Upcoming brief downtime (5 minutes) @ 0000 UTC / 1900 ET / 1600 PT

[mark]

This maintenance has been completed successfully. Thank you!

Hi all,

I need to restart one of our databases to change a setting in preparation for the upcoming database move. The short version of the story is that we're on fairly old hardware and it's time to refresh, so I need to move the data between one host and another.

This should only take around 5 minutes at the most. I'm planning to do this in a few hours around midnight UTC / 7PM Eastern / 4PM Pacific.

As always, please follow dreamwidth for updates during such events.

Thanks!

Comments

[paserbyp]

Thank you for heads up!

[dewline]

Understood!

[mellowtigger]

And do let us know when/if you choose a Mastodon presence too. I'll gladly add my follow for news.

[morrow]

Thanks for the forewarning!

[frith]

11 minutes to go! It's like on Legends of Equestria when a message in the chat box pops up stating a reboot will happen in five minutes! Better type fast people, the end is neigh! 8^D

[mark]

[the sound of bells ringing, signaling tidings... perhaps of great misfortune]

[mark]

Will do. We are debating what to do if Twitter becomes nonviable for communications updates.

[morrow]

That went smoothly! Hit the showers folks.

[zhiva]

Were there any portals to eldritch dimesion this time? :)

Perfectly spherical cows of uniform density

[frith]

I survived the end times by reading a very long dissection of Certificates, Certificate Authorities, Trustcor and trust itself. Misfortune averted!

Thank you for keeping Dreamwidth on an even keel!

[claudeb]

I second that.

[galadhir]

Noticed nothing at all. Thank you!

[mysockstotoro]

Just want to note that Denise doesn't like Mastodon/the whole fediverse thing (I found her opinions doing a Twitter search on Twitter alts) so probably don't expect it to be any time soon.

[denise]

My issues with it aren't really relevant for an offsite comms channel (the exact explanation takes a long time but basically boils down to: handling abuse is foisted off on instance admins who have varying levels of time, cope, and commitment, and that's extremely dangerous and bad for users) but the reason we picked Twitter for offsite comms was its reliable uptime, which Mastodon instances don't really have :/ we're still thinking about the best option.