diff --git a/bentoml/_internal/configuration/containers.py b/bentoml/_internal/configuration/containers.py
index a23eeee5513..333aa35fe18 100644
--- a/bentoml/_internal/configuration/containers.py
+++ b/bentoml/_internal/configuration/containers.py
@@ -25,7 +25,6 @@
 from ..context import component_context
 from ..resource import CpuResource
 from ..resource import system_resources
-from ...exceptions import InvalidArgument
 from ...exceptions import BentoMLConfigException
 from ..utils.unflatten import unflatten
 
@@ -39,37 +38,16 @@
 
 config_merger = Merger(
     # merge dicts
-    [(dict, "merge")],
+    type_strategies=[(dict, "merge")],
     # override all other types
-    ["override"],
+    fallback_strategies=["override"],
     # override conflicting types
-    ["override"],
+    type_conflict_strategies=["override"],
 )
 
 logger = logging.getLogger(__name__)
 
 
-RUNNER_CFG_KEYS = ["batching", "resources", "logging", "metrics", "timeout"]
-
-
-CONFIG_LATEST_VERSION = 2
-
-
-def migrate_up(from_version: int, to_version: int):
-    # All migrate up functions should only take a flattened config dict as input
-    migrate = getattr(
-        import_configuration_spec(version=from_version),
-        f"migrate_to_v{to_version}",
-        None,
-    )
-    if migrate is None:
-        raise InvalidArgument(
-            "Given version %d requires 'migrate_to_v%d' to update config to version %d"
-            % (from_version, to_version, to_version)
-        )
-    return migrate
-
-
 class BentoMLConfiguration:
     def __init__(
         self,
@@ -77,7 +55,7 @@ def __init__(
         override_config_values: str | None = None,
         *,
         validate_schema: bool = True,
-        use_version: int = CONFIG_LATEST_VERSION,
+        use_version: int = 1,
     ):
         # Load default configuration with latest version.
         self.config = get_default_config(version=use_version)
@@ -93,23 +71,16 @@ def __init__(
                 # If users does not define a version, we then by default assume they are using v1
                 # and we will migrate it to latest version
                 logger.debug(
-                    "User config does not define a version, assuming given config is version 1. Migrating to version %d..."
+                    "User config does not define a version, assuming given config is version %d..."
                     % use_version
                 )
-                current = 1
+                current = use_version
             else:
                 current = override["version"]
-            if current < use_version:
-                # Each version of configuration should have its own migration functions to the latest version
-                # i.e: v1 should have a migration function to v2 called "migrate_to_v2"
-                # Note that we should always migrate up, not down.
-                logger.debug(
-                    "Migrating from version %d to version %d..."
-                    % (current, use_version)
-                )
-                override = migrate_up(current, use_version)(
-                    override_config=dict(flatten_dict(override))
-                )
+            compat = getattr(import_configuration_spec(current), "compat_layer", None)
+            # Running compatibliity layer if it exists
+            if compat:
+                override = compat(override_config=dict(flatten_dict(override)))
             config_merger.merge(self.config, override)
 
         if override_config_values is not None:
@@ -135,19 +106,21 @@ def __init__(
             if "version" in override_config_map:
                 override_version = override_config_map["version"]
                 logger.debug(
-                    "Found defined 'version=%d' in BENTOML_CONFIG_OPTIONS. We will migrate up to latest configuration if possible."
+                    "Found defined 'version=%d' in BENTOML_CONFIG_OPTIONS."
                     % override_version
                 )
-                if override_version < use_version:
-                    override_config_map = migrate_up(override_version, use_version)(
-                        override_config=override_config_map
-                    )
+                compat = getattr(
+                    import_configuration_spec(override_version), "compat_layer", None
+                )
+                # Running compatibliity layer if it exists
+                if compat:
+                    override_config_map = compat(override_config=override_config_map)
             # Previous behaviour, before configuration versioning.
             try:
                 override = unflatten(override_config_map)
             except ValueError as e:
                 raise BentoMLConfigException(
-                    f"Failed to parse config options from the env var: {e}.\n *** Note: You can use '\"' to quote the key if it contains special characters. ***"
+                    f"Failed to parse config options from the env var:\n{e}.\n*** Note: You can use '\"' to quote the key if it contains special characters. ***"
                 ) from None
             config_merger.merge(self.config, override)
 
