--log-level ignored by C-chain logging

[mataebi]

Describe the bug
It seems that the --log-level setting is not observed by reporting activity on the C chain while on P and X chain this seems to work as expected. This produces huge amounts of unwanted log output wasting resources and slowing down the service unnecessarily.

To Reproduce
Run avalanchego ... --log-level=Info ... or any other setting for the log level. In the log files you will see INFO messages and above for P and X chain as expected:

==> P/0.log <==
INFO [06-06|17:48:56] <P Chain> avalanchego/vms/platformvm/vm.go#200: initializing last accepted block as 2gnCjpDPMxbZTHXtFtsDmMC7HGNyMTKwntJpiJJ4YtWdfZRCnu
INFO [06-06|17:48:56] <P Chain> avalanchego/snow/engine/snowman/transitive.go#64: initializing consensus engine
INFO [06-06|17:48:56] <P Chain> avalanchego/snow/engine/common/bootstrapper.go#71: Starting bootstrap...
INFO [06-06|17:49:00] <P Chain> avalanchego/snow/engine/common/bootstrapper.go#300: Bootstrapping started syncing with 1 vertices in the accepted frontier
...

==> X/0.log <==
INFO [06-06|17:49:01] <X Chain> avalanchego/snow/engine/avalanche/transitive.go#72: initializing consensus engine
INFO [06-06|17:49:01] <X Chain> avalanchego/snow/engine/common/bootstrapper.go#71: Starting bootstrap...
INFO [06-06|17:49:05] <X Chain> avalanchego/snow/engine/common/bootstrapper.go#300: Bootstrapping started syncing with 1 vertices in the accepted frontier
INFO [06-06|17:49:07] <X Chain> avalanchego/snow/engine/avalanche/bootstrap/bootstrapper.go#428: bootstrapping fetched 2 vertices. Executing transaction state transitions...

However, the C chain log shows all log levels including DEBUG messages independent of the setting of --log-level:

==> C/0.log <==
INFO [06-06|17:55:14.031] Imported new chain segment               blocks=1 txs=1 mgas=0.021 elapsed=2.987ms   mgasps=7.030   number=2233751 hash="1d7626…01ca44" dirty=0.00B
DEBUG[06-06|17:55:14.031] Reinjecting stale transactions           count=0
DEBUG[06-06|17:55:14.729] Accepting block 1d7626fd76f92d521cb63a58423c5e875d45eba2c319dca3d0e68b497b01ca44 (DyZD3c1FUhBuJkWk5UmSjFc2f1ZcqknFAFR9waKbKXVKfFJCV) at height 2233751 
DEBUG[06-06|17:55:28.825] crosschain X->C                          addr=0xB2224440da81c3D64f5Bb00E007ACd6AD5c1D05a amount=627360000      assetID=AVAX
DEBUG[06-06|17:55:28.876] crosschain X->C                          addr=0xB2224440da81c3D64f5Bb00E007ACd6AD5c1D05a amount=627360000      assetID=AVAX
DEBUG[06-06|17:55:28.878] Persisted trie from memory database      nodes=6   size=2.45KiB  time="596.748µs" gcnodes=0 gcsize=0.00B gctime=0s livenodes=1 livesize=-7462.00B
...

Expected behavior
The C chain log should show messages observing log-level as the other two chain logs do.

Operating System
Linux version 4.19.0-16-amd64 ([email protected]) (gcc version 8.3.0 (Debian 8.3.0-6)) #1 SMP Debian 4.19.181-1 (2021-03-19)

Additional context
Observed in 1.4.7 (but was most likely there before)

By submitting this issue I agree to the Terms and Conditions of the Developer Accelerator Program.

[ceyonur]

This was fixed with an external C-chain config: log-level

Release notes: https://docs.avax.network/build/release-notes/avalanchego#v1.5.3-view-on-github

Also see C-chain configs in: https://docs.avax.network/build/references/command-line-interface#c-chain-config