►
From YouTube: Online Meetup #5 - March 5, 2020
Description
Learn more:
https://speakerdeck.com/soloio/online-meetup-number-5-march-5-2020
https://www.solo.io
https://slack.solo.io
Featured guest speakers this meetup are Kevin DeLorey and Brian Vandre from Frasca Interntational, Inc. Kevin DeLorey is a Software Architect and Brian Vandre is a Lead Software Engineer at Frasca working on their adoption of Kubernetes, Gloo, Linkerd, Prometheus, Grafana, and Sealed Secrets.
About Frasca International:
For more than 60 years Frasca International, Inc. has been building flight simulations of all levels for companies and flight schools all over the world. To combat the ongoing pilot shortage, Frasca’s current goal is to make flight training more accessible to instructors and students by modernizing our tech stack and starting with bringing our simulators to the cloud.
A
Be
live
on
the
audio
as
well.
This
is
this
is
our
fifth
online
Meetup,
our
first
for
this
year
we
have
just
a
couple
updates
policy
from
us
on
the
solo
side,
and
then
we
have
some
guest
speakers
from
the
community
to
talk
about
what
they're
doing
with
cloud
native
tools
glue
being
one
of
them
and
we're
very
excited
to
have
them
here.
A
So
first
up
we
have
the
blog
for
this
went
out
late
last
night,
I
believe
I
should
know,
I
published
it,
but
there's
a
glue
so
Christian
I,
don't
know
if
you
want
to
give
a
little
more
insight
into
what
these
are.
The
Seavey's
that
were
addressed
and
the
upgrades
that
folks
should
be
considering
making.
A
B
B
Definitely
ask
questions
so
we're
gonna
be
like
Betty
said:
we'll
have
some
end
users
discussing
some
of
their
their
journey,
but
I
think
we're
going
to
kick
off
here
with
a
little
bit
of
an
update.
So
you
can
see
on
the
screening
that
we
have.
We
have
paps
blue,
which
is
a
open
source,
API
gateway
built
on
monthly
to
do
the
latest
patch
releases
of
Envoy.
B
B
It
can
do
all
of
the
proxying
things
that
you
would
expect
like
load,
balancing
and
TLS
termination
and
rate
limiting
and
so
forth,
and
then
what
would
build
on
top
of
it,
the
API
gateway
part.
It
allows
you
to
abstract
away
differences
in
the
way
you
implement
your
API
in
a
particular
cluster
or
set
of
services
versus
how
you
expose
that
API
out
to
clients
that
might
want
to
consume
it,
and
you
keep
that
decoupled.
B
While
you
can
continue
writing
micro
services
and
cloud
native
is
about
moving
fast
and
making
changes
and
so
forth,
and-
and
you
want
to
do
that
without
and
without
affecting
enforcing
your
clients
to
change.
You
need
that
decoupling.
That's
that's
what
it
does,
but
a
glue
and
envoy
are
very
specifically
built
to
live
in
this
cloud
native
and
including
kubernetes
native
environment,.
A
A
Yes,
thanks
hunter
for
sharing
the
for
sharing
the
version
numbers,
but
come
on
sit
releases.
Yes,
we
are
we're,
always
always
always
be
shipping,
but
there's
four
different
versions
there.
If
you
have
any
issues
or
questions
as
you
all
upgrade
pop
into
the
slack
and
ask
away
you
know
we're
there
to
help
you
out,
and
so
today,
special.
Yes,
we
have
Kevin
and
Brian
from
Frasca.
A
They
do
they
build
flight
simulation
software
and
they're
going
to
talk
about
kind
of
what
they've
been
doing
from
from
adopting
kubernetes
and
glue
and
cloud
cloud
native
ecosystem
of
tools.
So
Kevin
and
Brian.
Do
you
want
to
start
by
introducing
yourselves-
and
you
know
a
little
bit
about
your
background
and
how
y'all
came
to
work
at
fresca.
D
B
C
C
C
C
C
C
C
B
A
C
C
C
C
C
A
C
D
C
D
But
traditionally
the
way
like
qualification
works
with
the
various
different
FAA
they'll
come
in
and
they'll
they'll
test
the
simulation,
and
then
they
kind
of
put
their
stamp
on
and
qualify.
And
technically
you
can't
really
like
change
the
software.
All
that
much
or
any
like
any
of
the
models
are
so
there's
not
really
a
lot
of
yeah
updates
and
the
models
once
it's
kind
of
approve.