@@ -163,14 +136,20 @@ def __init__(
                 ) from None
 
     def _finalize(self):
-        global_runner_cfg = {k: self.config["runners"][k] for k in RUNNER_CFG_KEYS}
-        for key in self.config["runners"]:
-            if key not in RUNNER_CFG_KEYS:
-                runner_cfg = self.config["runners"][key]
+        GLOBAL_RUNNERS_KEY = ["batching", "resources", "logging", "metrics", "timeout"]
+        global_runner_cfg = {k: self.config["runners"][k] for k in GLOBAL_RUNNERS_KEY}
+        custom_runners_cfg = dict(
+            filter(
+                lambda kv: kv[0] not in GLOBAL_RUNNERS_KEY,
+                self.config["runners"].items(),
+            )
+        )
+        if custom_runners_cfg:
+            for runner_name, runner_cfg in custom_runners_cfg.items():
                 # key is a runner name
                 if runner_cfg.get("resources") == "system":
                     runner_cfg["resources"] = system_resources()
-                self.config["runners"][key] = config_merger.merge(
+                self.config["runners"][runner_name] = config_merger.merge(
                     deepcopy(global_runner_cfg),
                     runner_cfg,
                 )
diff --git a/bentoml/_internal/configuration/helpers.py b/bentoml/_internal/configuration/helpers.py
index fdf4a2012cf..cd099ee5610 100644
--- a/bentoml/_internal/configuration/helpers.py
+++ b/bentoml/_internal/configuration/helpers.py
@@ -67,6 +67,9 @@ def rename_fields(
             d[replace_with] = d.pop(current)
 
 
+punctuation = r"""!"#$%&'()*+,-./:;<=>?@[\]^`{|}~"""
+
+
 def flatten_dict(
     d: t.MutableMapping[str, t.Any],
     parent: str = "",
@@ -74,10 +77,8 @@ def flatten_dict(
 ) -> t.Generator[tuple[str, t.Any], None, None]:
     """Flatten nested dictionary into a single level dictionary."""
     for k, v in d.items():
-        # TODO: we probably need to find a better way
-        # to normalize slash and special characters keys.
-        key = f'"{k}"' if "/" in k else k
-        nkey = parent + sep + key if parent else key
+        k = f'"{k}"' if any(i in punctuation for i in k) else k
+        nkey = parent + sep + k if parent else k
         if isinstance(v, t.MutableMapping):
             yield from flatten_dict(
                 t.cast(t.MutableMapping[str, t.Any], v), parent=nkey, sep=sep
diff --git a/bentoml/_internal/configuration/v1/__init__.py b/bentoml/_internal/configuration/v1/__init__.py
index e70e1fbe41d..607c8b1c747 100644
--- a/bentoml/_internal/configuration/v1/__init__.py
+++ b/bentoml/_internal/configuration/v1/__init__.py
@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import re
 import typing as t
 
 import schema as s
@@ -9,44 +10,66 @@
 from ..helpers import rename_fields
 from ..helpers import ensure_larger_than
 from ..helpers import is_valid_ip_address
+from ..helpers import ensure_iterable_type
 from ..helpers import validate_tracing_type
 from ..helpers import validate_otlp_protocol
 from ..helpers import ensure_larger_than_zero
+from ...utils.metrics import DEFAULT_BUCKET
 from ...utils.unflatten import unflatten
 
-__all__ = ["SCHEMA", "migrate_to_v2"]
+__all__ = ["SCHEMA", "compat_layer"]
 
