Page MenuHomePhabricator

Give instance administrators and users greater visibility into maintenance windows
Closed, ResolvedPublic

Description

We don't have a clear picture of this yet, but once we do we should give users better visibility into when maintenance will occur.

My general feeling is that we should be scheduling maintenance to minimize the risk and damage associated with major errors: that is, during normal work hours so everyone is awake, on hand, well-rested, at a computer, etc., and then work to reduce the impact of maintenance. Overall, do major pushes at something like 1PM PST on Thursdays, not 1:30 AM on Sunday mornings.

However, reducing maintenance impact will require some technical work.

Revisions and Commits

Event Timeline

epriestley raised the priority of this task from to Needs Triage.
epriestley updated the task description. (Show Details)
epriestley added a project: Phacility.
epriestley moved this task to Backlog on the Phacility board.
epriestley added a subscriber: epriestley.

I'm pushing this out a bit because we don't have a clear picture of what we need to tell users yet, or anything to really tell them, and are still pushing arbitrary parts of the tier at arbitrary times throughout the day without anyone complaining.

epriestley lowered the priority of this task from High to Normal.Mar 16 2015, 5:32 PM

I've been pushing the cluster weekly, but it basically goes down for about 30 seconds some time early on a Saturday and no one notices. My tentative plan here is:

  • Register a Twitter handle for status updates.
  • twitter the updates
  • Keep doing that until we outgrow it.
epriestley closed this task as Resolved by committing Restricted Diffusion Commit.Mar 16 2015, 5:51 PM
epriestley added a commit: Restricted Diffusion Commit.
  • Registered @phacilitystatus.
  • Added a link to it to the admin console.
  • Credentials in K41.