This is an automated email from the ASF dual-hosted git repository.
nic-6443 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/apisix.git
The following commit(s) were added to refs/heads/master by this push:
new 494dc4229 feat: label WebSocket sessions with request_type=websocket
(#13909)
494dc4229 is described below
commit 494dc42299d3bd9e9edc5be13c713a5946d525a1
Author: Mohammad Izzraff Janius
<[email protected]>
AuthorDate: Thu Sep 3 15:24:51 2026 +0900
feat: label WebSocket sessions with request_type=websocket (#13909)
---
apisix/init.lua | 4 +
docs/en/latest/plugins/prometheus.md | 6 +-
docs/zh/latest/plugins/prometheus.md | 6 +-
t/plugin/prometheus-websocket.t | 183 +++++++++++++++++++++++++++++++++++
4 files changed, 193 insertions(+), 6 deletions(-)
diff --git a/apisix/init.lua b/apisix/init.lua
index 7e6675850..8953eba1b 100644
--- a/apisix/init.lua
+++ b/apisix/init.lua
@@ -1024,6 +1024,10 @@ function _M.http_header_filter_phase()
set_resp_upstream_status(up_status)
end
+ if ngx.status == 101 then
+ ngx_var.request_type = "websocket"
+ end
+
common_phase("header_filter")
local api_ctx = ngx.ctx.api_ctx
diff --git a/docs/en/latest/plugins/prometheus.md
b/docs/en/latest/plugins/prometheus.md
index a26787642..95d89f02f 100644
--- a/docs/en/latest/plugins/prometheus.md
+++ b/docs/en/latest/plugins/prometheus.md
@@ -197,7 +197,7 @@ The following labels are used to differentiate
`apisix_http_status` metrics.
| service | ID of the Service that the HTTP status originates from when
`prefer_name` is `false` (default), and name of the Service when `prefer_name`
to `true`. Default to the configured value of host on the Route if the matched
Route does not belong to any Service. |
| consumer | Name of the Consumer associated with a request. Default to an
empty string if no Consumer is associated with the request.
|
| node | IP address of the Upstream node. For AI Routes, this is the
LLM instance name reported by `ai-proxy` or `ai-proxy-multi`. |
-| request_type | Request category: `traditional_http`, `ai_chat`, or
`ai_stream`. |
+| request_type | Request category: `traditional_http`, `websocket`,
`ai_chat`, or `ai_stream`. A request answered with `101 Switching Protocols` is
`websocket`. |
| request_llm_model | Model name requested by the client.
|
| llm_model | Effective target model for the AI request. Uses the
model configured on the AI instance when present; otherwise uses the model
requested by the client. Empty for traditional HTTP traffic. |
| response_source | Response origin: `apisix` for responses generated by
APISIX, `nginx` for NGINX proxy errors, or `upstream` for responses received
from the Upstream. |
@@ -257,7 +257,7 @@ The following labels are used to differentiate
`apisix_bandwidth` metrics.
| service | ID of the Service that bandwidth corresponds to when
`prefer_name` is `false` (default), and name of the Service when `prefer_name`
to `true`. Default to the configured value of host on the Route if the matched
Route does not belong to any Service. |
| consumer | Name of the Consumer associated with a request. Default to an
empty string if no Consumer is associated with the request.
|
| node | IP address of the Upstream node. For AI Routes, this is the LLM
instance name reported by `ai-proxy` or `ai-proxy-multi`. |
-| request_type | Request category: `traditional_http`, `ai_chat`, or
`ai_stream`. |
+| request_type | Request category: `traditional_http`, `websocket`,
`ai_chat`, or `ai_stream`. A request answered with `101 Switching Protocols` is
`websocket`. |
| request_llm_model | Model name requested by the client.
|
| llm_model | Effective target model for the AI request. Uses the
model configured on the AI instance when present; otherwise uses the model
requested by the client. Empty for traditional HTTP traffic. |
@@ -386,7 +386,7 @@ The following labels are used to differentiate
`apisix_http_latency` metrics.
| service | ID of the Service that latencies correspond to when
`prefer_name` is `false` (default), and name of the Service when `prefer_name`
to `true`. Default to the configured value of host on the Route if the matched
Route does not belong to any Service. |
| consumer | Name of the Consumer associated with latencies. Default to an
empty string if no Consumer is associated with the request.
|
| node | IP address of the Upstream node associated with latencies. For
AI Routes, this is the LLM instance name reported by `ai-proxy` or
`ai-proxy-multi`. |
-| request_type | Request category: `traditional_http`, `ai_chat`, or
`ai_stream`. |
+| request_type | Request category: `traditional_http`, `websocket`,
`ai_chat`, or `ai_stream`. A request answered with `101 Switching Protocols` is
`websocket`. |
| request_llm_model | Model name requested by the client.
