ben at nerp.net
Mon Oct 21 07:22:06 UTC 2013
Yea, basically what happened was mailman went crazy and killed itself.
The difficulty is that it didn't take it all the way down, just one of the
queue processes. And I don't know mailman well enough to know why that
wasn't automatically re-spawned. Since the mailman controller was not
down, the automatic monitor that would have restarted it didn't think
anything was wrong.
I'm working on trying to figure out what inside mailman is broken, or if
there really is a need for 3x normal memory use from time to time.
On Sun, 20 Oct 2013, Andy Isaacson wrote:
> On Fri, Oct 18, 2013 at 09:34:35PM -0700, johnyradio wrote:
>> hi Rack
>> i'm sure you know Discuss is down.
> Nope, I didn't notice, due to life being overly busy recently ... and
> nobody mailing me about it outside of the various lists. Even if
> someone had emailed root@ or postmaster@ I would have noticed. Or mail
> my personal account directly. Or tweet at me.
> This particular failure mode resulted in non-list mails going through
> just fine, so I was still getting the automated spam from cron, which
> also lulled me into complacency.
> There are other people who in theory also do admin stuff, but in
> practice I'm the only one who does it in a regular basis, so if
> something's busted on the list/wiki server, please make sure I know.
> Rack mailing list
> Rack at lists.noisebridge.net
More information about the Rack