On Nov 7, 2017 1:12 AM, "el kalin" <ka...@el.net> wrote: hi guys... this is apparently happening across the board on aws with all netbsds. i have a couple of production machines and a couple a test ones. one of the test ones were restarted by aws and now unreachable. the logs show the root partition can not be loaded. kernel problems. this is most likely due to them updating their underlying xen layer... i'm in panic because i can not imagine what would happened if they decide to restart one of the production systems...
on the official netbsd aws page at https://wiki.netbsd.org/amazon_ec2/amis/ on the bottom says: "Last edited 1 year and 11 months ago"... my question is will netbsd actively produce aws images or should i just go straight to fbsd and forget netbsd for the future... if anybody has any answer/reply it'll be really appreciated... I heard that someone is working on getting things running again using HVM instead of PV mode, but I have no idea on the timeline. (Bcc'd him, so as not to put him on the spot if I got my rumor wrong) I ended up switching to Google's GCE, and while I couldn't find any pre-built images, the process defined in https://github.com/google/netbsd-gce worked pretty smoothly (though the lack of pre-built packages for netbsd-8 is annoying). Obviously that doesn't help if your production systems have AWS dependencies, but otherwise it seems like a good option. You should probably expect that your prod systems *will* be forcibly rebooted at some point, but you can look in amazon's console to see when/if that will be. (There's a scheduled tasks section) Eric
