[17:20:37] <cdanis>	 hi friends, anything I should worry about before helmfile apply'ing a 200-ish sized DaemonSet?  I have at least verified updatestrategy is rolling
[17:28:48] <akosiaris>	 cdanis: updating the opentelemetry thing on wikikube ?
[17:28:55] <cdanis>	 akosiaris: correct
[17:29:09] <akosiaris>	 I think you are going to be ok then. 
[17:29:27] <cdanis>	 I was also wondering about reasonable ways to have a canary deployment of the daemonset, preferably in a way that allows it to be co-located with the mw-debug pinkunicorn pods
[17:29:34] <cdanis>	 but I am guessing that's probably not very possible
[17:30:28] <akosiaris>	 you can start a kubectl -n <namespace> get events -w --sort-by='.lastTimestamp' in a second terminal to catch issues (if any) early
[17:30:41] <cdanis>	 oh that's helpful, thank you
[17:32:13] <akosiaris>	 regarding canary for daemonsets, yeah it is possible. In fact it has nothing to do with the daemonset, but rather with the Service k8s object and we have support already in sextant modules
[17:32:26] <akosiaris>	 it's the "routed_via" thing you might see in same yaml files
[17:32:49] <cdanis>	 ahhh right
[17:32:58] <akosiaris>	 we haven't tested it specifically with a daemonset ofc but you can have 2 different daemonsets and route both and do an green/blue kind of thing
[17:33:07] <akosiaris>	 you can't do percentages though
[17:33:22] <akosiaris>	 cause a daemonset is always N, where N is the number of nodes
[17:33:28] <cdanis>	 right
[17:34:11] <cdanis>	 does routed_by work with ClusterIP?
[17:34:25] <akosiaris>	 and NodePort
[17:34:36] <cdanis>	 neat
[17:34:37] <akosiaris>	 it's unrelated to the service type in fact
[17:34:51] <cdanis>	 thanks Alex :)
[17:34:56] <akosiaris>	 yw
[20:28:56] <cdanis>	   Error: UPGRADE FAILED: release main failed, and has been rolled back due to atomic being set: timed out waiting for the condition                                                                            
[20:28:58] <cdanis>	 🤔