C
C
C
We
next
or
the
one
inside
they'll
show
you
how
we're
going
to
think
that
so
yeah
we.
We
have
a
desire
to
kind
of
bring
the
simulation
of
the
cloud
for
a
few
reasons.
We
can't
talk
about
all
but
basically
want
to
bring,
looks
like
a
there's,
a
pilot,
soul
shortages
right
now,
and
we
want
to
try
to
come
back
that
a
little
bit
by
helping
you
bring
the
simulator
students
and
plan
structures
together,
be
in
the
cloud
and
that's
our
primary
injected.
C
C
Yeah,
basically,
our
software
will
still
our
simulation
will
still
live
on
the
device
itself.
Exo
simulator
we're
just
going
to
bridge
the
gap
where
we
take
that
simulation
and
like
feed
it
back
to
the
cloud
that
so
that
we
can
use
it
in
different
states
or
things
like
that,
and
essentially
people.
He
remotely
control
to
see
you
later
in
administrate,
the
simulator.
Thank
you
so
yeah,
so
we
have
within
the
cloud
or
bringing
clients
currently
running
on
a
loop
in
any
service.
Yes,
don't
take
Billy
reason
why?
C
C
C
B
C
Right
now
few
things
we
really
like
the
traffic
split
stuff
and
that's
kind
of
true
of
every
server
smashed,
but
it
seems
to
work
well,
CICP
system,
no
reason,
but
also
we're
like
in
developing
region.
So
we
don't
really
have
a
production
and
conserver,
but
just
for
debugging
purposes.
It's
like
really
convenient
to
just
like
buy
of
it
and
say
I'm
looking
at
this
server,
so
I
think
network.
So
this
route
what's
happening
and
it
just
like
works
yeah.
C
C
C
C
C
A
C
D
Our
it's,
our
our
old,
saw
our
software
and
now
simulation
software.
It
already
has
interfaces
for
manipulating
it
specifically.
It
was
developed
for
like
a
Nui
sitting
there,
so
kind
of
what
we've
done
is
left
that
all
alone
and
now
we're
having
a
little
piece
that
will
talk
to
that
interface
and
it's
basically
kind
of
forwarding
that
up
into
the
cloud
so
that
can
be
accessed
from
the
cloud.
It's
simplified
reasoning
there
in.
D
C
C
B
C
C
C
Yeah,
so
all
that's,
we
rewrote
our
Orchestrator
that
was
previously
in
our
simulation
black
box.
My
Orchestrator
now
handles
like
connecting
the
same
I'll,
fennekin
sim
with
the
cloud
that
the
cloud
returns,
some
JWT
tokens
back
and
the
orchestrator
pushes
those
out
to
the
software
component
sector
about
it.
C
C
So
yeah,
that's
all
goes
through
the
a
virtual
service
that
we
type
in
blue
and
grace
and
handles
all
the
simulation
specific
stuff,
and
it
is
configured
for
mutual
TLS
on
the
ingress
side,
so
the
orchestrator
will
actually
give
assert
to
glutes
and
hey.
This
is
go
ham
and
that's
forwarded
on
to
our
dinosaur
ear.
C
Yeah,
so
when
clue
that
glue
virtual
service
get
that
data
in
it
actually
pushes
it
up
to
a
server
service
called
right
here,
it's
called
a
simulation,
API
gateway,
but
I
think
ranae's,
something
about
that.
It
feeds
the
public
side
of
it
is
the
exact
same
thing
that
runs
on
the
simulation
on
the
left.
C
C
C
C
C
B
C
There's
so
all
of
our
services
are
based
the
GRP
security
enabled,
so
they
all
in
a
communicator,
the
GRC.
What's
a
handful
of
them
having
some
HTTP
like
this
is
designed
for
the
user
endpoints
we
do
like.
We
haven't
actually
done
even
practice,
I've
done
in
my
experiments,
but
we
do
like
the
ability,
like
translate,
Jeremy
C
to
HTTP
rest,
that's
kind
of
attractive
to
us.
C
C
D
All
handled
for
silikal
but
yeah,
so
we're
hoping
in
theory
like
if
we
ever
would
ever
have
to
you
know,
go
to
a
new
cluster
and
or
kind
of
thinking
we
might
have
to.
If,
if
they
like
go
to
Europe,
because
we
want
the
latency,
it's
a
pretty
pretty
nice
like,
we
might
need
clusters
in
different
areas
of
the
world.
