►
From YouTube: 2022 07 05 Git Cache Maintenance
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
B
A
B
Oh,
I
was
thinking
today's
meet.
We
would
do
like
you
know,
find
the
git
versions.
B
A
All
right!
Well,
so,
let's,
let's
be
sure
that
you
have
access!
My
apologies
that
I
blocked
your
work.
Let's
be
sure
that
you
have
access
and
we'll
do
that
during
today's
session.
So
tell
me
again
the
username
you
prefer
I
added
one
username,
but
if
that's
username
that
isn't
the
comfortable
one
that
you
that
you're
accustomed
to
tell
me
the
username
you
prefer
and
I'll
create
that.
B
B
Rather
than
no
creating
username,
can
we
like
just
figure
it
out
on
your
computer
right
now,
because
the
idealized
there
are
only
three
maintenance
tasks,
version
which
I
would
be
needing
because
because
prefetch
isn't
supported
for
versions
less
than
2.30
according
to
like
what
I
have
seen
and
lose
objects
also
isn't
supported.
So
the
other
commands
which
we
would
be
needing
are
you
know,
comic
graph,
gc
and
multi-pack
index.
A
A
You
should
see
my
screen
now.
It
should
have
a
web
browser
on
it,
yeah,
okay,
good
all
right.
So,
let's
go
to
and
prior
to
the
meeting,
I
had
prepped
by
starting
a
wide
collection
of
agents,
so
we've
got
centos
7,
red
hat
8
oracle,
linux,
8,
debian,
11,
debian,
10,
freebsd,
13,
suzy,
15,
ubuntu,
18,
20,
21
and
22
freebsd,
12
windows,
ubuntu,
something
on
powerpc
and
ubuntu,
something
else
on
system390
and
a
bunch
of
of
others,
raspberry,
pi's,
etc.
So
we've
got
a
a
pretty
reason:
oh
and
debian
testing.
A
A
A
And
this
thing
and
all
we
did,
if
I
remember
right,
was
we
copied
this
and
made
ourselves
a
a
job
that
would
answer
the
questions
we
wanted
so
that
we
could
iterate
on
it
right.
B
A
A
A
A
A
B
There
was
on
another
way
of
you
know
going
through
this
process
like
implementing
this.
I
was
thinking
you
know.
Rather
than
using
the
versions,
I
can
call
each
command
using
the
get
client.
You
know
tool
and
if
it
throws
an
exception,
then
I
can
figure
it
out
myself.
Whether
I
can
you
know
you
run
that
maintenance
task
or
not,
so
I'm
not
sure
which
is
the
better
way.
So.
A
And
and
either
either
would
be
fine
I've,
because
the
the
knowledge
of
which
versions
implement,
which
come
which
which
commands
is
static.
I've
rather
preferred
to
go
the
direction
of
having
a
table
that
tells
me
or
having
code
logic
in
the
plug-in.
That
tells
me
which
thing
is,
is
implemented
and
which
is
not,
but
but
I'm
open
to
either
that
that
would
your
your
technique
would
also
be
fine.
A
C
A
A
C
A
C
A
C
C
B
C
A
I
mean
it
could
be
taught
to
know
about
versions,
but
cli.
Really
there
isn't
a
version
concept
on
jet,
so
I
had
assumed
that
it
that
cli
get
was
a
reasonable
place
to
put
the
version
knowledge
but
but-
and
it
certainly
is
cached
there
in
the
sense
that
any
time
I
instantiate
a
cli
get
api
implementation,
it
will
call
this
once
and
then
future
use
of
that
instance
will
will
remember
the
cached
value.
B
Yeah
because
I
added
or
you
know,
get
I
added
the
logic
to
get
the
get
version,
and
this
thing
runs.
You
know
whenever
I'm
executing
on
maintenance
stars.
So
whenever
I'm
going
to
schedule
a
maintenance
task,
when
it's
added
into
the
queue
and
when
I'm
removing
it
from
the
queue,
then
I
go
get
the
git
version
like
what
exactly
is
the
version?
And
then
you
know
execute
the
task
based
on
the
version.
C
B
Here
we
decided
to
go
with
the
default
tool
right,
like
I,
and
mark
we
discussed
in
the
previous
meet
like
yeah.
C
B
Okay,
wait.
I
have
a
doubt
right
now:
okay,
after
this,
like
the
version
which
the
so
the
logic
which
I
have
written,
I'm
not
sure
if
that
is
the
same
tool
which
I
am
getting,
is
I'm
not
sure
whether
that
thing
gives
the
default
get
version
present
on
jenkins
controller,
because
I've
taken
it
from
a
test
file
like
and
then
I've
added
that
logic.
So
I'm
not
sure
if
that's
gonna
return
the
default
tool
which
we
are
using
in.
A
A
Okay,
so
then,
let's,
let's
take
a
look
at
it.
What
we
see
now
is
here's
the
output.
It
shows
us
each
each
of
these
is
what
it's
running
now.
We
probably
also
need
to
know
something
like
what
the
computer's
operating
system
version
number
is
right.
