►
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
Well
quickly,
say
hello:
as
tracy
mentioned,
one
of
the
cd
foundation
ambassadors,
I'm
also
working
for
a
company
called
intelligent
growth
solutions,
I'm
quite
interested
in
utilizing
all
tellius
in
the
early
days
of
kind
of
start
up,
the
growth
so
quite
excited
to
see
what's
going
on
and
try
and
contribute
in
any
way
I
can
and
yeah
tracy
very
quickly
twisted.
My
arm,
I
missed,
didn't,
take.
B
B
And
if
owen
is
working
is
this
new
company,
starting,
it
is
really
fascinating,
they
do
vertical
gardening,
and
all
of
their
sensors
will
be
is
running
in
a
kubernetes
environment.
B
You
know,
I
think,
about
a
vertical
garden.
I
think
about
something
you
could
put
in
your
house,
but
this
is.
These
are
like
massive
warehouses.
It's
very,
very
fascinating.
A
So
it's
kind
of
industrial
scale
life
our
first
big
one
was
going
live
next
year
for
a
french
company.
It
will
produce
about
480
tonnes
of
produce
per
year,
and
yesterday
said,
all
the
software
side
is
based
on
kubernetes,
with
the
hardware
side,
just
really
being
a
connection
to
that
kubernetes
cluster
under.
B
Control,
hey
steve:
could
you
swipe
the
the
link
to
the
dock
and
put
it
in
the
chat
yep.
D
Josh
is
here:
hi
josh
greetings,
one
and
all
I'll
go
ahead
and
go
next,
so
josh
garverick.
I
am
a
cloud
solution
architect
for
well.
Cognizant
now
was
10th
magnitude.
The
acquisition
finally
closed
yesterday,
so
we're
officially
cognizant
today
do
a
lot
of
work
in
the
microsoft
partner
world
lots
of
azure.
D
D
Dealing
with
all
kinds
of
stuff
from
every
pass
component
you
can
think
of
to
is,
unfortunately,
in
some
cases
really
looking
forward
to
getting
involved
with
this,
because
I
think
that
the
the
premise
behind
having
this
kind
of
a
tool
is
really
really
important
and
something
that
I
think
I
mean
I
personally
would
have
benefited
from
a
couple
of
years
ago,
when
I
was
trying
to
do
a
digital
transformation
for
an
application
at
the
company
that
I
used
to
work
for
so
looking
forward
to
it
and
nice
to
meet
everyone.
B
We're
glad
to
have
you,
but
I
mean,
like
I
said
this,
this
group,
I
think,
is
going
to
be
a
very
interesting
group
of
for
kind
of
a
mind
share.
I
think
all
of
you
have
been
in
microservices
in
the
past
and
are
bringing
a
lot
of
experience
and
that
this
is
exactly
what
we
need
to
be
able
to
create
a
solution
that
can
be
used
across
that
can
automate
a
lot
of
this,
clarify
the
confusion
and
automate
the
reporting
that
we
need
to
have
that
that
visibility
viv
off.
E
Hey
hi
hi,
I'm
here,
so
I'm
I'm,
I'm
sorry!
I
didn't
attend
the
meet
him
yodele
yodel,
so
I
got
a
little
bit
busy
these
days.
Just
want
to
keep
me
at
the
same
page
of
the
others.
F
E
Okay,
sorry
so
my
my
name
is
steve
john
and
I
I
did
a
lot
of
coaching
and
mentoring
work
based
on
the
hydra,
and
now
I
work
with
capital
cabin
con
consulting
capital,
carbon
consulting
is
a
a
trading
company
to
help
other
people
want
to
get
devops,
certification
and
capital
capital.
Consulting
is
a
partner
of
devops
in
institute
for
myself.
E
I
basically
I
am
a
10
years
java
developer
and
I,
I
hope,
the
client
to
help
them
to
get
the
certification
and
also
I
help
them
to
to
plan
and
design
some
micro
services
at
architecture,
and
so
this
open
source,
I'm
very
interested,
and
I
I
would
like
to
spend
some
time
working
on
that.
B
And
it
and
interesting
to
note
that
he
he
worked
for
thoughtworks
for
some
time,
so
he
has
a
really
strong
background
in
you
know.
You
know
ci
and
the
pipeline.
E
Yeah
yeah,
I
I
worked
for
star
wars
about
four
years.
C
Here,
like
I
said,
sign
yourself
in,
if
you
don't
have
access
to
the
doc,
we
just
need
to
get
you
added
to
the
google
groups.
The
google
group
is
called
orteliusdev.
B
C
So
that
will
give
you
access
to
the
documentation
just
easier
for
us
to
manage
it
that
way
so
the
two
documents.
