切换时序引擎¶
简介¶
本文将介绍观测云切换时序引擎,此操作适用于 InfluxDB 切换 GuanceDB。
前提条件¶
- 已部署 GuanceDB,未部署可参考 GuanceDB 安装
部署信息¶
名称 | 内容 |
---|---|
guance-insert 默认地址 | guancedb-cluster-guance-select.middleware |
guance-insert默认端口 | 8480 |
guance-select 默认地址 | guancedb-cluster-guance-select.middleware |
guance-select 默认端口 | 8481 |
切换步骤¶
备份表¶
CREATE TABLE `main_workspace_bak` LIKE `main_workspace`;
INSERT INTO `main_workspace_bak` SELECT * FROM `main_workspace`;
CREATE TABLE `main_influx_db_bak` LIKE `main_influx_db`;
INSERT INTO `main_influx_db_bak` SELECT * FROM `main_influx_db`;
插入 GuanceDB 信息¶
- 登录观测云 df_core 数据库,执行以下语句:
INSERT INTO `main_influx_instance` (`uuid`, `host`, `authorization`, `configJSON`,`dbcount`, `user`, `pwd`, `dbType`, `priority`, `status`, `creator`, `updator`, `createAt`, `deleteAt`, `updateAt`)
VALUES
('276f8404252347d18e469f65f5a1abdd', 'http://0.0.0.0', '{\"admin\": {\"password\": \"*\", \"username\": \"*\"}}','{"read": {"host": "http://guancedb-cluster-guance-select.middleware:8481"}, "write": {"host": "http://guancedb-cluster-guance-insert.middleware:8480"}}', 0, '', '', 'guancedb', 50, 0, '', '', UNIX_TIMESTAMP(), -1, UNIX_TIMESTAMP());
验证:
select count(1) from main_influx_instance where uuid='276f8404252347d18e469f65f5a1abdd';
+----------+
| count(1) |
+----------+
| 1 |
+----------+
1 row in set (0.00 sec)
查询结果大等于1,为执行成功
GuanceDB 替换 InfluxDB 信息¶
- 更改 InfluxDB 状态
验证:
select count(1) from main_influx_instance where status=3 and dbType='influxdb';
+----------+
| count(1) |
+----------+
| 1 |
+----------+
1 row in set (0.00 sec)
查询结果大于等于1,为执行成功
- 批量更新所有工作空间对应的 时序数据库信息
update main_influx_db set influxInstanceUUID=(select uuid from main_influx_instance where status=0 and dbType='guancedb' order by priority desc limit 1) , dbType='guancedb' where status=0;
验证:
SELECT count(1)
FROM main_influx_db AS db
JOIN main_influx_instance AS instance ON db.influxInstanceUUID = instance.uuid
WHERE db.status = 0
AND db.dbType = 'guancedb';
+----------+
| count(1) |
+----------+
| 26 |
+----------+
1 row in set (0.01 sec)
查询结果大于等于1,为执行成功
- 批量更新
main_workspace
表中的datastore
中的metric
值
update main_workspace set datastore=JSON_SET(datastore, "$.metric", "guancedb") where status=0 and datastore ->> "$.metric" = "influxdb";
验证:
SELECT count(1)
FROM main_workspace
WHERE status = 0
AND datastore ->> "$.metric" = "guancedb";
+----------+
| count(1) |
+----------+
| 26 |
+----------+
1 row in set (0.00 sec)
查询结果大于等于1,为执行成功
- 更换 DB ID
UPDATE df_core.main_influx_db
SET df_core.main_influx_db.db =(
SELECT df_core.main_workspace.id
FROM df_core.main_workspace
WHERE df_core.main_workspace.dbUUID = df_core.main_influx_db.UUID
);
验证:
SELECT count(1)
FROM df_core.main_influx_db AS db
JOIN df_core.main_workspace AS ws ON ws.dbUUID = db.UUID;
+----------+
| count(1) |
+----------+
| 26 |
+----------+
1 row in set (0.00 sec)
查询结果大于等于1,为执行成功
重启 Kodo 所有服务¶
清空 Redis¶
两种方法二选一:
- 重启 Redis
- 执行清空缓存命令
# 删除所有的 "dbinfo:*" 缓存记录,执行如下批量删除命令
redis-cli -h <host> -p <端口> -n 0 -a <密码> keys dbinfo:* | xargs -r -t -n1 redis-cli -h <host> -p <端口> -n 0 -a <密码> del
# 删除所有的 "wkspInfo:*" 缓存记录,执行如下批量删除命令
redis-cli -h <host> -p <端口> -n 0 -a <密码> keys wkspInfo:* | xargs -r -t -n1 redis-cli -h <host> -p <端口> -n 0 -a <密码> del
# 删除所有的 "tkn_info:*" 缓存记录,执行如下批量命令
redis-cli -h <host> -p <端口> -n 0 -a <密码> keys tkn_info:* | xargs -r -t -n1 redis-cli -h <host> -p <端口> -n 0 -a <密码> del