未验证 提交 eec05d86 编写于 作者: K kezhenxu94 提交者: GitHub

Set default connection timeout of ElasticSearch to 3000 milliseconds (#8560)

上级 bdf4ccd3
......@@ -60,6 +60,7 @@ Release Notes.
* Add Python FastApi component ID(7014).
* Support all metrics from MAL engine in alarm core, including Prometheus, OC receiver, meter receiver.
* Allow updating non-metrics templates when structure changed.
* Set default connection timeout of ElasticSearch to 3000 milliseconds.
#### UI
......
......@@ -86,7 +86,7 @@ core|default|role|Option values: `Mixed/Receiver/Aggregator`. **Receiver** mode
| - | - | namespace | Prefix of indexes created and used by SkyWalking. | SW_NAMESPACE | - |
| - | - | clusterNodes | ElasticSearch cluster nodes for client connection.| SW_STORAGE_ES_CLUSTER_NODES |localhost|
| - | - | protocol | HTTP or HTTPs. | SW_STORAGE_ES_HTTP_PROTOCOL | HTTP|
| - | - | connectTimeout | Connect timeout of ElasticSearch client (in milliseconds). | SW_STORAGE_ES_CONNECT_TIMEOUT | 500|
| - | - | connectTimeout | Connect timeout of ElasticSearch client (in milliseconds). | SW_STORAGE_ES_CONNECT_TIMEOUT | 3000|
| - | - | socketTimeout | Socket timeout of ElasticSearch client (in milliseconds). | SW_STORAGE_ES_SOCKET_TIMEOUT | 30000|
| - | - | responseTimeout | Response timeout of ElasticSearch client (in milliseconds), `0` disables the timeout. | SW_STORAGE_ES_RESPONSE_TIMEOUT | 1500 |
| - | - | numHttpClientThread | The number of threads for the underlying HTTP client to perform socket I/O. If the value is <= 0, the number of available processors will be used. | SW_STORAGE_ES_NUM_HTTP_CLIENT_THREAD | 0 |
......@@ -114,7 +114,7 @@ core|default|role|Option values: `Mixed/Receiver/Aggregator`. **Receiver** mode
| - | - | driver | H2 JDBC driver. | SW_STORAGE_H2_DRIVER | org.h2.jdbcx.JdbcDataSource|
| - | - | url | H2 connection URL. Defaults to H2 memory mode. | SW_STORAGE_H2_URL | jdbc:h2:mem:skywalking-oap-db |
| - | - | user | Username of H2 database. | SW_STORAGE_H2_USER | sa |
| - | - | password | Password of H2 database. | - | - |
| - | - | password | Password of H2 database. | - | - |
| - | - | metadataQueryMaxSize | The maximum size of metadata per query. | SW_STORAGE_H2_QUERY_MAX_SIZE | 5000 |
| - | - | maxSizeOfArrayColumn | Some entities (e.g. trace segments) include the logic column with multiple values. In H2, we use multiple physical columns to host the values: e.g. change column_a with values [1,2,3,4,5] to `column_a_0 = 1, column_a_1 = 2, column_a_2 = 3 , column_a_3 = 4, column_a_4 = 5`. | SW_STORAGE_MAX_SIZE_OF_ARRAY_COLUMN | 20 |
| - | - | numOfSearchableValuesPerTag | In a trace segment, this includes multiple spans with multiple tags. Different spans may have the same tag key, e.g. multiple HTTP exit spans all have their own `http.method` tags. This configuration sets the limit on the maximum number of values for the same tag key. | SW_STORAGE_NUM_OF_SEARCHABLE_VALUES_PER_TAG | 2 |
......@@ -206,7 +206,7 @@ core|default|role|Option values: `Mixed/Receiver/Aggregator`. **Receiver** mode
| - | - | restContextPath| Web context path of RESTful services. | SW_RECEIVER_ZIPKIN_CONTEXT_PATH|/|
| prometheus-fetcher | default | Prometheus fetcher reads metrics from Prometheus endpoint, and transfer the metrics into SkyWalking native format for the MAL engine. | - | - |
| - | - | enabledRules | Enabled rules. | SW_PROMETHEUS_FETCHER_ENABLED_RULES | self |