-_TRACING_CONFIG = {
-    "type": s.Or(s.And(str, s.Use(str.lower), validate_tracing_type), None),
+TRACING_CFG = {
+    "exporter_type": s.Or(s.And(str, s.Use(str.lower), validate_tracing_type), None),
     "sample_rate": s.Or(s.And(float, ensure_range(0, 1)), None),
+    "timeout": s.Or(s.And(int, ensure_larger_than_zero), None),
+    "max_tag_value_length": s.Or(int, None),
     "excluded_urls": s.Or([str], str, None),
-    s.Optional("zipkin"): {"url": s.Or(str, None)},
-    s.Optional("jaeger"): {"address": s.Or(str, None), "port": s.Or(int, None)},
-    s.Optional("otlp"): {
+    "zipkin": {
+        "endpoint": s.Or(str, None),
+    },
+    "jaeger": {
+        "protocol": s.Or(
+            s.And(str, s.Use(str.lower), lambda d: d in ["thrift", "grpc"]),
+            None,
+        ),
+        "collector_endpoint": s.Or(str, None),
+        "thrift": {
+            "agent_host_name": s.Or(str, None),
+            "agent_port": s.Or(int, None),
+            "udp_split_oversized_batches": s.Or(bool, None),
+        },
+        "grpc": {
+            "insecure": s.Or(bool, None),
+        },
+    },
+    "otlp": {
         "protocol": s.Or(s.And(str, s.Use(str.lower), validate_otlp_protocol), None),
-        "url": s.Or(str, None),
+        "endpoint": s.Or(str, None),
+        "compression": s.Or(
+            s.And(str, lambda d: d in {"gzip", "none", "deflate"}), None
+        ),
+        "http": {
+            "certificate_file": s.Or(str, None),
+            "headers": s.Or(dict, None),
+        },
+        "grpc": {
+            "insecure": s.Or(bool, None),
+            "headers": s.Or(lambda d: isinstance(d, t.Sequence), None),
+        },
     },
 }
 _API_SERVER_CONFIG = {
-    "host": s.And(str, is_valid_ip_address),
-    "port": s.And(int, ensure_larger_than_zero),
     "workers": s.Or(s.And(int, ensure_larger_than_zero), None),
     "timeout": s.And(int, ensure_larger_than_zero),
     "backlog": s.And(int, ensure_larger_than(64)),
-    "max_request_size": s.And(int, ensure_larger_than_zero),
-    s.Optional("ssl"): {
-        s.Optional("certfile"): s.Or(str, None),
-        s.Optional("keyfile"): s.Or(str, None),
-        s.Optional("keyfile_password"): s.Or(str, None),
-        s.Optional("version"): s.Or(s.And(int, ensure_larger_than_zero), None),
-        s.Optional("cert_reqs"): s.Or(int, None),
-        s.Optional("ca_certs"): s.Or(str, None),
-        s.Optional("ciphers"): s.Or(str, None),
-    },
     "metrics": {
         "enabled": bool,
         "namespace": str,
         s.Optional("duration"): {
+            s.Optional("buckets", default=DEFAULT_BUCKET): s.Or(
+                s.And(list, ensure_iterable_type(float)), None
+            ),
             s.Optional("min"): s.Or(s.And(float, ensure_larger_than_zero), None),
             s.Optional("max"): s.Or(s.And(float, ensure_larger_than_zero), None),
             s.Optional("factor"): s.Or(s.And(float, ensure_larger_than(1.0)), None),
@@ -59,16 +82,48 @@
             "request_content_type": s.Or(bool, None),
             "response_content_length": s.Or(bool, None),
             "response_content_type": s.Or(bool, None),
+            "format": {
+                "trace_id": str,
+                "span_id": str,
+            },
         },
     },
-    "cors": {
+    "http": {
+        "host": s.And(str, is_valid_ip_address),
+        "port": s.And(int, ensure_larger_than_zero),
+        "cors": {
+            "enabled": bool,
+            "allow_origin": s.Or(str, None),
+            "allow_origin_regex": s.Or(s.And(str, s.Use(re.compile)), None),
+            "allow_credentials": s.Or(bool, None),
+            "allow_headers": s.Or([str], str, None),
+            "allow_methods": s.Or([str], str, None),
+            "max_age": s.Or(int, None),
+            "expose_headers": s.Or([str], str, None),
+        },
+    },
+    "grpc": {
+        "host": s.And(str, is_valid_ip_address),
+        "port": s.And(int, ensure_larger_than_zero),
+        "metrics": {
+            "port": s.And(int, ensure_larger_than_zero),
+            "host": s.And(str, is_valid_ip_address),
+        },
+        "reflection": {"enabled": bool},
+        "max_concurrent_streams": s.Or(int, None),
+        "max_message_length": s.Or(int, None),
+        "maximum_concurrent_rpcs": s.Or(int, None),
+    },
+    "tracing": TRACING_CFG,
+    s.Optional("ssl"): {
         "enabled": bool,
-        "access_control_allow_origin": s.Or(str, None),
-        "access_control_allow_credentials": s.Or(bool, None),
-        "access_control_allow_headers": s.Or([str], str, None),
-        "access_control_allow_methods": s.Or([str], str, None),
-        "access_control_max_age": s.Or(int, None),
-        "access_control_expose_headers": s.Or([str], str, None),
+        s.Optional("certfile"): s.Or(str, None),
+        s.Optional("keyfile"): s.Or(str, None),
+        s.Optional("keyfile_password"): s.Or(str, None),
+        s.Optional("version"): s.Or(s.And(int, ensure_larger_than_zero), None),
+        s.Optional("cert_reqs"): s.Or(int, None),
+        s.Optional("ca_certs"): s.Or(str, None),
+        s.Optional("ciphers"): s.Or(str, None),
     },
 }
 _RUNNER_CONFIG = {
@@ -103,25 +158,18 @@
             **_RUNNER_CONFIG,
             s.Optional(str): _RUNNER_CONFIG,
         },
-        "tracing": _TRACING_CONFIG,
-        "logging": {
-            "formatting": {
-                "trace_id_format": str,
-                "span_id_format": str,
-            }
-        },
     }
 )
 
