[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: debian services and responsibility

On Fri, Mar 01, 2002 at 02:47:58PM -0700, Jason Gunthorpe wrote:
> On Fri, 1 Mar 2002, Branden Robinson wrote:
> > On Fri, Mar 01, 2002 at 11:25:15AM +0100, David N. Welton wrote:
> > > I would like to know the opinions of the DPL candidates on
> > > responsibility for Debian machines and services.
> > > 
> > > As it stands now, we have outages, and no one seems to really have a
> > > firm handle on the situation.  For instance, master has been
> > > unavailable for almost 48 hours, and the few people who have physical
> > > access to it seem to be saying "sorry, don't have time".  That is not
> > > an ideal situation for Debian services.
> > 
> > No, in fact I'd venture to say it's unacceptable.
> Huh?
> Not only is it acceptable, but it is a necessary situtation for an
> organization in our position. 

A situation where the unavailabilty was unaccompanied by any sort of
announcement to the developers after 2 days, "nobody has a firm handle
on the situation", and "the few people who do have physical access don't
have the time" is what I was calling unacceptable, not this particular
outage.  Sorry if that wasn't clear from my original message.  I would
think it goes without saying that we can "accept" an outage if Brainfood
was hit some sort of natural disaster, fire, flood, etc.  An intrusion
can be even worse than these because it's more difficult to assess the
damage -- no telltale physical signs.

I understood from IRC on Wednesday that Brainfood and DSA were both
aware of the situation and that access to the machines was going to be
available to DSA (only).  Unfortunately, this knowledge wasn't
communicated in Mr. Welton's mail and I didn't correct him on it.  I
thought we were dealing with more of a hypothetical situation since he
was asking for "the opinions of the DPL candidates on responsibility for
Debian machines and services."

> Brainfood is responding to a possible intrusion as best they can, and as
> efficiently as they can. They have done as much as possible to ensure the
> critical services will continue to operate, and I'm confident they will
> resolve the problem as soon as they can. They have always been very prompt
> with issues relating to murphy and master, they are just aware as the rest
> of us how important those machines are.

It's always been my understanding that Brainfood is just about
everything we could hope for in a sponsor.  If you were to ask me to
enumerate their deficiencies, the only one I could think of would be
that they're located in a country where U.S. controls on cryprography
export apply -- but that's not their fault.  :)

> Frankly, I'm very annoyed you decided to CC such an email to our sponsor. 
> I truely hope it is only because you and Ean have an existing
> relationship. Any other sponsor would have likely got very upset after
> getting an email like this. 

What I was trying to bring to Ean's attention was the fact that the
overwhelming majority of Debian developers didn't know the nature of the
outage, or why there was one, and yes, I Cc'ed him because we've served
together on the SPI Board for several months now.

> It is extremely difficult to get the serious dontations we do get, we
> absolultely do not need crap like this being sent to our sponsors. It is
> already pretty bad to see it in a public forum from a DPL candidate!

I think you have misinterpreted my message.  You're not alone, so I
posted an extensive clarification earlier today.  Unfortunately, list
mail appears to have slowed considerably, judging by the Received:
headers of the mails I'm getting via murphy.

> I should have ensured a posting was made to Debian-Announce when this came
> up, but I was away from my computer that day - now the issue is somewhat
> moot as it has been on -private/etc..

I personally don't think it's too late for an announcement.  If all you
need is encouragement, then please go ahead and do so.  Some of our
developers don't follow list traffic very religiously.

Thanks for speaking up; yours is the first word from a member of the DSA
team I've heard on the subject.

G. Branden Robinson                |      "To be is to do"   -- Plato
Debian GNU/Linux                   |      "To do is to be"   -- Aristotle
branden@debian.org                 |      "Do be do be do"   -- Sinatra
http://people.debian.org/~branden/ |

Attachment: pgpS4yq_n9Aca.pgp
Description: PGP signature

Reply to: