fix: aws/01_lambda failures

- fix: handle `str` internal ids with all digit chars
- rename the methods to handle internal ids

Signed-off-by: Varsha GS <[email protected]>

Author: GSVarsha

src/instana/collector/utils.py

@@ -5,7 +5,7 @@
 from opentelemetry.trace.span import format_span_id
 from opentelemetry.trace import SpanKind
 
-from instana.util.ids import hex_id, header_to_32
+from instana.util.ids import hex_id, internal_id
 
 if TYPE_CHECKING:
     from instana.span.base_span import BaseSpan
@@ -24,7 +24,7 @@ def format_span(
         span.t = format_span_id(span.t)
         span.s = format_span_id(span.s)
         span.p = format_span_id(span.p) if span.p else None
-        span.lt = hex_id(header_to_32(span.lt)) if hasattr(span, "lt") else None
+        span.lt = hex_id(internal_id(span.lt)) if hasattr(span, "lt") else None
         if isinstance(span.k, SpanKind):
             span.k = span.k.value if not span.k is SpanKind.INTERNAL else 3
         spans.append(span)

src/instana/instrumentation/aws/lambda_inst.py

@@ -11,6 +11,7 @@
 
 import wrapt
 from opentelemetry.semconv.trace import SpanAttributes
+from opentelemetry.trace.span import format_span_id
 
 from instana import get_aws_lambda_handler
 from instana.instrumentation.aws.triggers import enrich_lambda_span, get_context
@@ -44,7 +45,7 @@ def lambda_handler_with_instana(
             result = wrapped(*args, **kwargs)
 
             if isinstance(result, dict):
-                server_timing_value = define_server_timing(span.context.trace_id)
+                server_timing_value = define_server_timing(format_span_id(span.context.trace_id))
                 if "headers" in result:
                     result["headers"]["Server-Timing"] = server_timing_value
                 elif "multiValueHeaders" in result:

src/instana/propagators/base_propagator.py

@@ -8,7 +8,7 @@
 
 from instana.log import logger
 from instana.span_context import SpanContext
-from instana.util.ids import header_to_id, header_to_long_id, hex_id, header_to_32, header_to_16, hex_id_16
+from instana.util.ids import header_to_id, header_to_long_id, hex_id, internal_id, internal_id_limited, hex_id_limited
 from instana.w3c_trace_context.traceparent import Traceparent
 from instana.w3c_trace_context.tracestate import Tracestate
 
@@ -149,7 +149,7 @@ def _get_participating_trace_context(self, span_context: SpanContext):
         if span_context.suppression:
             return traceparent, tracestate
 
-        tracestate = self._ts.update_tracestate(tracestate, hex_id_16(span_context.trace_id), hex_id(span_context.span_id))
+        tracestate = self._ts.update_tracestate(tracestate, hex_id_limited(span_context.trace_id), hex_id(span_context.span_id))
         return traceparent, tracestate
 
     def __determine_span_context(
@@ -219,7 +219,7 @@ def __determine_span_context(
                 instana_ancestor = self._ts.get_instana_ancestor(tracestate)
 
             if disable_traceparent == "":
-                ctx_trace_id = hex_id_16(tp_trace_id)
+                ctx_trace_id = hex_id_limited(tp_trace_id)
                 ctx_span_id = tp_parent_id
                 ctx_synthetic = synthetic
                 ctx_trace_parent = True
@@ -241,11 +241,21 @@ def __determine_span_context(
             ctx_traceparent = traceparent
             ctx_tracestate = tracestate
 
+        if ctx_trace_id:
+            if isinstance(ctx_trace_id, int):
+                # check if ctx_trace_id is a valid internal trace id
+                if (ctx_trace_id <=  2**64 - 1):
+                    trace_id = ctx_trace_id
+                else:
+                    trace_id = internal_id(hex_id_limited(ctx_trace_id))
+            else:
+                trace_id = internal_id(ctx_trace_id)
+        else:
+            trace_id = INVALID_TRACE_ID
+
         return SpanContext(
-            # trace_id=int(ctx_trace_id) if ctx_trace_id else INVALID_TRACE_ID,
-            # span_id=int(ctx_span_id) if ctx_span_id else INVALID_SPAN_ID,
-            trace_id=header_to_32(hex_id_16(header_to_32(ctx_trace_id))) if ctx_trace_id else INVALID_TRACE_ID,
-            span_id=header_to_16(ctx_span_id) if ctx_span_id else INVALID_SPAN_ID,
+            trace_id=trace_id,
+            span_id=internal_id_limited(ctx_span_id) if ctx_span_id else INVALID_SPAN_ID,
             is_remote=False,
             level=ctx_level,
             synthetic=ctx_synthetic,