 
-def migrate_to_v2(*, override_config: dict[str, t.Any]):
+def compat_layer(*, override_config: dict[str, t.Any]):
     # We will use a flattened config to make it easier to migrate,
     # Then we will convert it back to a nested config.
     if depth(override_config) > 1:
         raise ValueError("'override_config' must be a flattened dictionary.") from None
 
-    # We will set version of the migration to 2
-    override_config["version"] = 2
+    if "version" not in override_config:
+        override_config["version"] = 1
 
     # First we migrate api_server field
     # 1. remove api_server.max_request_size (deprecated)
diff --git a/bentoml/_internal/configuration/v1/defaults.yaml b/bentoml/_internal/configuration/v1/defaults.yaml
index e92e18dc298..568c021c206 100644
--- a/bentoml/_internal/configuration/v1/defaults.yaml
+++ b/bentoml/_internal/configuration/v1/defaults.yaml
@@ -1,14 +1,33 @@
 version: 1
 api_server:
   workers: ~ # cpu_count() will be used when null
-  port: 3000
-  host: 0.0.0.0
-  backlog: 2048
   timeout: 60
-  max_request_size: 20971520
+  backlog: 2048
   metrics:
     enabled: true
     namespace: bentoml_api_server
+    duration:
+      # https://github.com/prometheus/client_python/blob/f17a8361ad3ed5bc47f193ac03b00911120a8d81/prometheus_client/metrics.py#L544
+      buckets:
+        [
+          0.005,
+          0.01,
+          0.025,
+          0.05,
+          0.075,
+          0.1,
+          0.25,
+          0.5,
+          0.75,
+          1.0,
+          2.5,
+          5.0,
+          7.5,
+          10.0,
+        ]
+      min: ~
+      max: ~
+      factor: ~
   logging:
     access:
       enabled: true
@@ -16,21 +35,75 @@ api_server:
       request_content_type: true
       response_content_length: true
       response_content_type: true
-  cors:
+      format:
+        trace_id: 032x
+        span_id: 016x
+  ssl:
     enabled: false
-    access_control_allow_origin: ~
-    access_control_allow_credentials: ~
-    access_control_allow_methods: ~
-    access_control_allow_headers: ~
-    access_control_max_age: ~
-    access_control_expose_headers: ~
-
+    certfile: ~
+    keyfile: ~
+    keyfile_password: ~
+    ca_certs: ~
+    version: 17 # ssl.PROTOCOL_TLS_SERVER
+    cert_reqs: 0 # ssl.CERT_NONE
+    ciphers: TLSv1 # default ciphers
+  http:
+    host: 0.0.0.0
+    port: 3000
+    cors:
+      enabled: false
+      allow_origin: ~
+      allow_credentials: ~
+      allow_methods: ~
+      allow_headers: ~
+      allow_origin_regex: ~
+      max_age: ~
+      expose_headers: ~
+  grpc:
+    host: 0.0.0.0
+    port: 3000
+    max_concurrent_streams: ~
+    maximum_concurrent_rpcs: ~
+    max_message_length: -1
+    reflection:
+      enabled: false
+    metrics:
+      host: 0.0.0.0
+      port: 3001
+  tracing:
+    exporter_type: ~
+    sample_rate: ~
+    excluded_urls: ~
+    timeout: ~
+    max_tag_value_length: ~
+    zipkin:
+      endpoint: ~
+    jaeger:
+      protocol: thrift
+      collector_endpoint: ~
+      thrift:
+        agent_host_name: ~
+        agent_port: ~
+        udp_split_oversized_batches: ~
+      grpc:
+        insecure: ~
+    otlp:
+      protocol: ~
+      endpoint: ~
+      compression: ~
+      http:
+        certificate_file: ~
+        headers: ~
+      grpc:
+        headers: ~
+        insecure: ~
 runners:
+  resources: ~
+  timeout: 300
   batching:
     enabled: true
     max_batch_size: 100
     max_latency_ms: 10000
-  resources: ~
   logging:
     access:
       enabled: true
@@ -41,22 +114,3 @@ runners:
   metrics:
     enabled: true
     namespace: bentoml_runner
-  timeout: 300
-
-tracing:
-  type: zipkin
-  sample_rate: ~
-  excluded_urls: ~
-  zipkin:
-    url: ~
-  jaeger:
-    address: ~
-    port: ~
-  otlp:
-    protocol: ~
-    url: ~
-
-logging:
-  formatting:
-    trace_id_format: 032x
-    span_id_format: 016x
diff --git a/bentoml/_internal/configuration/v2/__init__.py b/bentoml/_internal/configuration/v2/__init__.py
deleted file mode 100644
index d88be64b008..00000000000
--- a/bentoml/_internal/configuration/v2/__init__.py
+++ /dev/null
@@ -1,157 +0,0 @@
-from __future__ import annotations
-
-import re
-import typing as t
-
-import schema as s
-
-from ..helpers import ensure_range
-from ..helpers import ensure_larger_than
-from ..helpers import is_valid_ip_address
