be
updated,
but
the
plan
would
be
to
pull
there
the
logic
for
a
lot
of
the
backup
and
restore
work.
The
the
work
inside
the
controller
out
of
the
controller
into
their
own
code,
so
that
updating
the
controller
themselves
would
hopefully
be
less
work
so
that
we
we
kicked,
that
we
kick
the
business
logic
of
backing
up
or
restoring
out
of
the
controller.
C
E
G
Hey
Carly
Co:
maybe
we
can
set
up
some
time
for
us
to
alternative
talk
the
doriot
a
little
bit
further
and
then
I
think
that
you
know
the
from
a
requirements.
Perspective
I
mean
I,
think
it
sounds
great
and
I'm
very
much
on
board
with
it,
but
I
would
be
interested
to
see
how
it
all
kind
of
pans
out
once
the
planning
before
the
work
has
been
done
and
we
kind
of
spread
it
across
the
backlog.
Yeah.
A
H
Hello,
everyone
excuse
me
so
what
I
wanted
to
bring
up
was
creating
a
documentation
style
guide
that
we
can
use
throughout
all
of
the
docs
to
make
sure
that
everything
stays
consistent
whatever.
So,
basically,
a
the
style
guide
itself
is
just
a
collection
of
the
decisions
you've
made
about
how
a
document
how
you
want
things
to
be
documented
throughout
your
documentation,
so
yeah.
H
Like
you
know
the
final
word
and
law
of
how
things
should
be
added.
It's
it's
really
a
guide.
You
know
more
like
guidelines,
you
can
break
them,
be
as
strict
as
you
decide.
You
want
to
be
with
them,
but
I
think
it
will
really
help
make
everything
consistent
and
have
like
a
more
more
cohesive
things
on
the
site,
so
I
just
wanted
to
bring
it
up.
A
E
Think
it's
super
useful.
We
all
get,
we
always
get
stuck
on.
Even
reviewing
we
don't
we
don't
have
a
holistic
tour,
the
sightings
we
do
this
way.
Should
we
recognize
that
there
are
multiple
ways
it
can
be.
Things
can
be
saying,
but
we
don't
have
a
holistic
to
go
to
go.
Look
at
it
say:
okay,
that's
just!
Let's
just
go
this
way
because
to
be
consistent,
consistent,
I,
think,
consistency
for
at
least
can
prioritize.
Consistency
will
already
be
awesome.
E
H
A
H
C
H
H
Yes,
so
I
noticed
this
something
Nolan
talked
about
earlier
about
the
release,
processes
and
I
know
that
there's
a
release,
checklist
and
I
wanted
to
ask
if
that
was
something
that
was.
Is
everything
in
their
release
documented
there
or
is
there
still
work
to
be
done
to
create
documentation,
and
did
you
need
help
with
that?
That's
something
I
can
help
with
so.
C
C
C
So
my
personal
goal
is
to
get
this
document
as
short
as
possible,
like
run,
run
a
couple
scripts
and
then
do
the
announcements
is
my
hope,
but
I
think
right
now
it's
a
document
patched
release
process
and
then
see
where
we're
at
from
there.
So
I
don't
think.
We
need
a
whole
lot
of
box
help
here,
other
than
clarifying
the
patch
release
process
with
Steve
and
then
yeah
going
from
there.
Okay.
A
I
Doesn't
save
lera
community
thanks
for
having
me
on
today,
I
guess
so
so
some
background
we're
trying
to
do
worse,
we're
storage
so
we're
a
distributed
decentralized
cloud
or
CEOs,
the
former
founder
CEO
of
docker
and
before
that
cluster.
What
we
do
is
we
create
object,
storage
in
a
way,
that's
similar
to
you
like,
say:
Airbnb.
We
encrypt
data
client-side
and
break
it
up
an
eraser
coat
it
into
a
tea
pieces
which
you
need
to
any
29
of
the
80
pieces
to
reconstitute
the
file
nodes
are
constantly
measured
for
uptime.
I
If
they
go
offline,
your
files
are
repaired,
and
so
we've
had
100%
durability
were
featured
in
TechCrunch
like
we're
just
kind
of
getting
started
with
some
open
source
community
integrations.
Recently
we
launched
an
integration
and
got
merged
upstream
into
our
clone,
which
opened
up
the
door
for
a
stick,
and
so
I
was
researching
kind
of
like
what
you
know.
Other
open
source
communities
we
could
tap
into
and
discovered
Valero
and
I.
Think
having
a
you
know,
easy
to
use.
I
Kubernetes
backup
is
something
that's
really
exciting,
with
tardigrade
as
everything
is
by
default,
distributed
globally
and
setting
up
to
replicate
across
u.s.
East
us
West,
Asia,
Europe,
etc.
Your
data
is
globally
available
and
it's
hot
and
it's
what
the
price
of
glacier
with
the
availability
of
Amazon
s3,
and
so
you
know
with
that
I
kind
of
wanted
to
introduce
myself
we're
gonna,
probably
work
on
on
the
plugin
process
and
see
what
we
can
do.
We
kind
of
have
two
options
to
integrate
with.
I
A
C
I
I
mean
I
think
any
kind
of
advice
on
what
kind
of
the
best
path
for
it
is
we're
thinking.
Rasik
makes
the
most
sense,
but
we've
looked
at
the
plug
in
process.
We
just
need
to
kind
of
do
the
to
the
work
of
building
and
testing
integration
and
then
submitting
kind
of
the
plug
in
request.
Mr.
Lee
I
should
be
aware
of
like
within
that
process,
or
anything
that
would
be
helpful.
Is
someone
who's
not
as
familiar
with
Flair
so
for
the.
C
Plug-In
there's
not
really
a
request.
You
would
have
to
do
right
now
and
there's
no
plugins
inside
of
the
lyric
or
they're
all
separated
github
repos
that
you
would.
You
would
ultimately
control
the
rustic
integration.
We
don't
really
have
a
way
to
be,
because
rustic
is
built
directly
into
the
Valero
core.
We
don't
have
a
great
way
of
communicating
with
plugins
with
rustic
right
now,
so
the
closest
you
can
get
with
the
rustic
at
the
moment
is
through
our
claim,
so
you
probably
would
have
to
go
through
the
backstreet
gateway.
There.
C
For
so
I
think
with
the
plug-in
I
think
at
the
moment
the
thing
that
you
would
get
closest
with
is
getting
a
plugin
for
objects,
store
for
the
kubernetes
metadata
and
then
uploading
snapshots
of
volume
data.
It
would
probably
still
be
through
rustic,
but
we
could
keep
talking
about
that.
That's
the
current
design,
that's
not
to
say
it
couldn't
change,
but
off
the
top
of
my
head.
I
think
that's
as
close
as
we
can
get
in
things.