You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When using kubens $ns and kubens -, it's taking over ten seconds. This is caused by the relisting of namespaces, of which I have at least thousands. It'd be nice if this could be sped up, either by allowing me to disable the existence check or caching the namespaces.
The text was updated successfully, but these errors were encountered:
Oh man you might be the only person with thousands of namespaces. This sounds a lot like #83 which is reinforcing the point that we should probably implement.
#83 is similar, but I do have permissions to list namespaces and I appreciate autocomplete, interactive listing & choosing (via fzf) every so often. Here are some concrete cases where I don't need the namespace check:
Using a kubens X from my shell history. The check has already been done in the past and generally namespaces don't get deleted.
Using kubens -
I think one possible tweak is to replace the list check with a get check, i.e. instead of grep -q ^"${1}"\$ <(get_namespaces), do a $KUBECTL get namespace "${1}" and interpret the result/status of that.
Yeah, that makes perfect sense. It's likely we’ll fix this in the Go implementation. So I recommend you fork the bash implementation and use yourself in the meanwhile. Thanks for bringing it up.
When using
kubens $ns
andkubens -
, it's taking over ten seconds. This is caused by the relisting of namespaces, of which I have at least thousands. It'd be nice if this could be sped up, either by allowing me to disable the existence check or caching the namespaces.The text was updated successfully, but these errors were encountered: