►
From YouTube: CHAOSS Augur Workshop Jan 21 2023
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
D
C
A
B
I
think
I
should
share
the
screen
after
and
you
can
guide
me
through.
It.
A
Well,
either
we're
gonna
do
this
once
because
it's
a
brand
new
server,
so
you
can
guide
yourself
through
it
and
we
can
record
that
or
I
can
walk
through
it
and
we
can
afford
that.
It's
kind
of
your
call
and
which
of
those
we
do
so.
A
Do
you
want
me
to
walk
through
it
and
then
effectively
at
the
end
of
the
process?
If
I
do
it,
you
will
have
a
server.
That's
collecting
the
data
that
you
require
to
do
the
work
so
you'll
be
able
to
stop
and
start
this
instance
of
auger.
But
the
end
of
this
will
be
that
the
auger
instance
is
first
going
to
collect
all
of
the
data
that's
required
in
order
for
it
to
run
machine
learning.
A
A
Up
nonsense:
I'm
gonna,
I'm
gonna
set
up
the
instance
for
you,
but
the
reason
we
have
to
as
meat
pointed
out
the
reason
that
we
have
to
run
the
instance
before
you
can
do
the
machine.
Learning
Works
is
because
excuse
me,
so
you
can
see
like
in
this
database.
This
auger
instance
I've
got
1.4
million
because
it
Rose
I
have
300
000,
pull
request
files,
a
25,
000,
pull
requests,
2
000
releases.
A
Most
importantly,
pull
request.
Message
drops
to
get
35
000
of
those
issue
message:
ref
I've
got
36
000
of
those
messages,
I've
got
62
000,
and
so
that's
the
raw
data
that
you
need
in
order
to
do
clustering.
So
in
the
first
run
through
we're
just
going
to
get
the
data
that
you
need
to
then
use
to
do
the
clustering
work
on
a
second
run
through
okay.
B
B
C
A
B
A
A
A
A
Iowa
permission
here,
I
wonder
if
permissions
tonight,
because
I'm
already
in
it
and
it's
being
suspicious
of
that,
let
me
let
me
just
exit
my
terminal
session
and
try
to
command
again.
A
Not
accessible,
no.
C
A
Right
paste,
this
command
I,
don't
know
why
I
can't
see
it
we're
in
this
directory
paste
it
again
and
then
in
front
of
the
11.99b
put
dot,
slash
dot
forward,
slash.
A
B
A
A
I'm
able
to
do
it
so
now,
let
me
just
logs
are
recording
again
all
right.
So
the
issue
you
were
the
issue
you
encountered
was
I,
sent
you
this
command
without
putting
it
in
Unix
tick
marks
and
when
I,
sometimes
you
know
put
commands
in
the
tick
marks,
quotes,
use
a
different
utf-8
code
at
the
command
line
than
you
intend,
and
so
that's
probably
why
it
wasn't
working
earlier.
A
B
A
A
B
A
An
Amazon
ec2
instance
there's,
if
you
type
PS
space,
Dash
EF
you'll,
see
that
there
is
almost
nothing
right
here.
A
A
You're,
probably
also
going
to
find
that
this
has
pretty
good
internet
compared
to
your
regular
instance
to
do
with
postgres
yeah.
A
C
A
D
A
C
A
A
B
A
These
parameters
exist
so
that
when
the
worker
that
comps
commits
scans,
it
is
able
to
go
as
far
as
it
needs
to
go.
So
if
you
have
a
very
deep
Tree
in
your
repository,
this
ensures
that
it
works
when,
for
example,
somebody
doesn't
a
major
refactoring
and
every
file
in
the
repository
moves
what's
effectively
happening
is
in
a
single
process.
A
It's
renaming
all
of
the
files
and
when
you
do
a
get
pull
you
need
to
have
each
you
need
to
have
renames
set
to
true,
and
you
need
to
set
the
rename
mode
very
high
in
case
for
the
UN
I.
Guess
common
enough
that
we
need
to
do
this,
but
not
something
that
you'll
encounter
every
time
you
run
auditor,
but
anytime
somebody
who's
using
an
auger
repository
that
you've
installed
changes.
Everything
refactors
the
whole
repo,
then.
A
Yeah
and
I'm
not
sure
I
have
the
best
instructions
in
there.
For
that.
C
B
B
A
Yes,
that
is
correct,
so
what
I
do
you
can
follow?
Whatever
practice
you're
accustomed
to
in
the
user,
that
you're
in
right
now,
I
usually
create
a
GitHub
directory
and
then,
under
that
GitHub
directory
I'll
create
a
venv
directory
and
I'll?
