An update: > Dear client, > > Your server didn't show a video output and didn't respond to any keystrokes. > We have performed a BIOS update, replaced the complete RAM on suspicion of > defect > and boote your server back to the local system. > > Mit freundlichen Grüßen / Best Regards
I received that around 10:51 PM PST (UTC -8). Several of us had a discussion about this on IRC and some of the highlights that I recall from that: * OSUOSL and rackspace are good alternative data centers * We should definitely set up mirrors/fallback servers for critical bits (mail server, ghc.haskell.org, hackage.haskell.org) * We still don't know the actual cause here. What if it's software? * In an ideal world, we would use something like ansible to store our configuration so that bringing up new hosts (or moving data centers) is much easier. In my opinion, it's worth sticking with rock at hetzner for now to see if their changes make a difference. At the same time, I think we should make it a priority to bring up redundant servers at the alternatives (osuosl/rackspace). On Nov 16, 2013, at 8:15 PM, Gershom Bazerman <[email protected]> wrote: > As you may all have noticed, haskell.org is down again. We've contacted > them to actually investigate this time rather than just sending a reboot > request. But that's meant that we're staying down for the moment. I > guess worst-case-scenario tomorrow we just do the reboot to be back in > shape for monday. > > Our experiences with hetzner have not been great lately, and I've heard > bad experiences now from other members of the haskell community as well > as elsewhere (e.g. https://news.ycombinator.com/item?id=6577465). > > I was directed to the osu open source lab which provides hosting gratis > to lots of big projects [including spi itself] (http://osuosl.org/about) > and took it on myself to contact them to discuss hosting options for > some or all of our infrastructure. Migrating again will of course be > some work, but it could potentially buy us both stability and cheaper > hosting. They haven't even gotten back to me yet, so I'm not currently > proposing we do anything in particular, but I want to point out the option. > > Additionally, we have now, I believe, "bought out" ownership of abbot, > though it still resides at galois. We could potentially move it to the > osuosl and take the burden of hosting it from galois, potentially > bringing it back into rotation as an active server. > > Anyway, just wanted to fill folks in on the latest and solicit a round > of ideas/thoughts. > > Thanks, as always, to Austin, for actually being the point-person with > hetzner. > > --Gershom > _______________________________________________ > haskell-infrastructure mailing list > [email protected] > http://community.galois.com/mailman/listinfo/haskell-infrastructure
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ haskell-infrastructure mailing list [email protected] http://community.galois.com/mailman/listinfo/haskell-infrastructure