That
kind
of
thing,
so
so,
let's
configure
this
job
to
give
us
a
little
more
information
like.
A
A
A
A
A
A
A
A
B
Okay
and
we
we
have
to
con
like,
can
we
confirm
about
the
prefetch
command
once
because
when
I
tried
it,
wasn't
there
so
absolutely.
A
A
A
Then
we've
got
sento7.
A
A
A
B
A
A
A
Unknown
option:
prefetch:
oh
okay!
Well,
that's
actually
the
the
expected
thing!
Okay
good!
So
we
can
look
at
this
one
and
see
if
any,
oh,
no,
that's
odd.
None
of
them
recognize
get
prefetch
and
the
windows
one
even
shows
as
a
failure.
Why
would
the
okay,
the
or
the
windows?
One
is
not
in
my
list
right
right?
Is
it.
A
A
Yeah,
okay,
so
let's
look
for
2.30,
so
here's
one
that
has
2.30
it
says:
remote
origin
already
exists.
A
A
B
A
B
A
A
A
A
A
A
On
on
these
yeah,
the
linux
operating
systems
have
a
tend
to
have
a
tend
to
ship,
a
version
and
most
of
the
ones
that
I've
seen
anyway
choose
to
not
do
major
or
do
changes
to
the
minor
number.
Even
they
will
do
a
2.17.2
or
3,
but
they
won't
change
the
2.18.
A
C
But,
but
as
get
versions,
so
if
that
is
going
to
be
a
reference
point
to
decide,
if
we
want
to
execute
the
maintenance
task
or
not
a
particular
intense
task,
then
if
I
don't
have
a
version
there,
because
that
list
would
grow
as
git
versions
come
out
right
or
you
would
have
some
kind
of
I
mean
you
would
say
that
if
it's
greater
than
this,
then
you
would
obviously
execute
it.
A
A
Now
now,
well,
I
thus
far
it's
a
fair
assumption.
I
I
you
know
we
reserve
the
right
to
always
have
have
them
decide.
Hey
we've
had
to
drop
this
capability
because
it
was
a
a
total
disaster
or
it
was
a.
It
was
unacceptably
bad
performance
or
who
knows?
Okay,
so
2.30.2
on
debian
11
failed,
didn't
it
yeah,
it
did
okay,
so
that
was
a
no
for
prefetch,
okay
and
then
red
hat.
Eight.
A
B
A
A
A
B
A
A
A
A
Interesting,
okay,
yeah!
That's!
Okay!
That's
very
interesting!
All
right!
Okay!
So
let's
keep
looking
and.
A
A
A
A
A
B
A
A
A
I
I've
never
used
capital
h
on
a
unix
or
capital
letter
on
a
unix
login,
I'm
willing
to
try
it,
let's
see
if
it
out
so
khrushchev
20
like
that
yeah.
B
A
Okay,
all
right
russia,
cash
cash,
okay,
empty
for
defaults,
login
group.
That
class
is
that
sh
home
directory
default
permissions,
no
password-based
authentication,
no
we're
not
going
to
lock
out
the
account
yes,
okay,
so
this
means
we've
created
the
account,
but
you
can't
log
in
yet
because
we
haven't
added
your
public
key.
Yet.
A
A
A
A
B
B
A
B
A
A
Okay,
now
I
see
you're
logged
off
so
instead
do,
in
addition
to
the
the
minus
p
argument
that
we
had
before
we're
going
to
add
one
more
argument,
which
is
a,
I
think
it's
in
minus
l
argument:
okay,
just
a
minute,
while
I
read
it,
I
always
I
always
have
to
read
this
to
remember
how
to
do
it.
It
is
minus
capital,
l.
A
A
A
A
A
A
Yes,
oh,
oh
yes,
sorry!
I
did
my
username.
Didn't
I
yes,
yes,
you'll
have
to
use
your
username
instead
of
mine.
B
A
A
B
And
can
you
remove
like
the
right
access
kind
of
things
you
know
or
something
like
that
for
that.
A
That
I
cannot
you'll
just
have
to.
If
you
destroy
it,
don't
worry
about
it.
If
you
destroyed,
it
would
be
kind.
If
you
tell
me
that
you
destroyed
it,
but
it
should
be
fully
reconstructable.
Everything
is
managed
by
code
in
it
two
years
ago.
Rashab
didn't
do
any
really
serious
damage
other
you
know
so
you'll
be
fine.
Okay,
truly,
you
don't
need
to
worry.
If
you,
if
you
do
damage
and
don't
tell
me,
that's
okay
too,
it's
just
easier.
B
C
So
while
mark
is
doing
that
rishikesh,
I
wanted
to
know,
since
you've
implemented
how
to
run
these
metering
tasks
while
they
get
plugged
in
how
what
is
your?
How
are
you
thinking
to
test
this
on
mark's
machine?
I
mean:
do
you
have
a
plan
on
that.
B
Yeah
yeah,
I
was
that
was
what
I
wanted
to
discuss
next.
So
basically
I
wrote
all
the
you
know:
maintenance.
I
wrote
the
code
for
executing
the
maintenance
tasks
along
with
logs
okay.