| - | - | maxConvertWorker | The maximize meter convert worker. | SW_PROMETHEUS_FETCHER_NUM_CONVERT_WORKER | -1(by default, half the number of CPU core(s)) |
| - | - | maxConvertWorker | The maximize meter convert worker. | SW_PROMETHEUS_FETCHER_NUM_CONVERT_WORKER | -1(by default, half the number of CPU core(s)) |
| kafka-fetcher | default | Read SkyWalking's native metrics/logs/traces through Kafka server. | - | - |
| - | - | bootstrapServers | A list of host/port pairs to use for establishing the initial connection to the Kafka cluster. | SW_KAFKA_FETCHER_SERVERS | localhost:9092 |
| - | - | namespace | Namespace aims to isolate multi OAP cluster when using the same Kafka cluster. If you set a namespace for Kafka fetcher, OAP will add a prefix to topic name. You should also set namespace in `agent.config`. The property is named `plugin.kafka.namespace`. | SW_NAMESPACE | - |
......@@ -242,7 +242,7 @@ core|default|role|Option values: `Mixed/Receiver/Aggregator`. **Receiver** mode
| - | - | port | DCS server binding port. | SW_DCS_SERVER_PORT | 80 |
| - | - | clusterName | Cluster name when reading the latest configuration from DSC server. | SW_DCS_CLUSTER_NAME | SkyWalking|
| - | - | period | The period of reading data from DSC server by the OAP (in seconds). | SW_DCS_PERIOD | 20 |
| - | apollo| apolloMeta| `apollo.meta` in Apollo. | SW_CONFIG_APOLLO | http://localhost:8080 |
| - | apollo| apolloMeta| `apollo.meta` in Apollo. | SW_CONFIG_APOLLO | http://localhost:8080 |
| - | - | apolloCluster | `apollo.cluster` in Apollo. | SW_CONFIG_APOLLO_CLUSTER | default|
| - | - | apolloEnv | `env` in Apollo. | SW_CONFIG_APOLLO_ENV | - |
| - | - | appId | `app.id` in Apollo. | SW_CONFIG_APOLLO_APP_ID | skywalking |
......@@ -252,7 +252,7 @@ core|default|role|Option values: `Mixed/Receiver/Aggregator`. **Receiver** mode
| - | - | baseSleepTimeMs|The period of Zookeeper client between two retries (in milliseconds). |SW_CONFIG_ZK_BASE_SLEEP_TIME_MS|1000|
| - | - | maxRetries| The maximum retry time. |SW_CONFIG_ZK_MAX_RETRIES|3|
| - | - | period | The period of data sync (in seconds). | SW_CONFIG_ZK_PERIOD | 60 |
| - | etcd| endpoints | Hosts and ports for etcd cluster (separated by commas if multiple). | SW_CONFIG_ETCD_ENDPOINTS | http://localhost:2379 |
| - | etcd| endpoints | Hosts and ports for etcd cluster (separated by commas if multiple). | SW_CONFIG_ETCD_ENDPOINTS | http://localhost:2379 |
| - | - | namespace | Namespace for SkyWalking cluster. |SW_CONFIG_ETCD_NAMESPACE | /skywalking |
| - | - | authentication | Indicates whether there is authentication. | SW_CONFIG_ETCD_AUTHENTICATION | false |
| - | - | user | Etcd auth username. | SW_CONFIG_ETCD_USER | |
......
......@@ -133,7 +133,7 @@ storage:
namespace: ${SW_NAMESPACE:""}
clusterNodes: ${SW_STORAGE_ES_CLUSTER_NODES:localhost:9200}
protocol: ${SW_STORAGE_ES_HTTP_PROTOCOL:"http"}
connectTimeout: ${SW_STORAGE_ES_CONNECT_TIMEOUT:500}
connectTimeout: ${SW_STORAGE_ES_CONNECT_TIMEOUT:3000}
socketTimeout: ${SW_STORAGE_ES_SOCKET_TIMEOUT:30000}
responseTimeout: ${SW_STORAGE_ES_RESPONSE_TIMEOUT:15000}
numHttpClientThread: ${SW_STORAGE_ES_NUM_HTTP_CLIENT_THREAD:0}
......
......@@ -34,7 +34,7 @@ public class StorageModuleElasticsearchConfig extends ModuleConfig {
*
* @since 8.7.0
*/
private int connectTimeout = 500;
private int connectTimeout = 3000;
/**
* Socket timeout of ElasticSearch client.
*
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册