-from ..helpers import ensure_iterable_type
-from ..helpers import validate_tracing_type
-from ..helpers import validate_otlp_protocol
-from ..helpers import ensure_larger_than_zero
-from ...utils.metrics import DEFAULT_BUCKET
-
-TRACING_CFG = {
-    "exporter_type": s.Or(s.And(str, s.Use(str.lower), validate_tracing_type), None),
-    "sample_rate": s.Or(s.And(float, ensure_range(0, 1)), None),
-    "timeout": s.Or(s.And(int, ensure_larger_than_zero), None),
-    "max_tag_value_length": s.Or(int, None),
-    "excluded_urls": s.Or([str], str, None),
-    "zipkin": {
-        "endpoint": s.Or(str, None),
-    },
-    "jaeger": {
-        "protocol": s.Or(
-            s.And(str, s.Use(str.lower), lambda d: d in ["thrift", "grpc"]),
-            None,
-        ),
-        "collector_endpoint": s.Or(str, None),
-        "thrift": {
-            "agent_host_name": s.Or(str, None),
-            "agent_port": s.Or(int, None),
-            "udp_split_oversized_batches": s.Or(bool, None),
-        },
-        "grpc": {
-            "insecure": s.Or(bool, None),
-        },
-    },
-    "otlp": {
-        "protocol": s.Or(s.And(str, s.Use(str.lower), validate_otlp_protocol), None),
-        "endpoint": s.Or(str, None),
-        "compression": s.Or(
-            s.And(str, lambda d: d in {"gzip", "none", "deflate"}), None
-        ),
-        "http": {
-            "certificate_file": s.Or(str, None),
-            "headers": s.Or(dict, None),
-        },
-        "grpc": {
-            "insecure": s.Or(bool, None),
-            "headers": s.Or(lambda d: isinstance(d, t.Sequence), None),
-        },
-    },
-}
-_API_SERVER_CONFIG = {
-    "workers": s.Or(s.And(int, ensure_larger_than_zero), None),
-    "timeout": s.And(int, ensure_larger_than_zero),
-    "backlog": s.And(int, ensure_larger_than(64)),
-    "metrics": {
-        "enabled": bool,
-        "namespace": str,
-        s.Optional("duration"): {
-            s.Optional("buckets", default=DEFAULT_BUCKET): s.Or(
-                s.And(list, ensure_iterable_type(float)), None
-            ),
-            s.Optional("min"): s.Or(s.And(float, ensure_larger_than_zero), None),
-            s.Optional("max"): s.Or(s.And(float, ensure_larger_than_zero), None),
-            s.Optional("factor"): s.Or(s.And(float, ensure_larger_than(1.0)), None),
-        },
-    },
-    "logging": {
-        "access": {
-            "enabled": bool,
-            "request_content_length": s.Or(bool, None),
-            "request_content_type": s.Or(bool, None),
-            "response_content_length": s.Or(bool, None),
-            "response_content_type": s.Or(bool, None),
-            "format": {
-                "trace_id": str,
-                "span_id": str,
-            },
-        },
-    },
-    "http": {
-        "host": s.And(str, is_valid_ip_address),
-        "port": s.And(int, ensure_larger_than_zero),
-        "cors": {
-            "enabled": bool,
-            "allow_origin": s.Or(str, None),
-            "allow_origin_regex": s.Or(s.And(str, s.Use(re.compile)), None),
-            "allow_credentials": s.Or(bool, None),
-            "allow_headers": s.Or([str], str, None),
-            "allow_methods": s.Or([str], str, None),
-            "max_age": s.Or(int, None),
-            "expose_headers": s.Or([str], str, None),
-        },
-    },
-    "grpc": {
-        "host": s.And(str, is_valid_ip_address),
-        "port": s.And(int, ensure_larger_than_zero),
-        "metrics": {
-            "port": s.And(int, ensure_larger_than_zero),
-            "host": s.And(str, is_valid_ip_address),
-        },
-        "reflection": {"enabled": bool},
-        "max_concurrent_streams": s.Or(int, None),
-        "max_message_length": s.Or(int, None),
-        "maximum_concurrent_rpcs": s.Or(int, None),
-    },
-    "tracing": TRACING_CFG,
-    s.Optional("ssl"): {
-        "enabled": bool,
-        s.Optional("certfile"): s.Or(str, None),
-        s.Optional("keyfile"): s.Or(str, None),
-        s.Optional("keyfile_password"): s.Or(str, None),
-        s.Optional("version"): s.Or(s.And(int, ensure_larger_than_zero), None),
-        s.Optional("cert_reqs"): s.Or(int, None),
-        s.Optional("ca_certs"): s.Or(str, None),
-        s.Optional("ciphers"): s.Or(str, None),
-    },
-}
-_RUNNER_CONFIG = {
-    s.Optional("batching"): {
-        s.Optional("enabled"): bool,