This
is
the
architecture
and
marketplace
working
group
architecture
being
around
the
architecture
of
ortelius
and
the
marketplace
is
a
microservice
micro
marketplace
that
we're
looking
at
providing
to
the
the
world
around
microservices.
C
So
it
may
seem
weird
that
architecture
microplace
are
kind
of
disjointed,
but
the
one's
going
to
feed
into
the
other
one
and
you'll
see
when
we
get
into
the
individual
documents.
So
here's
the
links,
if
you're
in
the
doc,
you
can
go
over
to
them
I'll,
bring
them
up
here
and
I'm
actually
going
to
do
it
backwards.
We're
going
to
look
at
the
the
micro
marketplace
first
and
our
goal.
C
The
idea
behind
the
marketplace
is
to
have
an
area
somewhere
we
can
go
to
find
microservices
get
the
information
about
the
microservice.
C
You
know
how
to
use
it
all
the
information
on
you
know
where
issue
tracking
is
for
it.
Who's.
The
owner
is
all
that
information
that
you
normally
would
need.
That
is
missing
in
things
like
the
docker
registries
and
the
helm,
chart
museums
these
like
the
helm,
chart
and
docker
registries.
They
do
have
a
a
place
where
you
can
search
for
things,
but
I
wouldn't
call
it
very
efficient.
C
They
do
have
some
tagging
that
you
can
do
to
help
organize,
but
what
we're
really
looking
at
is
focusing
on
domain
driven
design
patterns,
how
we
can
use
that
concept
to
pull
together
a
hierarchy
of
of
services
that
people
can
can
utilize
and
as
part
of
that,
when
ortilius
would
be
involved,
we
can
actually,
when
we
get
an
update
to
like
a
docker
registry
of
a
new
service,
we
should
be
able
to
grab
that
in
new
information
that
there's
a
new
service,
that's
been
created
out
there
and
that
in
turn,
is
going
to
affect
the
downstream
who's
consuming
it.
C
I
may
not
do
anything
with
it,
but
I
know
that
it's
out
there
and
that
there's
a
new
version
that
I
may
need
to
work
with
and
and
test
like.
I
said
on
the
kind
of
diving
into
the
requirements,
some
of
the
information
that
I
see
that's
missing
out
there
and
there's
probably
going
to
be
more
so
if
you,
when
you
guys
start
reading
through
this,
feel
free
to
add
and
comment
to
the
documents.
That's
why
they're
out
out
on
a
google
docs.
C
So
we
could
everybody
can
contribute,
but
some
of
the
things
that
I
think
that
are
useful
is
you
know,
where's
the
documentation.
You
know
the
high-level
documentation
to
the
service,
how
to
use
it,
what
its
purpose
is:
where's
the
source
code
link
to
find
the
service.
So
if
we
actually
need
to
go
in
and
see
what
the
service
is
doing,
it's
always
nice
to
be
able
to
look
at
the
the
source
code,
who's,
the
maintainer
who's.
The
author
of
it,
where
are
we
getting
issues?
C
Where
do
we
log
issues
for
the
service
you
know?
Originally,
you
know
what
ends
up
happening
is
because
at
the
end
user
standpoint
we're
delivering
applications,
so
I
may
be
delivering
you
know
the
artelias
application
to
a
customer
and
they'll
log,
an
issue
against
the
application,
but
that
application,
what
we,
when
we
start
doing
like
our
root
cause
analysis,
may
be
a
microservice,
that's
written
by
somebody
else.
That's
external
to
us
that
we
have
to
go
and
open
an
issue
too.
C
So
we
have
to
have
that
that
pathway
to
the
source
and
who's
maintaining
it
and
how
we
can
help
on
that
front.
B
Hey
steve,
could
you
chat
a
little
bit
about
on
the
kind
of
the
same
vein
here
the
problem
set
that
netflix
has
talked
to
us
about.
C
Yep,
let
me
get
through
this
and
I'll
talk
about
what
they're
they're
running
into
and
then
just
other
things
like
where's
the
registry
api
documentation.
You
know
the
swagger
and
then
an
interesting
one
that
is
kind
of.
C
I
threw
this
out
there,
I'm
not
quite
sure
where
how
we
would
implement
it,
because
it's
more
at
the
runtime
side,
instead
of
on
the
catalog
side,
but
being
able
to
find
where.
C
If
we
have
an
issue,
we
need
to
link
that
and
push
it
back
to
the
author
at
that
level,
so
that
part's
going
to
vary
because
service
is
going
to
be
consumed
by
thousands
of
people,
and
we
need
to
be
able
to
have
that
type
of
information
tied
to
on
the
consuming
side.
But
that's
not
necessarily
on
the
marketplace
side
and
then
also
we
always
get
into,
and
I
you
can
tell
it's
on
the
very
bottom,
because
I
forgot
about
it:
the
devsecops
area.