|
| llm_model | Effective target model for the AI request. Uses the
model configured on the AI instance when present; otherwise uses the model
requested by the client. Empty for traditional HTTP traffic. |
diff --git a/docs/zh/latest/plugins/prometheus.md
b/docs/zh/latest/plugins/prometheus.md
index 48778dab4..86081f9ff 100644
--- a/docs/zh/latest/plugins/prometheus.md
+++ b/docs/zh/latest/plugins/prometheus.md
@@ -197,7 +197,7 @@ Prometheus 中有不同类型的指标。要了解它们之间的区别，请参
| service | HTTP 状态来源的服务 ID，当 `prefer_name` 为 `false`（默认）时，使用服务 ID，当
`prefer_name` 为 `true` 时，使用服务名称。如果匹配的路由不属于任何服务，则默认为路由上配置的主机值。 |
| consumer | 与请求关联的消费者名称。如果请求没有与之关联的消费者，则默认为空字符串。
|
| node | 上游节点的 IP 地址。对于 AI 路由，该值为 `ai-proxy` 或 `ai-proxy-multi` 上报的 LLM
实例名称。 |
-| request_type | 请求类别：`traditional_http`、`ai_chat` 或 `ai_stream`。
|
+| request_type | 请求类别：`traditional_http`、`websocket`、`ai_chat` 或
`ai_stream`。以 `101 Switching Protocols` 响应的请求为 `websocket`。 |
| request_llm_model | 客户端请求的模型名称。
|
| llm_model | AI 请求实际使用的目标模型。优先使用 AI 实例中配置的模型，否则使用客户端请求的模型；传统 HTTP
流量中为空字符串。 |
| response_source | 响应来源：`apisix` 表示由 APISIX 生成，`nginx` 表示 NGINX
代理错误，`upstream` 表示来自上游服务的响应。 |
@@ -245,7 +245,7 @@ UDP 没有关闭、FIN 或重置信号，因此只会出现其中一部分状态
| service | 请求对应的服务 ID，当 `prefer_name` 为 `false`（默认）时，使用服务 ID，当 `prefer_name`
为 `true` 时，使用服务名称。如果匹配的路由不属于任何服务，则默认为路由上配置的主机值。 |
| consumer | 与请求关联的消费者名称。如果请求没有与之关联的消费者，则默认为空字符串。
|
| node | 上游节点的 IP 地址。对于 AI 路由，该值为 `ai-proxy` 或 `ai-proxy-multi` 上报的 LLM
实例名称。 |
-| request_type | 请求类别：`traditional_http`、`ai_chat` 或 `ai_stream`。
|
+| request_type | 请求类别：`traditional_http`、`websocket`、`ai_chat` 或
`ai_stream`。以 `101 Switching Protocols` 响应的请求为 `websocket`。 |
| request_llm_model | 客户端请求的模型名称。
|
| llm_model | AI 请求实际使用的目标模型。优先使用 AI 实例中配置的模型，否则使用客户端请求的模型；传统 HTTP
流量中为空字符串。 |
@@ -374,7 +374,7 @@ APISIX 在开始 LLM 上游尝试时递增该 gauge，并在请求的日志阶
| service | 延迟对应的服务 ID，当 `prefer_name` 为 `false`（默认）时，使用服务 ID，当 `prefer_name`
为 `true` 时，使用服务名称。如果匹配的路由不属于任何服务，则默认为路由上配置的主机值。 |
| consumer | 与延迟关联的消费者名称。如果请求没有与之关联的消费者，则默认为空字符串。
|
| node | 与延迟关联的上游节点 IP 地址。对于 AI 路由，该值为 `ai-proxy` 或 `ai-proxy-multi` 上报的 LLM
实例名称。 |
-| request_type | 请求类别：`traditional_http`、`ai_chat` 或 `ai_stream`。
|
+| request_type | 请求类别：`traditional_http`、`websocket`、`ai_chat` 或
`ai_stream`。以 `101 Switching Protocols` 响应的请求为 `websocket`。 |
| request_llm_model | 客户端请求的模型名称。
|
| llm_model | AI 请求实际使用的目标模型。优先使用 AI 实例中配置的模型，否则使用客户端请求的模型；传统 HTTP
流量中为空字符串。 |
diff --git a/t/plugin/prometheus-websocket.t b/t/plugin/prometheus-websocket.t
new file mode 100644
index 000000000..b82d4fe98
--- /dev/null
+++ b/t/plugin/prometheus-websocket.t
@@ -0,0 +1,183 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements. See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+BEGIN {
+ if ($ENV{TEST_NGINX_CHECK_LEAK}) {
+ $SkipReason = "unavailable for the hup tests";
+
+ } else {
+ $ENV{TEST_NGINX_USE_HUP} = 1;
+ undef $ENV{TEST_NGINX_USE_STAP};
+ }
+}
+
+use t::APISIX 'no_plan';
+
+repeat_each(1);
+no_long_string();
+no_shuffle();
+no_root_location();
+
+add_block_preprocessor(sub {
+ my ($block) = @_;
+
+ if (!defined $block->request) {
+ $block->set_value("request", "GET /t");
+ }
+
+ if (!defined $block->yaml_config) {
+ $block->set_value("yaml_config", <<'EOC');
+plugin_attr:
+ prometheus:
+ refresh_interval: 0.1
