ticdc 4.0.9
今天中午突然任务延迟
重启cdc 后目前状态如下:
Starting component `ctl`: /home/tidb/.tiup/components/ctl/v4.0.9/ctl cdc changefeed list --pd=http://172.29.238.113:2379
[
{
"id": "kafka-task",
"summary": {
"state": "normal",
"tso": 425906039086907405,
"checkpoint": "2021-06-26 18:16:42.718",
"error": null
}
}
]
--- 共三个节点
[
{
"id": "97fb515b-97a0-4ded-ae4a-8a56ea366f95",
"is-owner": false,
"address": "172.29.238.137:8300"
},
{
"id": "b5ece9bd-f09e-4f85-97e1-6dec2c079524",
"is-owner": false,
"address": "172.29.238.131:8300"
},
{
"id": "df39138e-c755-4d2c-89df-3133fac22f3a",
"is-owner": true,
"address": "172.29.238.81:8300"
}
]
日志输出如下:
其中一个节点如下:
另外两个节点日志如下:
请问现在是正常的状态么 ?是否在追赶数据?
今天中午出现异常的时候日志大致如下
[2021/06/26 11:13:40.023 +08:00] [WARN] [client.go:1325] ["region not receiving resolved event from tikv or resolved ts is not pushing for too long time, try to resolve lock
"] [regionID=693438] [span="[7480000000000001ff6a5f72680000002bffffb1780000000000fa, 7480000000000001ff6a5f72680000002cff07a09b0000000000fa)"] [duration=21.2s] [resolvedTs=4
25899379606945820]
请问应该怎么处理比较合适,已经同步到下游kafak 的数据是否要重新全量同步一下。
我们这个任务已经跑了好久了