citi, linux-nfs.org downtime

citi, linux-nfs.org downtime

[J. Bruce Fields]

CITI's offices and machine room are moving to a new building tomorrow
(tuesday).  This means that citi.umich.edu and linux-nfs.org will be
going down late tonight or early tomorrow morning, and staying down at
least until early afternoon (and probably longer).

(Alert users may already have noticed one or two outages of several
hours over the last few weeks as we prepared the machine room for the
move.)

Apologies in advance for any inconvenience--getting services back up
will be our first priority, but it's a complicated operation.

Also we'll be very busy (and possibly not have access to email), so may
not be very responsive for a few days.

--b.

Re: citi, linux-nfs.org downtime

[Peter Chacko]

Very bad of you.... should have adopted virtualization and its
non-disruptive migration benefits :-)  :-)

On Mon, Jun 22, 2009 at 9:56 PM, J. Bruce Fields<bfields@fieldses.org> =
wrote:
> CITI's offices and machine room are moving to a new building tomorrow
> (tuesday). =A0This means that citi.umich.edu and linux-nfs.org will b=
e
> going down late tonight or early tomorrow morning, and staying down a=
t
> least until early afternoon (and probably longer).
>
> (Alert users may already have noticed one or two outages of several
> hours over the last few weeks as we prepared the machine room for the
> move.)
>
> Apologies in advance for any inconvenience--getting services back up
> will be our first priority, but it's a complicated operation.
>
> Also we'll be very busy (and possibly not have access to email), so m=
ay
> not be very responsive for a few days.
>
> --b.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" =
in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at =A0http://vger.kernel.org/majordomo-info.html
>

Re: citi, linux-nfs.org downtime

[Trond Myklebust]

On Mon, 2009-06-22 at 22:37 +0530, Peter Chacko wrote:
> Very bad of you.... should have adopted virtualization and its
> non-disruptive migration benefits :-)  :-)

It is non-disruptive only if you have a machine that can host the VM
while the contents of the machine room are being moved. It also assumes
a rich sugar-daddy who can pony up the cash for new processors with vm
extensions... ;-)

Trond

> On Mon, Jun 22, 2009 at 9:56 PM, J. Bruce Fields<bfields@fieldses.org> wrote:
> > CITI's offices and machine room are moving to a new building tomorrow
> > (tuesday).  This means that citi.umich.edu and linux-nfs.org will be
> > going down late tonight or early tomorrow morning, and staying down at
> > least until early afternoon (and probably longer).
> >
> > (Alert users may already have noticed one or two outages of several
> > hours over the last few weeks as we prepared the machine room for the
> > move.)
> >
> > Apologies in advance for any inconvenience--getting services back up
> > will be our first priority, but it's a complicated operation.
> >
> > Also we'll be very busy (and possibly not have access to email), so may
> > not be very responsive for a few days.
> >
> > --b.
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> >
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: citi, linux-nfs.org downtime

[J. Bruce Fields]

On Mon, Jun 22, 2009 at 12:26:11PM -0400, J. Bruce Fields wrote:
> CITI's offices and machine room are moving to a new building tomorrow
> (tuesday).  This means that citi.umich.edu and linux-nfs.org will be
> going down late tonight or early tomorrow morning, and staying down at
> least until early afternoon (and probably longer).
> 
> (Alert users may already have noticed one or two outages of several
> hours over the last few weeks as we prepared the machine room for the
> move.)
> 
> Apologies in advance for any inconvenience--getting services back up
> will be our first priority, but it's a complicated operation.

Stuff should be mostly working again now.

So, if I have this right:

	Total data moved: ~ 35 terabytes
	Time to get racks back online: ~ 28 hours
	Estimated total bandwidth of the CITI move: 2.8 gigabits/second

If we count only the time to physically move the racks, and not the time
to get them back online, bandwidth is closer to 20 gigabits per second.

--b.

Re: citi, linux-nfs.org downtime

[Brian J. Murrell]

[-- Attachment #1: Type: text/plain, Size: 497 bytes --]

On Wed, 2009-06-24 at 17:35 -0400, J. Bruce Fields wrote:
> 	Estimated total bandwidth of the CITI move: 2.8 gigabits/second
> 
> If we count only the time to physically move the racks, and not the time
> to get them back online, bandwidth is closer to 20 gigabits per second.

Never underestimate the bandwidth of a station wagon full of tapes
hurtling down the highway. 
—Tanenbaum, Andrew S. (1996). Computer Networks. New Jersey:
Prentice-Hall. pp. 83. ISBN 0-13-349945-6.

b.


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]