You
know
what
are
the
security
scans,
what
what's
the?
C
What's
the
vulnerabilities
cves,
you
know
licenses
that
the
service
is
using
and
then
potentially
service
levels
agreements.
What
what
are
those
those
may
be
again
tied
to
like
the
runtime
logs?
C
If
I
need
to
have
a
my
application
needs
to
be
24
7
with
a
99.9
percent
uptime
and
I'm
consuming
some
service
that
is
being
written
out
of
somebody's
garage
and
they
just
are
messing
around
it's,
not
a
good
fit
for
what
I
I
need
to
do,
some
other
of
the
other
pieces
on
the
the
cataloging
on
the
marketplace
we've
thrown
out.
There
is
like
how
to
rank
them.
C
Besides
the
domain
driven
design
in
organizing
it,
we
should
be
able
to
rank
them
and
find
out
things
like
which
ones
are
being
consumed
the
most
because
will
affect
service
level
agreements
and
possibly
issues
that
are,
we
need
to
address
stability
pop.
I
have
popularity
and
twice.
C
And
ease
of
use
in
maybe
uptime,
you
know
how
long?
How
long
has
this
been
around
and
running.
C
So
that's
kind
of
the
the
high
level
of
what
we're
looking
at
from
a
markets
place
perspective
now
tracy
brought
netflix
there's
another
piece
to
this.
C
I'm
not
sure
where
we
should
put
it
so
netflix
has
this
problem
where
they
will
create
a
microservice,
that's
dependent
upon
another
microservice
and
they're
tightly
coupled
so
they'll
have
a
group
of
five
or
six
microservices
that
need
to
be
tightly
coupled
and
versioned,
and
then
that
set
of
microservices
needs
to
be
moved
around,
and
so,
if
you
install
one,
you
have
to
install
the
other
ones
to
make
sure
you're
all
in
sync
or
or
things
are
going
to
go,
break
pretty
quick,
so
that
microservice
set
piece
is
another
thing
that
we
would
need
to
look
at
for
the
catalog.
C
I
don't
know
quite
know
where
we
would
how
we
would
describe
it
in
the
catalog
marketplace
view,
but
it's
definitely
something
that
from
a
domain-driven
design
perspective
those
patterns,
even
though
microservices
are
supposed
to
be
totally
independent.
It's
not
the
reality
in
into
your.
The
theoretical
world
do
diverge
there.
C
So
we
have
to
address
that
and
then
just
some
more
additional
information
about
the
the
the
marketplace:
we're
not
going
to
be
storing
the
microservices
themselves,
just
the
pointers
to
where
we
find
those
things
and
we'll
need
to
store
the
immutable
tags
like
the
the
image
digest
to
get
command.
So
we
can
get
back
to
the
single
source
of
truth.
C
So
we're
going
to
be,
like
I
said
more
of
the
the
aggregation
of
all
this
information
is
a
one
one
place
to
go
to
find
everything
about
your
servers,
so
we'll
have
all
these
pointers
different
places
to
handle.
What's
going
on
with
that
microservice,
then
another
thing
that
we
need
to
figure
out
is
in
the
domain
driven
design.
If
I
write
a
microscop,
microservice
called
single
sign-on
and
another
open
source
project
creates
one
called
microservice
called
single
sign-on.
C
C
Where
we
can
look
at
open
source
projects
that
are
creating
microservices
that
are
external
to
our
company,
that
we
want
to
consume
as
part
of
our
world
what
I've
been
working
with
some
of
the
the
national
labs
out
here
in
new
mexico
and
some
of
the
things
and
some
of
the
open
source
projects,
some
of
the
things
that
are
coming
about
is
normally
we
would
have
like
a
let's.
C
So
that
is
a
service
on
api
service.
That's
out
there
running.
What
we're
talking
about
is
taking
that
service
and
running
it
in
my
kubernetes
cluster,
where
this
comes
about
is
in
the
scientific
and
financial
arenas
where,
on
the
scientific
side,
we'll
get
like
sandia
national
labs
will
write
a
service
that
will
go
process.
C
Data
they'll
host
the
data,
but
they
don't
want
to
pay
for
the
kubernetes
cluster
and
be
worried
about
uptime
and
all
that
stuff.
So
what
they're
doing
is
they're
providing
the
service?
The
microservice
container
to
allow
you
to
run
it
inside
of
your
environment
and
then
they'll
link
back
to
their
data
store,
like
an
s3
bucket,
for
example.
C
So
that's
where
we're
seeing
these
universities,
laboratories,
financial
institutions
are
starting
to.