src/instana/propagators/http_propagator.py

@@ -4,7 +4,7 @@
 
 from instana.log import logger
 from instana.propagators.base_propagator import BasePropagator
-from instana.util.ids import define_server_timing, hex_id_16
+from instana.util.ids import define_server_timing, hex_id_limited
 
 from opentelemetry.trace.span import format_span_id
 
@@ -55,7 +55,7 @@ def inject_key_value(carrier, key, value):
             if span_context.suppression:
                 return
 
-            inject_key_value(carrier, self.HEADER_KEY_T, hex_id_16(trace_id))
+            inject_key_value(carrier, self.HEADER_KEY_T, hex_id_limited(trace_id))
             inject_key_value(carrier, self.HEADER_KEY_S, format_span_id(span_id))
             inject_key_value(
                 carrier, self.HEADER_KEY_SERVER_TIMING, define_server_timing(trace_id)

src/instana/span/base_span.py

@@ -5,7 +5,6 @@
 
 from instana.log import logger
 from instana.util import DictionaryOfStan
-from instana.util.ids import hex_id_16, header_to_32
 from instana.span.kind import ENTRY_SPANS
 
 if TYPE_CHECKING:
@@ -23,7 +22,7 @@ def __repr__(self) -> str:
 
     def __init__(self, span: Type["Span"], source, **kwargs) -> None:
         # pylint: disable=invalid-name
-        self.t = header_to_32(hex_id_16(span.context.trace_id)) 
+        self.t = span.context.trace_id
         self.p = span.parent_id
         self.s = span.context.span_id
         self.ts = round(span.start_time / 10**6)

src/instana/util/ids.py

@@ -92,61 +92,82 @@ def header_to_id(header: Union[bytes, str]) -> int:
 def hex_id(id: Union[int, str]) -> str:
     """
     Returns the hexadecimal representation of the given ID.
+    Left pad with zeros when the length is not equal to 16
     """
 
     hex_id = hex(int(id))[2:]
-    if len(hex_id) < 16:
+    length = len(hex_id)
+    # Left pad ID with zeros
+    if length < 16:
         hex_id = hex_id.zfill(16)
-    elif len(hex_id) > 16 and len(hex_id) < 32:
+    elif length > 16 and length < 32:
         hex_id = hex_id.zfill(32)
     return hex_id
 
-def hex_id_16(id: Union[int, str]) -> str:
+def hex_id_limited(id: Union[int, str]) -> str:
     """
     Returns the hexadecimal representation of the given ID.
+    Limit longer IDs to 16 characters 
     """
-
-    hex_id = hex(int(id))[2:]
-    length = len(hex_id)
-    if length < 16:
-        hex_id = hex_id.zfill(16)
-    elif length > 16:
-        # Phase 0: Discard everything but the last 16byte
-        hex_id = hex_id[-16:]
-    return hex_id
-
+    try:
+        hex_id = hex(int(id))[2:]
+        length = len(hex_id)
+        if length < 16:
+            # Left pad ID with zeros
+            hex_id = hex_id.zfill(16)
+        elif length > 16:
+            # Phase 0: Discard everything but the last 16byte
+            hex_id = hex_id[-16:]
+        return hex_id
+    except ValueError: # ValueError: invalid literal for int() with base 10:
+        return id
 
 def define_server_timing(trace_id: Union[int, str]) -> str:
     # Note: The key `intid` is short for Instana Trace ID.
-    return f"intid;desc={hex_id_16(trace_id)}"
+    return f"intid;desc={hex_id_limited(trace_id)}"
 
