Re: [foaf-dev] FOAF site downtime (including xmlns.com namespaces)

...and we're back. For the EC2-interested record, I shut everything
down, unmounted the root volume, attached it to a fresh Linux instance
so I could mount the filetree and disable the misconfigured ejabberd
that was somehow wedging the machine at boot time.

Lessons learned: there are costs associated with doing development
work and prototyping on the same machine as hosting main Web sites
(even if it saves on direct hosting costs).

cheers,

Dan


On 28 April 2011 21:25, Dan Brickley <[email protected]> wrote:
> Hi folks
>
> Just to note / confirm that we have some server problems with all the
> FOAF sites including the namespace at xmlns.com. They are hosted at
> Amazon EC2. I have no indication that this is related to Amazon's
> recent woes, ... it seems rather a misconfigured xmpp server is
> causing trouble after reboot. I prefer to take the time to figure
> things out carefully tommorrow, rather than race to get things back
> for sake of uptime statistics.
>
> As it happens this week I've been working with Tom (cc:'d) and Libby
> to work out some policies between Dublin Core and FOAF to help with
> longer term outages and persistence. One thing we talked about in
> addition was also having quick / short-term 'get the namespace back
> online' mechanisms for popular namespaces. So that's something we'll
> follow up on!
>
> Apologies for any inconvenience. Will send an update when I've figured
> out a bit more,
>
> Dan
>