-        s.Optional("max_batch_size"): s.And(int, ensure_larger_than_zero),
-        s.Optional("max_latency_ms"): s.And(int, ensure_larger_than_zero),
-    },
-    # NOTE: there is a distinction between being unset and None here; if set to 'None'
-    # in configuration for a specific runner, it will override the global configuration.
-    s.Optional("resources"): s.Or({s.Optional(str): object}, lambda s: s == "system", None),  # type: ignore (incomplete schema typing)
-    s.Optional("logging"): {
-        s.Optional("access"): {
-            s.Optional("enabled"): bool,
-            s.Optional("request_content_length"): s.Or(bool, None),
-            s.Optional("request_content_type"): s.Or(bool, None),
-            s.Optional("response_content_length"): s.Or(bool, None),
-            s.Optional("response_content_type"): s.Or(bool, None),
-        },
-    },
-    s.Optional("metrics"): {
-        "enabled": bool,
-        "namespace": str,
-    },
-    s.Optional("timeout"): s.And(int, ensure_larger_than_zero),
-}
-SCHEMA = s.Schema(
-    {
-        "version": s.And(int, lambda v: v == 2),
-        "api_server": _API_SERVER_CONFIG,
-        "runners": {
-            **_RUNNER_CONFIG,
-            s.Optional(str): _RUNNER_CONFIG,
-        },
-    }
-)
diff --git a/bentoml/_internal/configuration/v2/defaults.yaml b/bentoml/_internal/configuration/v2/defaults.yaml
deleted file mode 100644
index b952ea2feaf..00000000000
--- a/bentoml/_internal/configuration/v2/defaults.yaml
+++ /dev/null
@@ -1,117 +0,0 @@
-version: 2
-api_server:
-  workers: ~ # cpu_count() will be used when null
-  timeout: 60
-  backlog: 2048
-  metrics:
-    enabled: true
-    namespace: bentoml_api_server
-    duration:
-      # https://github.com/prometheus/client_python/blob/f17a8361ad3ed5bc47f193ac03b00911120a8d81/prometheus_client/metrics.py#L544
-      buckets:
-        [
-          0.005,
-          0.01,
-          0.025,
-          0.05,
-          0.075,
-          0.1,
-          0.25,
-          0.5,
-          0.75,
-          1.0,
-          2.5,
-          5.0,
-          7.5,
-          10.0,
-        ]
-      min: ~
-      max: ~
-      factor: ~
-  logging:
-    access:
-      enabled: true
-      request_content_length: true
-      request_content_type: true
-      response_content_length: true
-      response_content_type: true
-      format:
-        trace_id: 032x
-        span_id: 016x
-  ssl:
-    enabled: false
-    certfile: ~
-    keyfile: ~
-    keyfile_password: ~
-    ca_certs: ~
-    version: 17 # ssl.PROTOCOL_TLS_SERVER
-    cert_reqs: 0 # ssl.CERT_NONE
-    ciphers: TLSv1 # default ciphers
-  http:
-    host: 0.0.0.0
-    port: 3000
-    cors:
-      enabled: false
-      allow_origin: ~
-      allow_credentials: ~
-      allow_methods: ~
-      allow_headers: ~
-      allow_origin_regex: ~
-      max_age: ~
-      expose_headers: ~
-  grpc:
-    host: 0.0.0.0
-    port: 3000
-    max_concurrent_streams: ~
-    maximum_concurrent_rpcs: ~
-    max_message_length: -1
-    reflection:
-      enabled: false
-    metrics:
-      host: 0.0.0.0
-      port: 3001
-  tracing:
-    exporter_type: ~
-    sample_rate: ~
-    excluded_urls: ~
-    timeout: ~
-    max_tag_value_length: ~
-    zipkin:
-      endpoint: ~
-    jaeger:
-      protocol: thrift
-      collector_endpoint: ~
-      thrift:
-        agent_host_name: ~
-        agent_port: ~
-        udp_split_oversized_batches: ~
-      grpc:
-        insecure: ~
-    otlp:
-      protocol: ~
-      endpoint: ~
-      compression: ~
-      http:
-        certificate_file: ~
-        headers: ~
-      grpc:
-        headers: ~
-        insecure: ~
-
-runners:
-  timeout: 300
-  resources: ~
-  batching:
-    enabled: true
-    max_batch_size: 100
-    max_latency_ms: 10000
-  logging:
-    access:
-      enabled: true
-      request_content_length: true
-      request_content_type: true
-      response_content_length: true
-      response_content_type: true
-  metrics:
-    enabled: true
-    namespace: bentoml_runner
diff --git a/bentoml/_internal/runner/runner_handle/remote.py b/bentoml/_internal/runner/runner_handle/remote.py
index d557cb5a84c..b1dbfc8d72a 100644
--- a/bentoml/_internal/runner/runner_handle/remote.py
+++ b/bentoml/_internal/runner/runner_handle/remote.py
@@ -89,9 +89,7 @@ def _get_conn(self) -> BaseConnector:
         return self._conn
 