+EOC
+ }
+});
+
+run_tests;
+
+__DATA__
+
+=== TEST 1: set up the metrics route and a websocket-enabled route that logs
$request_type
+--- config
+ location /t {
+ content_by_lua_block {
+ local t = require("lib.test_admin").test
+
+ local code = t('/apisix/admin/routes/metrics',
+ ngx.HTTP_PUT,
+ [[{
+ "plugins": {"public-api": {}},
+ "uri": "/apisix/prometheus/metrics"
+ }]])
+ if code >= 300 then
+ ngx.status = code
+ ngx.say("failed to create metrics route")
+ return
+ end
+
+ local code, body = t('/apisix/admin/routes/ws',
+ ngx.HTTP_PUT,
+ [[{
+ "plugins": {
+ "prometheus": {},
+ "serverless-post-function": {
+ "phase": "log",
+ "functions": [
+ "return function(conf, ctx) ngx.log(ngx.WARN,
\"request_type=\", ngx.var.request_type) end"
+ ]
+ }
+ },
+ "enable_websocket": true,
+ "upstream": {
+ "nodes": {"127.0.0.1:1980": 1},
+ "type": "roundrobin"
+ },
+ "uri": "/websocket_handshake"
+ }]])
+ if code >= 300 then
+ ngx.status = code
+ ngx.say(body)
+ return
+ end
+
+ ngx.sleep(0.5)
+ ngx.say(body)
+ }
+ }
+--- response_body
+passed
+
+
+
+=== TEST 2: a completed upgrade is logged as request_type=websocket
+--- config
+ location /t {
+ content_by_lua_block {
+ local client = require("resty.websocket.client")
+ local wb = client:new()
+ local ok, err =
wb:connect("ws://127.0.0.1:1984/websocket_handshake")
+ if not ok then
+ ngx.say("failed to connect: ", err)
+ return
+ end
+
+ local data, typ, err = wb:recv_frame()
+ if not data then
+ ngx.say("failed to receive frame: ", err)
+ return
+ end
+
+ wb:close()
+ ngx.say("received: ", data, " (", typ, ")")
+ }
+ }
+--- response_body
+received: hello (text)
+--- error_log
+request_type=websocket
+
+
+
+=== TEST 3: the session's status is labelled request_type=websocket
+--- request
+GET /apisix/prometheus/metrics
+--- response_body eval
+qr/apisix_http_status\{code="101",route="ws",[^}]*request_type="websocket"[^}]*\}
1\n/
+
+
+
+=== TEST 4: the session's latency is labelled request_type=websocket
+--- request
+GET /apisix/prometheus/metrics
+--- response_body eval
+qr/apisix_http_latency_count\{type="request",route="ws",[^}]*request_type="websocket"[^}]*\}
1\n/
+
+
+
+=== TEST 5: the session's bandwidth is labelled request_type=websocket
+--- request
+GET /apisix/prometheus/metrics
+--- response_body eval
+qr/apisix_bandwidth\{type="ingress",route="ws",[^}]*request_type="websocket"[^}]*\}
\d+\n/
+
+
+
+=== TEST 6: a refused handshake on the same route stays
request_type=traditional_http
+--- request
+GET /websocket_handshake
+--- error_code: 400
+--- error_log
+failed to new websocket: bad "upgrade" request header
+request_type=traditional_http
+
+
+
+=== TEST 7: the refused request lands in its own series
+--- request
+GET /apisix/prometheus/metrics
+--- response_body eval
+qr/apisix_http_status\{code="400",route="ws",[^}]*request_type="traditional_http"[^}]*\}
1\n/
+
+
+
+=== TEST 8: the websocket series is untouched by the refused request
+--- request
+GET /apisix/prometheus/metrics
+--- response_body_like eval
+qr/apisix_http_latency_count\{type="request",route="ws",[^}]*request_type="websocket"[^}]*\}
1\n/
+--- response_body_unlike eval
+qr/apisix_http_latency_count\{type="request",route="ws",[^}]*request_type="websocket"[^}]*\}
2\n/