So
now
we
have
to
see
how
efficient
it
is.
That's
the
only
way
like
the
logs
were
written.
So
I
thought
if
we
run
it
on
mark's
machine
and
then
check
the
log
station.
B
A
A
It
first,
then,
you
tell
me
reshab,
clicked
it
first
and
we
okay,
so
if
you'll
click
that
link
that
hyperlink
or
russia,
what
you'll
see
is
it
should
open
a
web
page
and
on
that
web
page
will
be
a
text
string
of
more
than
15
characters.
That
is
your
password
okay,
that
is
my
password.
Okay,
fine!
That's
your
password
on
that
jenkins
system
and
you
can
change
it
to
whatever
you'd
like,
but
but
that's
the
password
that
I
set
there.
B
A
B
B
B
A
A
No
need
to
worry
okay,
so
now
let
me
be
sure
that
I've
got
a
copy
of
your
environment.
A
A
A
C
A
Don't
don't
assume
that
if
you
had
installed
the
one
you
wanted
to
use
yesterday
that
it's
still
there
because
frequently,
I
will
reset
things.
B
A
Okay
and
if
we
need,
if
we
need,
if
you
need
access
to
a
weekly
version,
the
version
that's
running
there
isn't
a
long-term
support
version.
If
you
need
a
weekly
version,
I
can
pretty
easily
start
and
stop
weekly
versions
on
other
hosts.
A
B
B
A
Open
that
job
and
click
the
configure
link,
okay
and
you'll,
see
as
you
scroll
through
it.
It
lists
the
steps
that
we
were
running
okay.
A
A
A
A
A
A
B
Oh
great
before
that
I
have
like
few
more
questions.
If
you
don't
oh.
A
B
Ahead,
yes,
what
other
questions
do
you
have?
I
have
I've
written
the
logic
for
logging,
the
files,
okay
and
then
I
realized
that
you
know
there
are
few
locks
which
are
written
every
minute,
because
the
cron
scheduler
runs
every
minute
and
I
think
that
would
have
an
impact
on
the
performance,
because
when
I
was
only
using
the
software,
I
could
feel
it
very
slow
on
my
computer.
So
are
there
any
suggestions
on.
A
A
B
A
Know,
oh
well,
so
as
far
as
that.
Well
so
so
there
are
some
techniques
to
keep
the
cost
of
logging
as
low
as
possible,
and
if
you
want
to
do
a
little
bit
of
studying
about
those
java,
logging
has
some
techniques
that
will
let
you
by
choosing,
which
which
method
you
call
you
can
reduce
the
expense
of
a
log
message
that
is
not
being
actually
written
to
any
log.
B
A
A
A
A
B
A
A
B
A
A
A
B
A
B
A
B
A
A
Us
to
manage
jenkins
find
a
new
thing
which
is
yeah
the
good
maintenance.
That's
dead.
You
obviously
see
it
and
I'm
not
seeing
it
oh
there.
It
is
good.
Okay,.
B
So
get
maintenance?
Can
you,
like
you,
know,
open
the
terminal
and
you
know
see
where
exactly
these
caches
are
present
like
sure
path,
so
that
you
know
we
can
look
whether
the
maintenance
command
is
running
or
not.
You
bet.
A
B
B
When
I
run
the
maintenance
command,
a
lock
file
is
written
to
one
of
the
caches.
So
if
is
there
any
chance
that
we
can
keep
watching
that
so
that
when
we
run
the
maintenance
command,
you
know
that
thing.
A
B
You
have
to
go
into
that
directory
and
then
there's
a
dot
kit.
Oh.
B
A
B
B
B
B
A
A
B
I
think
a
task
executor
or
a
task
scheduler.
You
know.
A
A
B
B
B
A
B
A
B
A
A
B
B
B
A
A
B
A
B
B
A
C
A
Right,
I
see
so
it
it
wasn't
enough
to
clear
the
schedule.
I
really
needed
to
say
hey
blank
and.
B
I
can
improve
the
ui.
This
was
just
an
initial
version.
C
A
B
A
B
B
A
A
B
A
I
and
that
I
don't
know,
I'm
not
sure,
what's
causing
it
to
be
created,
but
something
is
causing
it
to
be
created.
It
could
just
be
the
the
branch
source
branch
source
is
periodically
generating
that
thing
it
could
be
pipeline.
That's
doing
it,
there
could
be
lots
of
things
that
are
are
creating
it.
B
A
we're
now
in
june,
right
yeah.
Why
would
that
be
from
that
long
ago?
A
So
russia,
okay,
we've
got
no,
it's
still
processing
the
temp
directories.
A
A
B
A
A
B
A
A
B
A
A
This
one
is
ubuntu
20.,
yeah
good,
so
we're
we're
seeing
we're
seeing,
I
think
the
results
we
hope
for
in
that
job.
Now
I
don't
know
if
this
log
has
more
things
to
tell
us
it's
doing
commit
graph
and
iterating
over
the
temp
temp
folders
still
so
ushakesh.
Are
you
comfortable
that
this
gives
you
a
way
to
continue
exploring
and
to
see
things.