     @property
-    def _client(
-        self,
-    ) -> ClientSession:
+    def _client(self) -> ClientSession:
         import aiohttp
 
         if (
@@ -111,7 +109,7 @@ def strip_query_params(url: yarl.URL) -> str:
                 trace_configs=[
                     create_trace_config(
                         # Remove all query params from the URL attribute on the span.
-                        url_filter=strip_query_params,  # type: ignore
+                        url_filter=strip_query_params,
                         tracer_provider=BentoMLContainer.tracer_provider.get(),
                     )
                 ],
diff --git a/tests/e2e/bento_server_http/configs/cors_enabled.yml b/tests/e2e/bento_server_http/configs/cors_enabled.yml
index d774d957050..f891666791f 100644
--- a/tests/e2e/bento_server_http/configs/cors_enabled.yml
+++ b/tests/e2e/bento_server_http/configs/cors_enabled.yml
@@ -1,4 +1,4 @@
-version: 2
+version: 1
 api_server:
   http:
     cors: # standard: https://fetch.spec.whatwg.org/#http-cors-protocol
diff --git a/tests/e2e/bento_server_http/configs/default.yml b/tests/e2e/bento_server_http/configs/default.yml
index 3a8d295bd1e..94aba7999bf 100644
--- a/tests/e2e/bento_server_http/configs/default.yml
+++ b/tests/e2e/bento_server_http/configs/default.yml
@@ -1,4 +1,4 @@
-version: 2
+version: 1
 api_server:
   http:
     cors: # standard: https://fetch.spec.whatwg.org/#http-cors-protocol
diff --git a/tests/unit/_internal/configuration/test_containers.py b/tests/unit/_internal/configuration/test_containers.py
index bcda02ec316..253857f8aa3 100644
--- a/tests/unit/_internal/configuration/test_containers.py
+++ b/tests/unit/_internal/configuration/test_containers.py
@@ -14,7 +14,7 @@
 @pytest.mark.usefixtures("container_from_file")
 def test_validate_configuration(container_from_file: t.Callable[[str], ConfigDictType]):
     CONFIG = """\
-version: 2
+version: 1
 api_server:
   http:
     host: 0.0.0.0
@@ -23,7 +23,7 @@ def test_validate_configuration(container_from_file: t.Callable[[str], ConfigDic
     assert config["api_server"]["http"]["host"] == "0.0.0.0"
 