So
we're
kind
of
open
that
we've
set
up
something
with
blue,
especially,
and
we
can
quickly
spin
up
new
clusters.
They
different.
D
B
Such
an
important
consideration
right,
where
your
source
of
truth
for
your
infrastructure
is
actually
in
in
document
like
yeah,
mole
files
or
code,
or
something
and
not
the
actual
deployment
itself,
and
so
when
we
build
glue
adhering
to
that
that
declarative
configuration
was
was,
you
know,
super
important.
It.
C
We'd,
like
the
idea
of
this,
is
very
important
to
us,
and
this
will
come
back
to
our
like
traditional
solutions
out.
There's
like
there
is
something
there's
some
history
and
I
can
go
to
and
say.
This
is
exactly
what's
running
in
the
cloud
right
or
in
this
case,
or
a
running
simulator
or
whatever.
So
having
that
source.
The
truth.
C
What's
going
on
that
topic
is
like
when
I
think
that
were
attractive
is
clear
for
me,
as
well
as
it
be
like
the
gamehole
definitions
and
speed.
My
were
just
a
to
ative
like
I,
can
get
in
there
and
change
it.
So
we
we
previously
like
we're
running
before
glue
right
in
the
nginx
egress
and
like
configuring
alive,
it
was
like.
C
C
B
A
D
C
So,
like
the
one
thing
was
helpful
and
the
docks
for
sure
was
that,
like
at
the
time
I
was
doing
this,
there
was
nice
instructions
for
how
to
like
set
up
a
building
wearing,
especially
this
is
pre
pre
piping
modules
and
go
so
I'm
gonna
go
expert
so
like
it
was.
It
was
nice
to
have
a
little
walkthrough
of
what
I'm
doing.
A
C
C
A
Okay,
let's
put
in
a
pin
in
that
we're
gonna
have
to
look
into
that
for
the
future,
but
also
I
have
tagged
in
here
linked
to
your
website,
because
there's
some
really
amazing
pictures
in
there
for
the
the
types
of
simulators
you
have
now
and
also
like
the
history
super
interesting.
Okay,
you,
like
those
giant,
you
know,
computers
that
take
up
the
entire
wall
nearly
days
and
then,
if.
D
A
Twitter
could
go,
has
a
lot
of
great
pics
from
I
thought
they
were
super
cool
because
there
not
you,
know
software
conferences
there,
but
but
you're
saying
it's
very
run-of-the-mill,
for
you
know
aviation
conferences,
but
you
know
having
it
one
of
those
simulates
the
simulators
in
your
show
floor
seemed
pretty
neat
to
me.
So
if
you
don't
want
to
kick
out
on
some
cool
aviation
photos,
there's
some
links
are
for
you.
C
A
Will
post
the
video
with
your
with
your
slack
handle
in
the
in
the
channel
so
that
people
can
find
you
so
yeah?
Thank
you,
Brian
and
Kevin.
Thank
you
so
much
for
taking
the
time
to
you
know
it
takes
time
to
prep
for
things
like
this
and
also
speak
and
I'm.
Sure
lots
of
folks
will
be
hitting
you
up
on
slack
and
for
those
online
with
us.
We'll
have.
We've
got
a
lot
of
upcoming
events,
most
of
them
virtual
online
events.
A
We
have
a
couple
of
like
a
hoot
session
next
next
week
or
the
week
after
and
then
another
chat
with
Christian
and
Nick
Jackson
of
a
Corp
continuing
their
service
mesh
for
the
developer
series
and
then
more
webinars
coming
so
lots
of
things,
opportunities
to
learn
and
dig
deeper.
If
there
are
topics
you
specifically
want
to
hear
about
just
damn
me
on
slack
and
then
we
can
get
those
on
the
schedule.
You
know
we'll
just
create
a
backlog
of
all
the
things
people
want
to
know
about
and
resource
wise.
A
If
you
go
to
there's
a
section
on
the
website
where
we
are
posting
a
lot
of
new
resources,
the
blog
has
moved
under
there.
So
it's
no
longer
on
medium
and
then
also
the
all
the
the
hoot
live
streams
that
we've
been
doing
to
dig
into
each
service
mesh
I've
kind
of
batch
them
all
together
into
one
page.
So
it's
easy
to
browse
and
then
all
of
niihka
and
christians
chats
are
in
one
page
as
well,
and
then,
if
you're,
not
in
our
slack,
there's
a
link
to
our
slack
there
and
then
also.