[10:43:10] claime, Raine: following up on our conversation yesterday, whart would be a good time to deploy an unmerged branch to staging to experiment with `helmfile test`? [10:43:47] effie and I are deploying some stuff to the api/rest-gateway [10:44:00] I think once we're done you would be g2g [10:45:17] I am applying on eqiad rest-gateway, and I reckon we are done [10:45:35] redioscope, ratelimit and rest-gateway have been migrated [14:57:09] didn't get around to it today. would it be ok if I did it in the morning tomorrow? like 8:00 utc or so? [15:01:42] duesen: I should be online around that time [15:01:55] please ping me here and/or on slack [15:12:12] I'm trying to deploy linked-artifacts to production (eqiad/codfw) for the first time (thus far, it's only been in staging), and I'm getting i/o timeouts when connecting to Cassandra (restbase cluster). I'm guessing I did something wrong WRT the external service, which is all still very opaque to me. How does that all work again? [15:12:29] There is some puppet-ry somewhere, no? [15:19:03] oh, I think I might have found it after all [15:19:16] urandom: what was it? [15:19:29] (sorry, just catching up) [15:19:40] I think I used the wrong name for the external service [15:23:13] I still don't think I understand how all of that works, but I found the yaml output on the deployment host [15:23:33] s/found/reunited with/ (because I think someone tried to explain this to me earlier) [15:24:02] effie: ok, thanks. this is a low priority experiment, so something else may come up. But I'll block an hour tomorrow morning. [15:25:02] urandom: ok, lmk if you want me to look at anything, I'll be around relatively late today [15:26:47] Raine: thanks!