This
is
very
early
on
where
they're
going
to
start
putting
out
these
public
services
that
are
that
they're
going
to
let
you
run
inside
of
your
kubernetes
cluster
and
then
how
do
we
notify
people
that
there's
update
and
again
we
need
to
be
able
to
search
across
not
only
our
artilleries
installation.
So
this
is
easy.
C
If
you
have,
everybody
run
off
of
the
same
place
all
at
once,
but
if
we
have
like
my
ortiz
installation
and
then
we
have
our
assass
version
and
we
have
one
coming
from
los
alamos
national
lab
that
they've
running
one
of
theirs
that
we
should
be
able
to
concatenate
them
together
kind
of
like
the
rpm
repositories
for
the
linux.
C
So
that's
kind
of
the
idea.
So
any
questions
about
what
we're
trying
to
do.
C
And
you'll
you'll
see
where
this
drives
into
the
into
the
next
one
about
architecture.
C
Okay,
so
let's
go
ahead
and
read
through
the
documentation
and
make
any
comments
and
stuff
add
to
it.
This
is
just
the
first
pass.
You
won't
hurt
my
feelings
if
you
change
something
you
know
it's,
it's
gonna
be
one
of
these
things
that
we're
gonna
evolve.
So,
let's
go
over
to
the
architecture.
C
C
Just
to
give
you
some
background,
so
artelias
is,
is
currently
more
or
less
a
monolithic
architecture,
and
one
of
the
things
that
we
need
to
be
able
to
do
is
we
need
to
be
able
to
trans
form
it
to
a
a
microservice
implementation
and
as
part
of
that
path,
we'll
actually
move
from
a
monolithic
to
hybrid
and
eventually
we'll
get
to
a
pure
microservices
world
under
kubernetes.
C
The
today's
architecture
that
we
have
the
back
end
part
of
it,
is
all
written
in
java.
There's
some
jsps
out
there
as
well,
but
it
runs
under
tomcat.
The
restful
apis
are
all
served
up
as
one
of
the
the
servlets
under
the
the
tomcat
world.
The
front
end
is
there's
a
cut
like
I
said,
a
couple
jsp
pages,
but
the
front
end
part
is
99
javascript
based
on
jquery
and
there's
a
bunch
of
jquery
plugins.
C
C
The
is
the
true
jenkins
plug-in
is
a
java
plugin,
and
then
we
do
have
a
groovy
library
for
doing
the
the
pipelines
as
part
of
that,
and
then
we
have
a
deployment
engine
that
handles
incremental
deployments
and
figures
out,
roll
forward
roll
backs
and
that's
written
in
c
and
c,
plus
plus
so
that's
kind
of
what
the
the
stack
looks
like
right
now
when
we
deploy
you
can
download
a
single
docker
image,
and
most
of
you
have
done
this,
where
you
can
stand
up
and
that's
going
to
include
the
web
application,
the
top
cap
piece
and
the
deployment
engine
it's
all
kind
of
like
bundled
into
one
big,
docker
image.
C
I
think
it's
like
three
gig
right
now,
because
all
the
other
supporting
programs
we
put
in
there,
then
we
have
another
docker
image
for
our
documentation,
server,
which
we
run
under
hugo
and
doxy,
and
then
we
create
another
docker
image
for
the
cli,
which
is
basically
just
wrappering,
the
python
pi
pi
package
into
that.
C
We
do
that
just
to
make
the
dependencies
easy
as
far
as
that
and
then
when
we
host
it
on
a
sas
version,
we
do
split
out
these
guys
into
different,
smaller
images,
so
we
have
a
front
end
image
and
a
back
end,
and
then
we
have
an
nginx
for
routing.
C
That's
what
kind
of
weighted
it
is
today
the
proposed
architecture
that
I'm
throwing
out
there
right
now.
Initially,
we
want
to
take
new
functionality
and
start
piecing.
It
transitioning
the
new
functionality.
C
Instead
of
going
in
updating
the
monolith,
we'll
just
go
ahead
and
start
writing
new
microservices
and
bring
them
into
the
front
end
piece.
So
that's
where
we'll
actually
have
that
hybrid
world
for
a
while,
eventually,
I'd
like
to
see
us
get
to
where
the
front
end
is
also
broken
into
smaller
components
and
we
can
plug
and
play
at
the
front
end
level
with
component
web
web
components
for
front
design
work
which
we
can
push
into
microservices
and
reuse
there,
but
that's
way
down
the
road
as
part
of
that.
C
So
right
now,
looking
at
we
have
as
part
of
our
oktoberfest
there's.
I
think
five
or
six
reports
that
we
need
to
write
and
those
are
going
to
fit
perfectly
into
new
microservices,
because
they're
pretty
much
really
standalone
they're,
not
a
lot
of
dependencies
between
one
and
one
service
and
another
service
or
tied
to
the
the
monolith.