Do
my
git
clone
of
auger
inside
the
GitHub
directory,
and
then
I'll
have
like
two
two
subdirectories
under
GitHub,
the
EnV
and
the
auger
clone?
It's
entirely
up
to
you.
What
strategy
that
you
use
or
where
you
put
your
virtual
environments,
got
virtual
envs
and
home
directory
is
common
as
well.
A
The
the
critical
thing
just
supports
it.
You
put
it
in
a
place
where
you'll
remember
that
it's
there
and
you
won't
come
back
to
this
computer
and
be
like.
Where
is
that.
B
So
inside
this
directory
directory
I
should
clone
the
repository
and
also
make
the
environment
correct.
Okay,
yeah.
B
And
also
like
every
time,
I
work,
I
I
get
off
work.
I
have
should
I,
you
know,
remove
the
I
should
exit
the
instance.
A
Let's,
let's
stop
a
minute
here:
I'm
gonna,
have
you
update
your
fork
of
auger,
because
I
just
did
a
merge
into
doubt
that
you
don't
have
so
what's
critical
here
is
making
sure
that
your
virtual
environment
is
where
you
you
will
remember
that
it
exists.
B
A
Me
know,
which
is
fine
I
just
I,
want
if
you
think
that
we're
seeing
what
you're
saying.
E
B
B
A
Yeah,
go
ahead:
the
command
Dash
m
space,
no,
oh
yeah!
So
now
you
do
python3
Dash
M,
and
then
you
just
get
rid
of
everything,
except
for
the
there.
No,
no,
no
leave
that
yeah!
So
the
dash
MV
EnV
is
actually
calling
the
python
module
vgnb
to
create
the
virtual
environment.
And
now
it's
going
to
do
that
inside
this
directory,
so
go
ahead
and
hit
enter.
A
A
E
A
Now
you're
ready
to
do
your
make
and
install
it.
Then
we
create
the
database
user
already
yeah
yeah
we
did
so
now
you
can.
You
should
be
able
to
install
and.
A
Actually
put
in
wheel,
it's
a
little
faster
if
you,
what
are
you
saying
this
command
that
you
have
here?
That's
helpful.
A
It's
the
the
speed,
is
linear
with
your
internet
connection
speed.
So
here
we're
on
ec2
server,
and
it's
going
really
fast.
If
you
have
a
severely
bandwidth,
Limited
internet
connection,
that
I
don't
know
what
you
would
consider,
it's
really
badly
included,
but
let's
say
30
Mig
makes
second
or
less
it's
going
to
take
a
lot
longer.
A
Well,
here,
you're
fine,
because
you
are
logged
into
a
computer,
that
is
an
Amazon's
data
center
in
London,
and
that's
just
gonna
go
super
fast,
okay,
but
if
you're
doing
this
locally,
this
process
and
you've
probably
experienced
this
at
this
process.
It
takes
much
longer
as
you
go
sort
of
fast
here,
and
this
is
this-
is
actually
the
fact
that
it's
building
that's
fine.
B
B
Before
closing,
my
PC
should
I
do
something
or
just
close
it
directly.
A
A
Yeah
I'll,
you
know
we
can
leave
it
run
for
a
month
or
two
okay,.
A
Yeah
yeah,
basically
and.
B
A
Here's
all
the
issue,
maybe
we
just
keep
it
boring
and
we're
working
on
different.
It's
you
know
using
these
suggestions,
it's
like
possible,
so.
A
A
Yeah
yeah,
it's!
Let's
do
that
so
we're
gonna
pause
the
installation
and
look
at
the
pull
requests
that
our
girl
has
I'm.
A
Third,
one,
the
third
one,
the
third
one
not
this
is
the
third
one.
Okay
I'll
be
a
feature:
yeah
repository
traffic.
D
C
A
Where,
in
order
for
auger
to
be
what's
called
item
potent
we
need
anytime,
there's
an
addition
to
the
table
as
there
is
in
this
case,
we
knew
there
would
be
an
update
to
the
database
script.
A
Yeah,
it's
it's
well,
the
the
I
guess
the
the
trick
is
actually
we
could
make
it
I
mean.
Let's
see
this
is
being
merged
into
Main
right
now,
and
then
we
could
make
that
I
could
make
the
changes
afterwards,
because
Maine
still
the
main
branch
still
only
has
two
I
believe
right.
E
I
guess
yes,
the
dead
brand
has
to
be.
