[11:36:13] <stevemunene>	 Hi arturo noticed we are both doing some decommissions, have you merged the dns deletions of analytics1063, cloudswift1001, cloudswift1002 or should I proceed to do it?
[11:36:30] <arturo>	 stevemunene: I think I have already
[11:37:28] <arturo>	 I saw this: !log stevemunene@cumin1001 START - Cookbook sre.hosts.decommission for hosts analytics1061.eqiad.wmnet
[11:38:27] <arturo>	 sorry, this: !log stevemunene@cumin1001 START - Cookbook sre.hosts.decommission for hosts analytics1063.eqiad.wmnet
[11:38:37] <arturo>	 and the diff matched just that
[11:39:34] <stevemunene>	 https://www.irccloud.com/pastebin/MyetO1T6/
[11:40:20] <stevemunene>	 aborting and restarting since it is already merged arturo 
[11:41:06] <arturo>	 ack
[11:41:56] <arturo>	 I don't think you strictly need to restart the decomission cookbook, but should be harmless anyway
[12:00:10] <stevemunene>	 np restarted it and so far all good on this end arturo :)
[12:00:23] <arturo>	 👍
[12:50:25] <stevemunene>	 Still working on some decommissions, is this from Cookbook sre.puppet.sync-netbox-hiera generate netbox hiera data okay to merge? https://usercontent.irccloud-cdn.com/file/dOoCf2cZ/image.png
[12:56:38] <volans>	 topranks maybe? ^^^
[12:57:19] <topranks>	 volans: yes correct, should have anticipated that 
[12:57:30] <topranks>	 stevemunene: you can merge that it looks correct 
[12:57:36] <topranks>	 apologies for the confusion 
[12:57:49] <stevemunene>	 great, thanks topranks volans 
[15:19:31] <elukey>	 volans, sukhe o/
[15:19:47] <sukhe>	 elukey: hellooo
[15:19:48] <elukey>	 I'd like to test https://gerrit.wikimedia.org/r/c/operations/deployment-charts/+/935772, that is a new kafka setting for changeprop
[15:19:49] <volans>	 elukey: what's up?
[15:20:09] <sukhe>	 go for it as long as you fix it :P 
[15:20:10] <sukhe>	 (kidding)
[15:20:21] <elukey>	 it should just change the batching size (so wait 20 ms max before sending msgs to increase throughput)
[15:20:26] <volans>	 all yours I'm offcall in 10m :D
[15:20:39] <sukhe>	 haha
[15:22:01] <elukey>	 ahahah
[15:22:03] <elukey>	 okok
[15:55:41] <elukey>	 rolled out, I'll avoid to deploy the change to the jobqueue as well to limit the blast radious if anything goes wrong
[15:55:46] <sukhe>	 elukey: <3
[15:56:06] <elukey>	 I'll check metrics later on, if anything looks weird it is sufficient to revert my last two changes in deployment-charts
[17:09:01] <sukhe>	 hi folks, if you see any DNS issues, please ping me immediately
[17:09:31] <sukhe>	 the reason is that we just reimaged a DNS host (dns1004) and took out dns1001, but this is the first time with automatic config generation (NTP and resolv.conf)
[17:09:44] <sukhe>	 while everything looks and should be fine, if you see something, please let me know thanks
[17:17:08] <mutante>	 stops using "ssh ns0.wikimedia.org" as I should have, heh
[17:18:02] <sukhe>	 haha, you might get a key error but it now points to dns100[2-4] now
[17:18:04] <mutante>	 key change, that is kind of historic if dns1001 is down
[17:18:10] <sukhe>	 yeah, truly
[17:18:21] <mutante>	 congrats
[19:21:21] <sukhe>	 ---
[19:21:29] <sukhe>	 no pending DNS work, so all is fine :) 
[19:22:08] <sukhe>	 I am on on-call so if I jinx it, that's on me