So
be
a
nice
little
transition
as
part
of
right.
C
Now,
I'm
thinking
because
one
of
the
the
people
that
stepped
up
trishiti
she
knows
python
and
basically
pi
flask.
So
that's
what
I'm
thinking
right
now
is
we
just
start
with
a
a
python
based
microservice
for
the
back
end
and
then
the
front
end
will
still
be
the
java
script
and
jquery.
C
We
won't
be
able
to
break
the
front
end
right
away
apart,
just
because
the
way
it
is
architected
it's
kind
of
tangled
together
too
tightly
at
this
point,
and
then
when
we
go
to
run
it
we'll
actually
create
an
image
for
the
front
end
an
image
for
the
the
back
end
and
then
we'll
do
one
per
microservice
to
throw
some
istio
routing
in
there
at
that
level.
C
Now
the
on-prem
install
I
this
one,
I
don't
know
how
we
should
address
it.
If
we
should
have
right
now,
people
can
run
just
a
docker
container.
They
can
get
up
and
going.
They
don't
have
to
worry
about
kubernetes
or
anything
like
that.
C
And
then
we
can
roll
forward
to
the
production
which
would
be
console.rto.io
and
we'll
either
use
ortilius
or
deploy
hubs
to
deploy.
Artelius
to
to
this,
the
this
kubernetes
cluster
that's
running
out
there.
I
was
trying
to
figure
out.
I
was
thinking
about
this
last
night
if
we
can
actually
have
artelias
deploy
to
ortelius,
while
it's
all
running
in
the
same
cluster,
so
deploy
back
to
itself.
I'm
not
sure.
C
If
that's
going
to
work
because
we
may
kill
the
the
pod,
that's
actually
doing
the
the
deployment,
so
we
may
have
to
break
out
and
just
have
deploy
hub.
Do
it
as
part
of
that,
so
that's
kind
of
the
the
idea
that
I
have.
I
I
think
one
of
the
other
languages
that
we
could
look
at
would
be
going
for.
One
of
these,
I
think
what
I
really
like
to
do
is
to
get
at
the
most
three
languages
out
there.
C
C
So
right
now
the
initial
ones
I
was
thinking
in
in
python,
and
maybe
we
can
throw
in
a
golang
one
as
well
to
get
started
there.
So
just
thoughts
on
this
and
you'll
see
because
we're
talking
about
this
microservice
architecture,
how
it's
going
to
affect
our
marketplace.
C
So
if
we
look
at
the
requirements
that
we
have,
you
know
finding
all
these
links
doing
all
that
stuff.
These
are
going
to
be
microservice.
Calls
at
that
level
as
well
being
able
to
you
know,
look
at
multiple
domains
and
multiple
installations,
that's
all
going
to
be
microservices
related
as
part
of
that
process.
C
So
you
can
see
where
the
marketplace
will
build
out
the
marketplace
based
on
our
architecture
that
we
decide
here.
B
Yeah,
I
think
that
we
need
to
have
some
discussion.
I
think
the
good
place
to
start
would
be
on
the
language.
Does
anybody
have
any
strong
feelings
about
what
languages
we
should
sort
of
standardize
on
or
if
we
need
to.
E
So
deep,
I
have
a
question
for
the
alternators
today's
architecture.
What
can
you
give
me
some
example
for
the
cri
python
3,
what
we
use
python
3.
C
So
this
is
I'll,
show
you
it's.
C
So
there
is
a
repository
out
here,
this
guy
okay,
so
this
guy
is
the
this
basically
the
cli
and
it
it
has
a
bunch
of
different
actions.
If
you
want
to
deploy
something
approve,
it
move,
it
update
information
about
a
component
assign
a
component
version
to
an
application
version,
deal
with
key
value
pairs
for
configuration,
management,
import
and
export
domains.
C
So
this
is
the
python
script
for
the
cli,
and
then
it
is
leveraging
a
python
library
of
of
restful
api
calls
basically.
So
this
is
the
the
client
side
of
the
cli.
So
like
there's
the
login,
if
we're
going
to
deploy
an
application,
those
type
of
things-
this
is
all
the
cli.
The
restful
api
calls
that
are
being
made.
So
this
is
all
in
in
python,.
C
So
we've
everything
in
is
in
python
3.,
with
sasha's
help.
We
were
able
to
get
the
containers
up
to
the
latest
version
of
fedora,
which
has
which
has
dropped
the
python
2
7.
So
everything's
python
3.
threw
a
couple
kinks
in
that
process,
but
he
figured
it
out
so
we're
all
on
we're
all
good
on
python3.
E
Okay,
okay,
thank
you
and
another
thought.