 
-def header_to_32(header) -> int:
-    if isinstance(header, int):
-        return header
+def internal_id(id: Union[int, str]) -> int:
+    """
+    Returns a valid id to be used internally. Handles both str and int types.
+    """
+    if isinstance(id, int):
+        return id
+
+    if isinstance(id, str) and id.isdigit():
+        return int(id)
 
     try:
-        if len(header) < 16:
+        if len(id) < 16:
             # Left pad ID with zeros
-            header = header.zfill(16)
+            id = id.zfill(16)
 
-        return int(header, 16)
+        # hex string -> int
+        return int(id, 16)
     except ValueError:
         return INVALID_TRACE_ID
 
-def header_to_16(header) -> int:
-    if isinstance(header, int):
-        return header
+def internal_id_limited(id: Union[int, str]) -> int:
+    """
+    Returns a valid id to be used internally. Handles both str and int types.
+    Note: Limits the hex string to 16 chars before conversion.
+    """
+    if isinstance(id, int):
+        return id
+
+    if isinstance(id, str) and id.isdigit():
+        return int(id)
 
     try:
-        length = len(header)
+        length = len(id)
         if length < 16:
             # Left pad ID with zeros
-            header = header.zfill(16)
+            id = id.zfill(16)
         elif length > 16:
             # Phase 0: Discard everything but the last 16byte
-            header = header[-16:]
+            id = id[-16:]
 
-        return int(header, 16)
+        # hex string -> int
+        return int(id, 16)
     except ValueError:
-        return INVALID_SPAN_ID
+        return INVALID_SPAN_ID

tests/propagators/test_http_propagator.py

@@ -14,7 +14,7 @@
 
 from instana.propagators.http_propagator import HTTPPropagator
 from instana.span_context import SpanContext
-from instana.util.ids import header_to_id
+from instana.util.ids import header_to_long_id, internal_id
 
 
 class TestHTTPPropagator:
@@ -62,7 +62,6 @@ def test_extract_carrier_dict(
         span_id: int,
         _instana_long_tracer_id: str,
         _instana_span_id: str,
-        _long_tracer_id: int,
         _trace_id: int,
         _span_id: int,
         _traceparent: str,
@@ -82,7 +81,7 @@ def test_extract_carrier_dict(
         assert ctx.correlation_type == "web"
         assert not ctx.instana_ancestor
         assert ctx.level == 1
-        assert ctx.long_trace_id == header_to_id(_instana_long_tracer_id)
+        assert ctx.long_trace_id == header_to_long_id(_instana_long_tracer_id)
         assert ctx.span_id == _span_id
         assert not ctx.synthetic
         assert ctx.trace_id == _trace_id
@@ -92,8 +91,8 @@ def test_extract_carrier_dict(
 
     def test_extract_carrier_list(
         self,
-        trace_id: int,
-        span_id: int,
+        _trace_id: int,
+        _span_id: int,
         _instana_long_tracer_id: str,
         _instana_span_id: str,
         _traceparent: str,
@@ -106,8 +105,8 @@ def test_extract_carrier_list(
             ("connection", "keep-alive"),
             ("traceparent", _traceparent),
             ("tracestate", _tracestate),
-            ("X-INSTANA-T", f"{trace_id}"),
-            ("X-INSTANA-S", f"{span_id}"),
+            ("X-INSTANA-T", f"{_trace_id}"),
+            ("X-INSTANA-S", f"{_span_id}"),
             ("X-INSTANA-L", "1"),
         ]
 
@@ -118,9 +117,9 @@ def test_extract_carrier_list(
         assert not ctx.instana_ancestor
         assert ctx.level == 1
         assert not ctx.long_trace_id
-        assert ctx.span_id == span_id
+        assert ctx.span_id == _span_id
         assert not ctx.synthetic
-        assert ctx.trace_id == trace_id
+        assert ctx.trace_id == internal_id(_trace_id)
         assert not ctx.trace_parent
         assert ctx.traceparent == f"00-{_instana_long_tracer_id}-{_instana_span_id}-01"
         assert ctx.tracestate == _tracestate
@@ -199,7 +198,7 @@ def test_extract_carrier_dict_corrupted_level_header(
         assert ctx.correlation_type == "web"
         assert not ctx.instana_ancestor
         assert ctx.level == 1
-        assert ctx.long_trace_id == header_to_id(_instana_long_tracer_id)
+        assert ctx.long_trace_id == header_to_long_id(_instana_long_tracer_id)
         assert ctx.span_id == _span_id
         assert not ctx.synthetic
         assert ctx.trace_id == _trace_id
@@ -209,16 +208,16 @@ def test_extract_carrier_dict_corrupted_level_header(
 