E
E
Mean
we
have
a
Google
doc
we're
using
for
that
so
yeah
I'm,
going
through
that
draft
ER
that
we
have
enabling
rabbit
images
so
yeah.
A
A
B
A
Perhaps
safer
yeah.
Let
me
take
this
too
for
a
minute.
A
So
auger
news
and
that's
that
branch
has
sort
of
been
abandoned
now
that
we've
run
into
released
it.
Oh
okay,
okay,
it's
not
just
trying
to
think
of
the
best
way
to
do
this.
I
think
the
best
way
to
do
it.
This.
A
A
So
all
right,
so
this
is
fine
and
there's
no
risk
now,
because
I've
created
a
separate
branch,
foreign.
E
Yeah
I
mean
I
just
do
we
allow
push
permissions
as
well
when
we
are
making
up
the
token
generally.
A
E
E
A
All
right
is
there
and
then,
let's
see
if
I
do
call
get
stashed.
A
We'll
make
rebuild,
which
should
be
interesting,
because
this
re
this
instance
was
actually
running
with
mq
previously,
so
we
had
to
jump
to
mq
for
the
rabbitmq
we're
using
for
large-scale
data
collection
like
when
you
get
above
about
about
15
000
repos
redis
doesn't
actually
have
the
oomph
that
we
need
to
manage
that
much
work.
So,
okay,
so
now
we're
going
to
do
I've
rebuilt
it
I'm,
just
gonna,
make
sure
I've
killed
all
the
existing
processes.
E
Yeah
actually
Sean
I
wanted
to
ask
this.
One
question
specifically,
like
you
know,
when
we
are
testing
a
particular
worker,
is
it
possible
to
test
a
single
worker
or
do
we
have
to
again
rebuild
everything.
A
A
The
clue
to
how
you
might
test
an
individual
worker,
okay,
okay,
because
I
haven't
actually
obviously
I-
haven't
I'm,
not
running
now,
I'm,
just
tailing,
the
output
of
your
instance
to
make
sure
it
starts
well
yeah,
and
it's
going
through
everything,
so
I'll,
just
let
this
run
and
if
it
works,
then
I'll
do
a
pull
request
into
Denver,
main,
probably
Dev.
A
So
that's
part
of
the
next
release.
Okay
right,
first
I'll
make
sure
that
everything
runs
this
way
and
then
I'll
make
sure
that
it
runs
in.
E
Dev,
just
just
let
me
know
if
you
know
anything
breaks
down
absolutely.
A
Yeah,
absolutely
yeah,
I've
I've
merged
I've
closed
and
merged
to
pull
requests.
It
just
hasn't
yet
hit
the
main
branch.
So
you
won't
quote
unquote,
get
GitHub
credit
for
again
because
it
hasn't
been
merged
into
the
main
branch,
but
it
will
be
I'm
testing,
a
copy
of
the
main
branch
to
make
sure
it
works.
And
if
that
works
and
there's
no
problems,
I'll
just
merge
into
the
main
branch
and
then
we'll
test
it
and
do
a
release
of
auger
and
then
we'll
test.
A
It
wait
we'll
test
it
yeah
and
Dev
as
well.
So
then
I'll
do
I'll,
create
another
pull
request
where
I
change
the
name
of
your
bullying,
big
script
to
for
underscore
for.
A
Dev
one
okay,
one
way
or
the
other.
Something
obviously
will
have
to
change
so
yeah
right.
C
A
So
is
that
good
done
me.
A
Yeah
yeah,
thank
you
for
that
by
the
time
we're
done
here
this
morning
in
a
couple
hours
I
should
know
if
well
maybe
I
won't
actually
buy.
A
Maybe
I
just
actually
stupidly
started
it
pretty
large
set.
It
would
probably
be
smarter
if
I
tested
it
against
a
assuming
of
less
than
seven
thousand
repos
in
it.
E
A
A
it's
going
to
be
a
very
small
change.
Actually,
yeah
there'll
be
two
system
level
package
manager,
things
that
have
to
be
installed.
One
is
erlang
which,
for
whatever,
for
whatever
reason,
rabbitmq
doesn't
install
it
automatically
as
something
that
rather
than
the
Q
requires.
But
if
you
run
rabbitmq
without
first
installing
erlang
at
the
operating
system,
level,
rabbitmq
doesn't
work
and
it's
in
the
documentation.
But
what's
ambiguous
to
me
is
why
rabbitmq
hasn't
made
it
part
of
their
own
apt
install
to
go
get
erlang,
since
it
won't
really
work
without
it.