Another
suggestion
is
about
the
on-prem
thing,
so
my
stature
is
how
about
we
build
some
automation,
scripts
on
aws
or
and
on
our
adroll.
So
that
means
if
people
want
to
set
up
an
your
moment,
they
just
run
the
script
and
everything
will
be
done
in
our
best,
so
we
gave
them
two
kinds
of
work
and
another
one.
I
get
a
comment
of
ayasha
that
that
is
also
a
good
point.
C
We
would
we
use
something
like
terraform
to
stand,
something
like
that
up.
C
F
C
Well,
that's
the
way
we
ship
it
today,
it's
just
a
very
big
image
that
you
volume
mount
the
database
and
it
has
the
postgres
database
engine
in
it.
Tomcat
the
deployment
engine
are
all
kind
of
smashed
together
into
that
single
image,
so
the
command
line
to
run
it
is
just
a
single
command
line
with
a
volume
mount.
So
it's
really
quick
and
easy
for
people
to
get
up
and
running
to
kick
the
tires,
but
for
a
more
of
a
production
world
you
want
to
put
it
into
kubernetes.
A
Really
important
there
as
far
as
the
different
tiers
of
on-prem,
because
someone
who's
just
looking
to
explore
it
and
have
a
play
around
it.
The
barrier
of
aws
and
the
potential
costs
of
that
is
fairly
minimal.
It's
probably
worth
avoiding
versus
natural
on-prem
production
deployment,
so
we've
probably
got
those
two
different
tiers
and
then
the
sass
solution
as
well,
which.
C
G
G
C
Yeah,
so
on
the
when
we
like
on
the
sas
version,
and
it's
it's
really
a
bad,
I
should
really
just
say:
hosted
version
because
you
can.
We
have
a
unisys
actually
runs
the
deploy
hub,
sas
version
in
their
environment
instead
of
using
our
gke
environment.
So
you
could
take
the
the
sas
architecture
and
you
can
run
it
on
your
own
own
cloud
and
that's
what
unisys
did
for
that.
So
basically,
this
is
sas
equals
kubernetes.
C
Wherever
it
is,
you
know
it
doesn't
matter.
If
it's
on
on
on-prem,
you
know
it,
it
could
be
their
own,
you
know
an
open
shift
or
it
could
be
on
their
aws
install,
but
that's
kind
of
the.
C
Idea
and
then,
when
we
get
into
these
these
two
versions,
then
we
have
to
deal
with
where
the
database
is
because
the
database,
when
we
bring
it
up
as
just
a
single
docker
image,
we're
just
doing
a
volume
mount
to
the
postgres
database,
where
these
guys
actually
are
using
a
separate
container
for
running
the
database
itself.
So
postgres
in
like
when
we
run
or
deploy
hub
on
gke,
we
have
postgres
as
a
separate
container
and
a
pg
bouncer
as
a
entry
point
to
do
the
transaction.
C
So
in
as
part
of
it,
we
can
kind
of
like
what
I'm
saying
here
is
and
josh
helps
stand.
Part
of
this
up
right
now
is
getting
ortulius
up
and
running
on
azure
and
that's
my
goal
for
this
week
is
to
take
what
we
have
today,
the
monolith
part,
and
so
these
two
pieces
here.
C
Actually,
if
I
go
back
up
to
the
top
here,
I'll,
take
these
these
pieces
here
and
I'll
push
them
over
to
to
azure.
So
we
have
a
place
that
we
can
get
some
data
in
there.
People
can
start
testing
doing
those
pieces
as
part
of
that,
and
then
we
can
start
transitioning
that
into
adding
in
istio
and
adding
in
the
different
services.
C
So
some
of
the
things
that
we'll
need
here
in
the
the
short
term
is
getting
istio
installed,
working
with
some
basic
service
mesh
routing
at
that
level,.
C
B
C
B
H
C
I
C
B
F
The
other
part
is
a
lot
more
ghouling,
especially
if
you're
looking
like
operator
development
and
things
like
that.
But
I
mean
with
the
microservices
approach.
We
want
to
just
use
python
and
every
once,
while
do
something
a
little
different
for
a
use
case.
I
mean
we
could
probably
do
that
too.
If
we
wanted
to.
C
Yeah,
because
one
of
the
things
I'd
like
to
be
able
to
do
is
it's
kind
of
like
a
a
secondary
motivation
is
to
because
most
people
don't
know
how
to
write
microservices
and
what
they
look
like
and
get
something
bigger
than
a
hello
world.
C
C
A
Hey
so
so,
to
be
honest
day
today,
I'm
mainly
using
gonang
okay
nowadays,
but
I
do
see
a
lot
of
golang
and
python
in
use.
So
I
think
both
are
pretty
pretty
valid
and
a
lot
of
teams.
