From 2f8dc96bb740fb879f0af7e6ac81794c0e5cc4e8 Mon Sep 17 00:00:00 2001 From: Shannon Quinn Date: Thu, 4 Aug 2022 15:31:32 -0400 Subject: [PATCH] Initial attempt at addressing the NEBARI_KUBECONFIG in tmp issue raised in #83 and #88. May need some more detail. --- docs/docs/troubleshooting.md | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/docs/docs/troubleshooting.md b/docs/docs/troubleshooting.md index 43cf250fa..2c045ca82 100644 --- a/docs/docs/troubleshooting.md +++ b/docs/docs/troubleshooting.md @@ -6,6 +6,7 @@ Invariably you will encounter behavior that does not match your expectations. Th - DNS domain=1your_nebari_domain1 record does not exist - User instances on AWS occasionally die ~30 minutes after spinning up a large dask cluster + - Why is the `NEBARI_KUBECONFIG` file in `/tmp`? [_Errors_](#errors) @@ -39,6 +40,10 @@ AWS uses Elastic Kubernetes Service for hosting the Kubernetes cluster. [It requ To stop this from happening, the autoscaler service "AZRebalance" needs to be manually suspended. You can [follow the steps in the AWS documentation to suspend the AZRebalance service](https://docs.aws.amazon.com/autoscaling/ec2/userguide/as-suspend-resume-processes.html). +### Why is the `NEBARI_KUBECONFIG` file in `/tmp`? + +Nebari regenerates this file on every run. Yes, it will be trashed by the operating system's tmpdir cleanup, but it will be repopulated. + ## Errors ### Conda-Store environment fails to build