A
I
I
don't
have
an
answer
to
that.
I'm
sure
there's
a
reason
like.
Maybe
it's
a
package
used
by
other
packages
that
the
version
has
to
be
managed
with
is
the
most
likely
explanation
that
makes
sense
to
me.
There
could
also
be
a
stupid
explanation,
but
so
and
and
then
there's
one
line
that
you
have
to
add,
there's
a
there's,
a
script,
there's
a
script
that
I
will.
We
will
provide
for
creating
a
specific
mq
for
each
instance.
Okay,
Iran
right,
but
that's
we
have.
E
I
guess
like
if
you
need
some
help
with
writing
down
documentation
to
add
tasks.
I.
Can
you
know
help
with
that,
because
I've
just
added
that
Graphics
right
exactly
know
like
you
know
what
changes
are
to
be
made
and
where
the
editions
are
to
be
made
so
yeah
around
that
I
think
I'll
probably
do
okay.
A
Basically,
when
I
have
a
stable
update
like
I,
try
to
run
I
try
to
test
everything,
I
mean
I
run
it
so
we'll
run
all
of
our
tests
and
when
they
pass,
then
I'll
do
what
I
call
a
field
test
where
I'll
run
the
auger
on
three
different.
It's
got
sizes
of
repository,
and
that
takes
a
couple
of
days
just
to
make
sure
that
there
aren't
any
issues
that
emerge
at
scale
that
are
not
present
when
I'm
doing
what
I
would
like
200
repositories.
A
A
I
do
a
valley
like
15
000,
8,
000,
four
thousand.
A
Those
are,
those
are
some
of
the
bigger
collections.
Those
are
the
three
biggest
collections,
I've
done
and
then
there's
like
you
know
many
instances
that
have
like
a
thousand
repos.
E
Is
it
like
the
the
devtrans
that
we
have
now?
Is
it
like?
The
maintainer
is
the
new
previously
yeah
Dev
branch
is
the
new
auger.
A
A
A
A
B
A
Let
me
can
we
can
I
ask
if
we
can
take
like
a
five
minute
break.
A
A
But
let's
do
a
cat
space
now.
A
Let's
do
a
make
Let's
so
do
make
space
rebuild.
A
C
A
A
It
looks
like
auger
died,
yeah,
there's,
no
blogger
isn't
running
yeah
so
that
that
that
so,
let's
try
the
new
Hub
dot
out
running
it
again
coming
down
the
rebuild
the
the
command.
But
you
know
the
last
command
that
you
actually
need
to
do
both
I'm
going
to
start
the
flower
dashboard
and
the
know
how
to
help
one.
A
B
A
A
A
A
All
right,
this
has
actually
happened
before
in
the
world
I
see.
So
there
is
some
kind
of
binary
this.
So
the
reason
that
augers
we
say
in
our
documentation,
but
now
Ubuntu
22-
includes
python
3.10
by
default.
The
issue
with
python
3.10
is
that
the
machine
learning
libraries
are
not
all
updated
yet,
so
we
actually
have
to
deactivate
the
virtual
environment
that
you're
using.
D
A
D
C
C
D
C
A
I'm
just
going
to
share
with
you
so
that
Aaron
I'm
getting
is
I'm
trying
to
create
the
virtual
environment
and
install
the
python
3.9,
and
it
should
work,
but
I
can
install
it's
like
it's.
It's
the
same
when
I
try
to
install
it
when
I
try
to
create
a
virtual
environments,
telling
me
that
I,
it's
not
working.
A
Trying
to
create
this
I'm
actually
trying
to
create
the
median
v,
it's
giving
me
an
error
and
it's
not
letting
me
install
pip.
It's
so.
A
A
That's
okay,.
E
A
D
D
D
C
A
B
B
B
A
A
A
D
D
C
C
A
A
No
I
have
many
computers,
okay,.
B
A
A
A
All
right,
well,
I,
gotta
I
am
I'm
stuck
here,
so
I'll
leave
this
open
and
come
back
to
it
and
let
you
know
okay
I
am
stuck
I.
It's
like
my
brain
is
just
like
I'm
wound
into
this
problem,
and
I
have
no
idea
why
the
problem
I've
never
seen
this
issue
before
so
I
just
need
to
step
back
from
it,
because
I
have
no
idea.
What's
going
on.
Okay,.
A
A
Wait
I
guess
right
now.
You
just
have
to
wait
to
hear
from
me:
okay,
because
you
don't
have
any
data
yet!
Okay,.