I've
worked
with
over
the
last
few
years,
even
if
they're,
using
something
like
golang
or
even
if
they're,
using
something
like
java.
They
still
have
some
python
code
in
there
for
a
variety
of
reasons.
So
I
think
it's
a
good
common
denominator
to
start
with
and
then
potentially
go
nang
all
even
later
downline.
A
If
we're
looking
at
examples,
I'm
seeing
quite
a
few
teams
using
net
core
as
well,
so
once
we're
a
bit
more
stabilized
being
able
to
bring
a
few
working
examples
of
other
languages
like
that,
I
think
will
be
really
powerful.
C
Yeah
yeah
yeah,
I
noticed
the
our
our
hipster
store.
Example
came
out
of
google
that
that
has
a
I've
seen
a
couple
dot
net
core
services
in
there.
So
it's
it's
interesting
world
to
explore.
E
Basically,
I'm
I'm
a
java
guy,
but
python
is
good.
I
I
also
did
a
little
a
little
bit
python
code,
so
I
think
we
can
start
with
python.
I
know
also
one
one:
one
question
python
2
python
2.7
and
after
python
3
they
are
a
little
bit
different.
So
what
do
you
prefer
on
three
or
two.
E
C
C
Yep,
that's
the
one,
that's
the
one
I
prefer
to
use,
because
it's
a
lot
of
everybody's
finally
moving
up
and
they
skipped
over
some
of
the
earlier
versions
of
three
so
that
that
one,
the
3.8.6
seems
to
be
a
good,
stable
version
that
is
available
now.
E
C
Right
now,
yes,
just
because
the
way
the
front
end
was
architected,
it
basically
is.
C
It's
not
the
best
design.
It
came
out
of
some
developers
I
had
in
the
uk
and
they
were
not
front
end
developers,
so
it's
kind
of
a
mess
over
on
the
front
end
side.
So
we'll
need
to
break
that
apart.
C
There's
no
like
framework
underneath
it
like
like
view
or
react,
or
anything
like
that,
so
it's
pure
pure
javascript,
some,
like
I
said
some
jsp
and
mostly
around
jquery
plugins,
so
we'll
have
to
see
where
we
can
start
splitting
apart
the
front
end
and
figure
out
how
we
can
maybe
like
these
reports
right
now.
C
Instead
of
embedding
the
reports
into
the
existing
ui
that
we
create
another
page
for
the
reports
that
we
link
to
at
that
level,
so
that
may
be
an
option
to
and
that
that
would
be
a
whole
new
front
end
that
we'd
stand
up
from
scratch.
You
know
like
using
reactoring,
probably
react
something
like
that.
C
What
I
really
like
to
do
is
like
on
the
front
end,
is,
is
really
focus
on.
If
we're
going
to
go
somewhere,
new
is
web-based
components,
so
the
concept
behind
web-based
components
is
I'm
trying
to
remember
the
name
of
the
project.
C
Oh
polymer,
polymer
out
of
google,
is
where
one
of
the
ideas,
but
basically
you
put
your
div
in
there
and
there
and
you
can
tag
your
div
with
a
with
basically
a
widget
name
and
then,
when
you
go
to
fetch
the
front
end
makes
a
call
to
the
back
end
and
says
I
need
you
to
fill
in
this
div.
The
back
end
actually
sends
html
code
in
to
expand
and
replace
into
that
div,
and
it
has
to
do
with
the
the
dom
and
the
shadow.
C
C
Quick,
so
if
you
think
about
like
a
calendar,
widget
or
you'll,
see
here
in
deploy
hub
because
I'm
doing
a
screen
recording
it's
running
slow.
C
So
if
we
go
into
like
this
guy
here,
each
one
of
these
boxes
would
be
its
own
web-based
component,
and
so,
when
we
lay
out
the
page,
we
just
have
our
seven
divs
here
and
the
back
end
would
provide
us
the
html
to
pop
into
place.
As
part
of
that,
then
you
have
a
css
to
style.
It
that's
unique
for
each
each
website,
but
that's
look
up
web-based
component
web-based
components
along
with
the
polymer
project
and
you
get
an
idea
of
what
I'm
thinking
of.
On
that
front.
C
C
If
this
is
being
served
up
just
this
one
box
is
being
served
up
by
a
microservice,
we
can
have
a
qa
version
of
it.
That
may
add
in
our
github
owner
as
another
detail
and
because
of
istio
we
can
route
the
q8
testers
to
the
new
version
and
they'll
watch
they'll
bring
up
the
the
new
box
with
the
the
new
field
on
it,
and
everybody
else's
production
would
get
routed
through
istio
to
the
old
version
of
that
service.
