TiDB Operator

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: tidb operator

| username: liyuntang

[TiDB Usage Environment] Testing
[TiDB Version] 5.4.1
[Reproduction Path]

[Encountered Issues: Issue Phenomenon and Impact]
[Resource Configuration]
[TiDB Operator Version]:
Currently version 1.3.3, planning to upgrade to version 1.4.6
[K8s Version]:
1.19
[Attachments: Screenshots/Logs/Monitoring]

Questions:

  1. Will the change 4683 affect the scaling functionality of PVC for tidb, pd, ticdc, tiflash, tikv? If so, what specific impacts will it have? If not, which component’s PVC is being referred to here?

  2. What impact will the change 4494 have on tasks during the task reconstruction process? Will there be synchronization failures, transaction losses, etc.?

| username: tidb狂热爱好者 | Original post link

As I understand it, your synchronization tasks need to be reset because CSC is new.

| username: liyuntang | Original post link

Does this mean we need to manually rebuild all tasks?

| username: Billmay表妹 | Original post link

According to the literal understanding: the storage modification function of all components is disabled by default. If you want to scale up or down the PVC of any component, you can manually enable this function.

| username: liyuntang | Original post link

I have verified this, it should not be the case.

| username: yiduoyunQ | Original post link

  1. Will change 4683 affect the PVC scaling functionality of TiDB, PD, TiCDC, TiFlash, and TiKV? If so, what specific impacts will it have? If not, which component’s PVC is being referred to here?

No, it won’t. This refers to the size and performance parameters of EBS, including IOPS/throughput, etc. For specifics, you can refer to the EBS driver documentation: aws-ebs-csi-driver/docs/parameters.md at master · kubernetes-sigs/aws-ebs-csi-driver · GitHub

  1. What impact will change 4494 have on tasks during the task reconstruction process? Will there be synchronization failures, transaction losses, etc.?
  • What impact will it have on tasks?
    It’s hard to say. Normally, it should be online, but it’s best to test it.
  • Will there be synchronization failures, transaction losses, etc.?
    It’s hard to say. Normally, there shouldn’t be, but it’s best to test it.
| username: redgame | Original post link

No impact. Avoid extreme scenarios.

| username: liyuntang | Original post link

I tested it, no impact.

  1. Observe the status of ticdc synchronization tasks between 19:17:20 — 19:17:30

>>>>>>>>>>>>>>>>> Monday, July 24, 2023 19:17:19 CST
[
    {
        "id": "kafka222",
        "state": "normal",
        "checkpoint_tso": 443075117123633153,
        "checkpoint_time": "2023-07-24 19:17:17.758",
        "error": null
    },
    {
        "id": "kafka281",
        "state": "normal",
        "checkpoint_tso": 443075117123633153,
        "checkpoint_time": "2023-07-24 19:17:17.758",
        "error": null
    },
    {
        "id": "mysql",
        "state": "normal",
        "checkpoint_tso": 443075117123633153,
        "checkpoint_time": "2023-07-24 19:17:17.758",
        "error": null
    }
]

>>>>>>>>>>>>>>>>> Monday, July 24, 2023 19:17:21 CST
[
    {
        "id": "kafka281",
        "state": "normal",
        "checkpoint_tso": 443075117647921153,
        "checkpoint_time": "2023-07-24 19:17:19.758",
        "error": null
    },
    {
        "id": "mysql",
        "state": "normal",
        "checkpoint_tso": 443075117386039297,
        "checkpoint_time": "2023-07-24 19:17:18.759",
        "error": null
    },
    {
        "id": "kafka222",
        "state": "normal",
        "checkpoint_tso": 443075117647921153,
        "checkpoint_time": "2023-07-24 19:17:19.758",
        "error": null
    }
]

