xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 17:35:48 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Fix None value handling in trace span creation for cancelled requests (#20272)

Browse Source
This commit is contained in:
Bram 2025-07-30 14:44:02 -07:00 committed by GitHub
parent 287f527f54
commit 601f856d56
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 20 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
vllm/engine/llm_engine.py
View File

@ -1862,8 +1862,14 @@ class LLMEngine:
context=trace_context, context=trace_context,
start_time=arrival_time_nano_seconds) as seq_span: start_time=arrival_time_nano_seconds) as seq_span:
metrics = seq_group.metrics metrics = seq_group.metrics
ttft = metrics.first_token_time - metrics.arrival_time
e2e_time = metrics.finished_time - metrics.arrival_time # Handle potential None values for cancelled/aborted requests
ttft = (metrics.first_token_time - metrics.arrival_time
if metrics.first_token_time is not None else None)
e2e_time = (metrics.finished_time - metrics.arrival_time
if metrics.finished_time is not None else None)
seq_span.set_attribute(SpanAttributes.GEN_AI_RESPONSE_MODEL, seq_span.set_attribute(SpanAttributes.GEN_AI_RESPONSE_MODEL,
self.model_config.model) self.model_config.model)
seq_span.set_attribute(SpanAttributes.GEN_AI_REQUEST_ID, seq_span.set_attribute(SpanAttributes.GEN_AI_REQUEST_ID,
@ -1886,11 +1892,18 @@ class LLMEngine:
seq.get_output_len() seq.get_output_len()
for seq in seq_group.get_finished_seqs() for seq in seq_group.get_finished_seqs()
])) ]))
seq_span.set_attribute(SpanAttributes.GEN_AI_LATENCY_TIME_IN_QUEUE,
metrics.time_in_queue) # Only set timing attributes if the values are available
seq_span.set_attribute( if metrics.time_in_queue is not None:
SpanAttributes.GEN_AI_LATENCY_TIME_TO_FIRST_TOKEN, ttft) seq_span.set_attribute(
seq_span.set_attribute(SpanAttributes.GEN_AI_LATENCY_E2E, e2e_time) SpanAttributes.GEN_AI_LATENCY_TIME_IN_QUEUE,
metrics.time_in_queue)
if ttft is not None:
seq_span.set_attribute(
SpanAttributes.GEN_AI_LATENCY_TIME_TO_FIRST_TOKEN, ttft)
if e2e_time is not None:
seq_span.set_attribute(SpanAttributes.GEN_AI_LATENCY_E2E,
e2e_time)
if metrics.scheduler_time is not None: if metrics.scheduler_time is not None:
seq_span.set_attribute( seq_span.set_attribute(
SpanAttributes.GEN_AI_LATENCY_TIME_IN_SCHEDULER, SpanAttributes.GEN_AI_LATENCY_TIME_IN_SCHEDULER,