C
So
it
allows
us
to
really
mix
and
match
basically
makes
your
your
web
page
plug
and
play
type
of
scenario,
but
I've
diverged.
This
is
way
down
the
road
as
part
of
that.
But
it's
something
to
think
about
from
an
architecture
perspective.
E
Okay,
okay,
so
I'll
touch
so
pop
polymer
you
mean
use.
Is
this
one.
H
That
has
the
the
samples
around
it.
Let
me
see.
C
So
it's
it's
based
more
off
of
it
does
says,
as
some
more
of
a
node.js
feel
to
it,
but
and
there's
a
whole
data
binding
piece
and
stuff
like
that.
But
this
is
one
of
them.
E
C
I'll
put
this
in
the
in
our
document.
C
But
it's
based
on
these
different,
oh
material,
web
components,
is
the
other
piece
of
it
and
the
interesting
part
about
this.
When
you
look
at
like
these
web-based
components,
google
spent
tons
of
time
trying
to
figure
out
how
something
should
be
spaced.
You
know
that,
like
this,
this
tag
here,
how
many
pixels
away
should
the
icon
be
from
the
edge
of
the
box.
They've
spent
a
shitload
of
time
on
these
components
as
part
of
that,
but
material
design
is
where
I
think
down.
C
And
you
you
can
plug
it
into
like
view
and
react,
and
some
of
those
frameworks
to
make
it
work.
J
C
All
right,
so
anybody
else
have
any
comments
on
a
programming
language.
D
For
what
it's
worth,
I'm
a
c-sharp
guy
for
quite
some
time
have
been
for
quite
some
time,
dotnation.net
core.
So
when
we
get
to
the
point
where
we
look
at
that
net
core
for
anything,
if
we
do
I'm
happy
to
help
out
there
python's
fine
with
me,
though,
okay.
B
Yeah
that
I
think
it'll
take
us
a
while
to
get
rid
of
that
back
and
ninja,
because
there
are
people
who
use
it
for
managing
database
updates
and
it
does
a
better
job
than
a
script
of
doing
the
rollbacks
and
back
outs
between
for
database
updates
built
into
like,
and
I
have
a
suspect.
I
have
a
suspicion
that,
as
people
move
to
a
poly
kind
of
a
poly
database
format
that
the
database
will
be
more
of
an
issue
and
needs
to
be
included
in
the
configuration.
C
Yeah
and
right
now
I
did
not
put
in
anything
about
us
changing
the
database
up
drastically
at
this
point.
It
may
make
sense
down
the
road
to
do
some
database
transitions
at
that
point,
to
make
things
easier
for
the
the
micro
services
in
the
front
end
we'll
cross
that
bridge
when
we
get
there.
B
B
Would
do
that
or
he
said
swagger
doesn't
doesn't
solve
it
for
him,
so
there
may
be
something
else,
and
because
we
have
been
asked
about
producing
information
on
a
service
to
service
dependency,
which
we
can
do
is
just
a
it's
just
some
queries
we
have
to.
You
know,
find
all
the
applications
that
they
that
I
think
they
share.
That's
one
way
we
can
do
it,
but
service
to
service
dependencies
is
an
area
that
we
we
should
be
thinking
about.
B
C
Okay,
I
think
that
is
what
I
wanted
to
cover
so
you'll,
see
out
on
since
we're
in
october.
C
C
You'll
see
that
we
created
a
bunch
of
issues
out
here
and
they're
going
to
be
tagged
with
a
hacktoberfest,
so
the
report
ones
are
these
down
here
as
part
of
it
and
there's
also
a
landing
page
now
for
the
hector
fest.
B
C
So
just
what
we're
looking
at-
and
these
are
the
the
repos
and
the
issues
and
stuff
that
we're
gonna
get
started
with
with
the
microservices
and
in
python.
C
So
just
let
me
know,
go
ahead
and
get
if
you
don't
have
access
to
the
repository
or
if
you
can't
assign
yourself
to
an
issue
that
you
want
to
work
on.
Some
of
these
are
pretty
general,
so
they'll
be
multiple
pieces
that
need
to
come
into
play
on
this.
So
if
we
have
more
than
one
person
assigned
to
it,
that's
no
problem
again.
C
Let
me
know
if
you're
having
problems
with
getting
access
to
any
of
the
repositories
need
to
and
preajot
can
help
you
out,
they're
going
to
be
doing
the
project
management
side
of
things
so
they're
the
go-to
people
as
well.
If
you
need
need
to
get
access
anything
find
out,
you
have
a
dependency
upon
somebody
else,
those
type
of
things
we're
early
on,
so
we
still
have
to
build
out
a
lot
of
this
into
finer
grain
detail.
C
C
And
mark
did
to
get
our
did
everybody
see
the
new
logo.