     INVALID_CONFIG = """\
-version: 2
+version: 1
 api_server:
   cors:
     max_age: 12345
@@ -44,7 +44,7 @@ def test_containers_from_envvar(
     assert config["api_server"]["http"]["port"] == 5000
 
 
-@pytest.mark.parametrize("version", [None, 1, 2])
+@pytest.mark.parametrize("version", [None, 1])
 @pytest.mark.usefixtures("container_from_file")
 def test_bentoml_configuration_runner_override(
     container_from_file: t.Callable[[str], ConfigDictType], version: int | None
diff --git a/tests/unit/_internal/configuration/test_v1_migration.py b/tests/unit/_internal/configuration/test_v1_migration.py
index f631ab04cdf..d3ede9886aa 100644
--- a/tests/unit/_internal/configuration/test_v1_migration.py
+++ b/tests/unit/_internal/configuration/test_v1_migration.py
@@ -45,7 +45,7 @@ def test_backward_from_envvar(
     envvar = 'version=1 tracing.type="jaeger" tracing.jaeger.address="localhost" tracing.jaeger.port=6831 tracing.sample_rate=0.7'
     with caplog.at_level(logging.WARNING):
         bentoml_cfg = container_from_envvar(envvar)
-    assert bentoml_cfg["version"] == 2
+    assert bentoml_cfg["version"] == 1
     assert bentoml_cfg["api_server"]["tracing"]["exporter_type"] == "jaeger"
 
     assert (