     def test_extract_carrier_dict_level_header_not_splitable(
         self,
-        trace_id: int,
-        span_id: int,
+        _trace_id: int,
+        _span_id: int,
         _traceparent: str,
         _tracestate: str,
     ) -> None:
         carrier = {
             "traceparent": _traceparent,
             "tracestate": _tracestate,
-            "X-INSTANA-T": f"{trace_id}",
-            "X-INSTANA-S": f"{span_id}",
+            "X-INSTANA-T": f"{_trace_id}",
+            "X-INSTANA-S": f"{_span_id}",
             "X-INSTANA-L": ["1"],
         }
 
@@ -229,9 +228,9 @@ def test_extract_carrier_dict_level_header_not_splitable(
         assert not ctx.instana_ancestor
         assert ctx.level == 1
         assert not ctx.long_trace_id
-        assert ctx.span_id == span_id
+        assert ctx.span_id == _span_id
         assert not ctx.synthetic
-        assert ctx.trace_id == trace_id
+        assert ctx.trace_id == internal_id(_trace_id)
         assert not ctx.trace_parent
         assert ctx.traceparent == _traceparent
         assert ctx.tracestate == _tracestate
@@ -304,7 +303,7 @@ def test_w3c_off_x_instana_l_0(
         # Assert that the traceparent is propagated when it is enabled
         if "traceparent" in carrier_header.keys():
             assert ctx.traceparent
-            tp_trace_id = header_to_id(carrier_header["traceparent"].split("-")[1])
+            tp_trace_id = header_to_long_id(carrier_header["traceparent"].split("-")[1])
         else:
             assert not ctx.traceparent
             tp_trace_id = ctx.trace_id

tests_aws/01_lambda/test_lambda.py

@@ -221,9 +221,9 @@ def test_custom_service_name(self, trace_id: int, span_id: int) -> None:
 
         span = payload["spans"].pop()
         assert span.n == "aws.lambda.entry"
-        assert span.t == hex(trace_id)[2:]
+        assert span.t == hex_id(trace_id)
         assert span.s
-        assert span.p == hex(span_id)[2:]
+        assert span.p == hex_id(span_id)
         assert span.ts
 
         server_timing_value = f"intid;desc={hex_id(trace_id)}"
@@ -293,9 +293,9 @@ def test_api_gateway_trigger_tracing(self, trace_id: int, span_id: int) -> None:
 
         span = payload["spans"].pop()
         assert span.n == "aws.lambda.entry"
-        assert span.t == hex(trace_id)[2:]
+        assert span.t == hex_id(trace_id)
         assert span.s
-        assert span.p == hex(span_id)[2:]
+        assert span.p == hex_id(span_id)
         assert span.ts
 
         server_timing_value = f"intid;desc={hex_id(trace_id)}"
@@ -405,9 +405,9 @@ def test_application_lb_trigger_tracing(self, trace_id: int, span_id: int) -> No
 
         span = payload["spans"].pop()
         assert span.n == "aws.lambda.entry"
-        assert span.t == hex(trace_id)[2:]
+        assert span.t == hex_id(trace_id)
         assert span.s
-        assert span.p == hex(span_id)[2:]
+        assert span.p == hex_id(span_id)
         assert span.ts
 
         server_timing_value = f"intid;desc={hex_id(trace_id)}"
@@ -803,9 +803,9 @@ def __validate_result_and_payload_for_gateway_v2_trace(self, result: Dict[str, A
 
         span = payload["spans"].pop()
         assert span.n == "aws.lambda.entry"
-        assert span.t == hex(int("0000000000001234"))[2:].zfill(16)
+        assert span.t == hex_id("0000000000001234")
         assert span.s
-        assert span.p == hex(int("0000000000004567"))[2:].zfill(16)
+        assert span.p == hex_id("0000000000004567")
         assert span.ts
 
         server_timing_value = f"intid;desc={hex_id(int('0000000000001234'))}"