>>>>>>>>>>>>>>>>> Monday, July 24, 2023 19:17:22 CST
[
    {
        "id": "mysql",
        "state": "normal",
        "checkpoint_tso": 443075117910327297,
        "checkpoint_time": "2023-07-24 19:17:20.759",
        "error": null
    },
    {
        "id": "kafka222",
        "state": "normal",
        "checkpoint_tso": 443075117910327297,
        "checkpoint_time": "2023-07-24 19:17:20.759",
        "error": null
    },
    {
        "id": "kafka281",
        "state": "normal",
        "checkpoint_tso": 443075117910327297,
        "checkpoint_time": "2023-07-24 19:17:20.759",
        "error": null
    }
]

>>>>>>>>>>>>>>>>> Monday, July 24, 2023 19:17:24 CST
[
    {
        "id": "kafka281",
        "state": "normal",
        "checkpoint_tso": 443075118434615297,
        "checkpoint_time": "2023-07-24 19:17:22.759",
        "error": null
    },
    {
        "id": "mysql",
        "state": "normal",
        "checkpoint_tso": 443075118434615297,
        "checkpoint_time": "2023-07-24 19:17:22.759",
        "error": null
    },
    {
        "id": "kafka222",
        "state": "normal",
        "checkpoint_tso": 443075118434615297,
        "checkpoint_time": "2023-07-24 19:17:22.759",
        "error": null
    }
]

>>>>>>>>>>>>>>>>> Monday, July 24, 2023 19:17:26 CST
[
    {
        "id": "kafka222",
        "state": "normal",
        "checkpoint_tso": 443075118696759297,
        "checkpoint_time": "2023-07-24 19:17:23.759",
        "error": null
    },
    {
        "id": "kafka281",
        "state": "normal",
        "checkpoint_tso": 443075118696759297,
        "checkpoint_time": "2023-07-24 19:17:23.759",
        "error": null
    },
    {
        "id": "mysql",
        "state": "normal",
        "checkpoint_tso": 443075118696759297,
        "checkpoint_time": "2023-07-24 19:17:23.759",
        "error": null
    }
]

>>>>>>>>>>>>>>>>> Monday, July 24, 2023 19:17:27 CST
[
    {
        "id": "mysql",
        "state": "normal",
        "checkpoint_tso": 443075118958641153,
        "checkpoint_time": "2023-07-24 19:17:24.758",
        "error": null
    },
    {
        "id": "kafka222",
        "state": "normal",
        "checkpoint_tso": 443075119221047297,
        "checkpoint_time": "2023-07-24 19:17:25.759",
        "error": null
    },
    {
        "id": "kafka281",
        "state": "normal",
        "checkpoint_tso": 443075119221047297,
        "checkpoint_time": "2023-07-24 19:17:25.759",
        "error": null
    }
]

>>>>>>>>>>>>>>>>> Monday, July 24, 2023 19:17:28 CST
[
    {
        "id": "kafka281",
        "state": "normal",
        "checkpoint_tso": 443075119483191297,
        "checkpoint_time": "2023-07-24 19:17:26.759",
        "error": null
    },
    {
        "id": "mysql",
        "state": "normal",
        "checkpoint_tso": 443075119483191297,
        "checkpoint_time": "2023-07-24 19:17:26.759",
        "error": null
    },
    {
        "id": "kafka222",
        "state": "normal",
        "checkpoint_tso": 443075119483191297,
        "checkpoint_time": "2023-07-24 19:17:26.759",
        "error": null
    }
]

>>>>>>>>>>>>>>>>> Monday, July 24, 2023 19:17:30 CST
[
    {
        "id": "kafka222",
        "state": "normal",
        "checkpoint_tso": 443075120007479297,
        "checkpoint_time": "2023-07-24 19:17:28.759",
        "error": null
    },
    {
        "id": "kafka281",
        "state": "normal",
        "checkpoint_tso": 443075120007479297,
        "checkpoint_time": "2023-07-24 19:17:28.759",
        "error": null
    },
    {
        "id": "mysql",
        "state": "normal",
        "checkpoint_tso": 443075120007479297,
        "checkpoint_time": "2023-07-24 19:17:28.759",
        "error": null
    }
]

| username: liyuntang | Original post link

Well, there are no extreme scenarios; the business SQLs are all CRUD operations.

| username: system | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.