Long index updates inside of custom containers

[Mark Fine]

I've run into issues updating a package index inside of a container lately - it's a custom container that is based on haskell:8.2.1, but with stack-1.6.3 and ghc-8.2.2. Any known issues or things to help troubleshoot?

root@1d5d51c3d6f0:/# stack update --verbose

Version 1.6.3, Git revision b27e629b8c4ce369e3b8273f04db193b060000db (5454 commits) x86_64 hpack-0.20.0

2018-01-11 00:47:32.336874: [debug] Checking for project config at: /stack.yaml

@(Stack/Config.hs:842:9)

2018-01-11 00:47:32.337174: [debug] No project config file found, using defaults.

@(Stack/Config.hs:872:13)

2018-01-11 00:47:32.339614: [info] Selected mirror https://s3.amazonaws.com/hackage.fpcomplete.com/

@(Stack/PackageIndex.hs:309:24)

2018-01-11 00:47:32.345377: [info] Downloading timestamp

@(Stack/PackageIndex.hs:309:24)

2018-01-11 00:47:32.767299: [info] Downloading snapshot

@(Stack/PackageIndex.hs:309:24)

2018-01-11 00:47:32.854847: [info] Updating index

@(Stack/PackageIndex.hs:309:24)

Updating package index Hackage (mirrored at https://s3.amazonaws.com/hackage.fpcomplete.com/) ...

this just hangs forever now, and I'm not sure what's going on. Thanks for any help!

Mark

[Mark Fine]

An update - I had a call to `stack update` in my container's layer (like, this https://github.com/freebroccolo/docker-haskell/blob/master/8.2/Dockerfile#L20 but with a call to `stack update` appended to the layer). As far as I could tell, this might have lead to poor performance at the index got updated between the container's layer and subsequent builds.

Mark