Skip to content

fix(plugin): fix ai latency metric doesn't correctly calculated issue #14639

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jun 16, 2025
+15 −9
Merged

fix(plugin): fix ai latency metric doesn't correctly calculated issue #14639

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9e76523
fix(plugin): fix ai latency metric doesn't correctly calculated issue
ProBrian Jun 6, 2025
b086fcc
fix(e2e): add missing comment
ProBrian Jun 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions kong/plugins/prometheus/exporter.lua
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -401,15 +401,15 @@ local function log(message, serialized)
metrics.ai_llm_cost:inc(ai_metrics.usage.cost, labels_table_ai_llm_status)
end

if ai_metrics.meta and ai_metrics.meta.llm_latency and ai_metrics.meta.llm_latency > 0 then
if ai_metrics.meta and ai_metrics.meta.llm_latency and ai_metrics.meta.llm_latency >= 0 then
metrics.ai_llm_provider_latency:observe(ai_metrics.meta.llm_latency, labels_table_ai_llm_status)
end

if ai_metrics.cache and ai_metrics.cache.fetch_latency and ai_metrics.cache.fetch_latency > 0 then
if ai_metrics.cache and ai_metrics.cache.fetch_latency and ai_metrics.cache.fetch_latency >= 0 then
metrics.ai_cache_fetch_latency:observe(ai_metrics.cache.fetch_latency, labels_table_ai_llm_status)
end

if ai_metrics.cache and ai_metrics.cache.embeddings_latency and ai_metrics.cache.embeddings_latency > 0 then
if ai_metrics.cache and ai_metrics.cache.embeddings_latency and ai_metrics.cache.embeddings_latency >= 0 then
metrics.ai_cache_embeddings_latency:observe(ai_metrics.cache.embeddings_latency, labels_table_ai_llm_status)
end

Expand Down
18 changes: 12 additions & 6 deletions spec/03-plugins/26-prometheus/02-access_spec.lua
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,8 @@
local helpers = require "spec.helpers"
local shell = require "resty.shell"
local pl_file = require "pl.file"
local timeout = 10
local step = 1

local UUID_PATTERN = "%x%x%x%x%x%x%x%x%-%x%x%x%x%-%x%x%x%x%-%x%x%x%x%-%x%x%x%x%x%x%x%x%x%x%x%x"

Expand Down Expand Up @@ -820,14 +822,16 @@ describe("Plugin: prometheus (access) AI metrics", function()
assert.res_status(200, res)

local body
-- wait until the histogram observe finished and get the correct metrics.
helpers.wait_until(function()
local res = assert(admin_client:send {
method = "GET",
path = "/metrics",
})
body = assert.res_status(200, res)
return res.status == 200
end)
return body:find('ai_llm_provider_latency_ms_bucket{ai_provider="openai",ai_model="gpt-3.5-turbo",cache_status="",vector_db="",embeddings_provider="",embeddings_model="",workspace="default",le="+Inf"} 1',
nil, true)
end, timeout, step)

assert.matches('kong_nginx_metric_errors_total 0', body, nil, true)
assert.matches('http_requests_total{service="empty_service",route="http-route",code="200",source="service",workspace="default",consumer=""} 2', body, nil, true)
Expand Down Expand Up @@ -864,8 +868,9 @@ describe("Plugin: prometheus (access) AI metrics", function()
path = "/metrics",
})
body = assert.res_status(200, res)
return res.status == 200
end)
return body:find('ai_llm_provider_latency_ms_bucket{ai_provider="openai",ai_model="gpt-3.5-turbo",cache_status="",vector_db="",embeddings_provider="",embeddings_model="",workspace="default",le="+Inf"} 2',
nil, true)
end, timeout, step)

assert.matches('kong_nginx_metric_errors_total 0', body, nil, true)
assert.matches('http_requests_total{service="empty_service",route="http-route",code="200",source="service",workspace="default",consumer=""} 3', body, nil, true)
Expand Down Expand Up @@ -898,8 +903,9 @@ describe("Plugin: prometheus (access) AI metrics", function()
path = "/metrics",
})
body = assert.res_status(200, res)
return res.status == 200
end)
return body:find('ai_llm_provider_latency_ms_bucket{ai_provider="openai",ai_model="gpt-3.5-turbo",cache_status="",vector_db="",embeddings_provider="",embeddings_model="",workspace="default",le="+Inf"} 2',
nil, true)
end, timeout, step)

assert.matches('http_requests_total{service="empty_service",route="http-route",code="400",source="kong",workspace="default",consumer=""} 1', body, nil, true)
assert.matches('kong_nginx_metric_errors_total 0', body, nil, true)
Expand Down
Loading