►
From YouTube: 2022-02-28 Kubernetes Migration Working Group
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
Hey
everyone
so
welcome
to
the
february
28
kubernetes
migration
working
group.
I
will
kick
off
with
a
couple
of
things
that
we
have
done
in
the
last
few
weeks.
So
first
one
is
on
the
rate
limiting
redis
migration.
So
I
continue
to
progress
here.
We've
completed
the
required
redis
upgrade
and
we
also
have
a
migration
plan
that
will
enable
host
name
support
for
sentinels.
So
both
of
these
things
are
allow
around
enabling
the
hybrid
deployment
that
we'll
need,
as
we
go
through
to
move
everything
over
to
kubernetes.
A
Great
so
the
other
thing
that's
done
last
week,
mark
myself
dylan
met
up
to
discuss
the
testing
of
getaly
on
kubernetes,
and
we've
agreed
that
for
now
the
testing
of
a
of
well,
we
were
talking
about
testing
of
italy
service.
I
think
josh's
comment
of
testing
a
single
node
gidly
service
probably
makes
sense
as
the
smallest
piece.
A
We
discussed
the
fact
that
gitly
cluster
may
not
be
suitable
or
even
wanted
on
kubernetes.
So
we
shouldn't
focus
on
testing
that
right
now.
A
A
Yes,
okay,
cool,
okay!
Thank
you.
D
Gotcha
I
saw
I
was
looking
for
the
the
what
at
the
top
of
that
issue,
ability?
Okay,
it's
functionality
performance.
I
think
the
performance
one
is
we're
particularly
interested
in
that
on
kubernetes,
jason,
sc,
just
joined
he's
got
all
sorts
of
context
on
on
what
we've
seen
with
with
gideon
and
kubernetes.
But
I
know
that's
something
that
we're
definitely
wanting
to
understand.
You
know
what
the
workloads,
how
they
impact
the
infrastructure
and
whatnot.
E
E
Not
super
retail,
I
mean
there's
another
case
in
point
where
we
actually
just
have
a
customer.
That's
trying
to
do
extremely
large,
ci
workloads
and
I've
been
working
with
them,
not
in
kubernetes,
but
they
given
a
perfect
example
of
their
users
are
fine,
but
when
they
load
their
ci
pipelines
to
load
against
gitaly
and
prefect,
actually
results
in
the
entire
application
starting
to
slow
down.
E
A
Okay,
so
what's
happening
next
from
the
rate
limiting
redis
side,
we
are
continuing
to
test
our
planned
migration
path.
We
are
working
to
have
as
much
sort
of
automated
tooling
and
sort
of
understanding
the
observability
as
we
can
before
we
move
over
to
pre.
So
that's
in
progress
and
jason.
Do
you
want
to
verbalize
your
point.
E
A
Okay
and
then
possibly
related
to
our
lack
of
blockers.
Recently,
I
have
a
proposal
that
perhaps
we
consider
reducing
the
frequency
of
this
meeting
at
the
moment,
we're
focusing
in
on
the
platform
side
on
redis
migration,
which
seems
to
be
a
little
bit
less
like
dependent
on
other
teams,
and
we
also
have
the
italy
testing.
It
looks
like
we
have
a
path
forwards
and
I
think
that's
working
fairly
well
async.
So
does
anyone
have
any
kind
of
thoughts
on
on
moving
this
to
monthly.
A
A
Great
thanks
for
that-
and
I
think
you
know
if
we
go
into
migrating
a
service-
that's
more
has
more
dependencies
like
web
or
api
are
things
we
can
always
shift
the
frequency
back
up
if
needed,
but
it
sounds
good
for
now
cool
great,
that's
up.
Are
there
any
other
things
that
anybody
would
like
to
discuss
today.