Re: [bitfolk] Scheduled maintenance, 2016-09-02, 03 and 05

Hi, On Thu, Aug 25, 2016 at 09:11:51PM +0000, Andy Smith wrote: I realise I did not follow up on how the work went on 2016-09-02 and today. Particularly important since there was a mistake made on 2016-09-02. During the reboot of the first host, "jack", on 2016-09-02 I attempted to be clever and stab the "Enter" key during the grub prompt to shave off 5 seconds and have it boot the correct kernel immediately. Unfortunately I missed the correct key and caused the cursor to move, selecting an older, incorrect kernel. As the machines boot quite fast with SSDs now, by the time I realised what had happened it was already at the point where it was going to start mounting filesystems and I judged that it would be a bad idea to attempt to reboot or power off at that stage. I decided to let it boot and then reboot it again. This regrettably meant an additional boot/shutdown cycle was seen by some customers with VPSes on "jack" and a couple of minutes were added to the outage length. In future I will not interact with grub unless it is necessary. The correct kernel to boot is already set from userland with grub-editenv. The work for the remaining hosts that day ("macallan" and "leffe") as well as work for "hen" and "hobgoblin" today went without incident. Remaining work for "paradox" and "snaps" is taking place tomorrow, 2016-09-05, as scheduled. Cheers, Andy -- http://bitfolk.com/ -- No-nonsense VPS hosting