neatnik@social.lol ("Neatnik") wrote:
All done! We’re now on our new server, with tons of extra RAM and disk space ("so much room for activities"), running 4.6.3+glitch, and all sidekiq queues are now fully caught up.
Elasticsearch indexes are being rebuilt right now, and that’ll take a while (even at ~2500 docs/second, we're looking at ~21 hours to fully cover the 207,680,460 items in our index).
Overall, everything went well! Thanks once again for your patience while I worked through this move, and apologies — again — for the unrelated (but sort of tangentially related) disruptions earlier. The good news is that our migration addresses the memory issues that led to those hiccups earlier, so thing should be so much better from here on out. :prami_happy: