server: fix interceptConfigs

[alexanderbez]

Description

1. cleanup error messages
2. fix issue where app.toml was overriding config.toml

---

Before we can merge this PR, please make sure that all the following items have been
checked off. If any of the checklist items are not applicable, please leave them but
write a little note why.

• Targeted PR against correct branch (see CONTRIBUTING.md)
• Linked to Github issue with discussion and accepted design OR link to spec that describes this work.
• Code follows the module structure standards.
• Wrote unit and integration tests
• Updated relevant documentation (docs/) or specification (x/<module>/spec/)
• Added relevant godoc comments.
• Added a relevant changelog entry to the Unreleased section in CHANGELOG.md
• Re-reviewed Files changed in the Github PR explorer
• Review Codecov Report in the comment section below once CI passes

[alexanderbez]

This was causing any user value set in Tendermint's config to be wiped out. We need to merge configurations, not overwrite them.

[tac0turtle]

utACK

[alessio]

Argh - tests are failing :-/

[alexanderbez]

Error: Canelled in prerun due to TestInterceptConfigsPreRunHandlerDoesNotMixConfigFiles failing?

[alexanderbez]

So it seems TestInterceptConfigsPreRunHandlerDoesNotMixConfigFiles is actually wrong in the invariant its trying to test.

Recall, interceptConfigs reads configuration from config.toml (tendermint) and app.toml (application) and its intention is to merge them into a single viper instance.

So in TestInterceptConfigsPreRunHandlerDoesNotMixConfigFiles, where we check:

	if 0 != serverCtx.Viper.GetInt("halt-time") {
		t.Error("Halt time is not using default")
	}

But that is not the case.

/cc @hydrogen18 we should be using assertions via testfiy and no need for constants in these tests.

[hydrogen18]

@alexanderbez is this PR related to an issue I opened? I don't really have any context here

[alexanderbez]

@hydrogen18 its in relation to the TestInterceptConfigsPreRunHandlerDoesNotMixConfigFiles. We're trying to test and assert invariants that are not true. We should probably just remove this test all together

[hydrogen18]

If what you're saying is correct, there's no difference between the two configuration files at all. You can specify any value in any file. Which leads me to the question of why have two files?

[alexanderbez]

If what you're saying is correct, there's no difference between the two configuration files at all. You can specify any value in any file. Which leads me to the question of why have two files?

One is managed by Tendermint and the other by the application. What exists in the application is completely opaque to Tendermint and is completely customizable, so having one file would not make sense.

[amaury1093]

sgtm, utACK

[codecov]

Codecov Report

Merging #8641 (3e4f4c7) into master (7b52767) will decrease coverage by 0.04%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #8641      +/-   ##
==========================================
- Coverage   61.47%   61.43%   -0.05%     
==========================================
  Files         659      658       -1     
  Lines       37876    37769     -107     
==========================================
- Hits        23283    23202      -81     
+ Misses      12159    12143      -16     
+ Partials     2434     2424      -10     

Impacted Files	Coverage Δ
x/authz/genesis.go	66.66% <0.00%> (ø)
x/authz/types/msgs.go	53.84% <0.00%> (ø)
x/authz/types/codec.go	0.00% <0.00%> (ø)
x/authz/keeper/keeper.go	71.29% <0.00%> (ø)
x/authz/types/genesis.go	0.00% <0.00%> (ø)
x/authz/client/cli/query.go	75.90% <0.00%> (ø)
x/authz/types/generic_authorization.go	0.00% <0.00%> (ø)
x/authz/types/authorization_grant.go
x/staking/types/authz.go
x/bank/types/send_authorization.go
... and 3 more