tidb数据写入到分区表因为时区问题落错分区

[Billdi表弟]

【 TiDB 使用环境】测试
【 TiDB 版本】v5.4.0
【复现路径】建表+写入数据+查询
注意我的时区 SYSTEM是上海，time_zone环境变量时utc，数据库里面存入的时间戳也是utc

CREATE TABLE `history_wti_alarm` (
  `id` bigint not null,
  `alarm_id` varchar(64) NOT NULL  COMMENT 'unique alarm id',
  `start_time` timestamp(3) NOT NULL,
  `end_time` timestamp(3) NULL DEFAULT NULL COMMENT 'Last sample timestamp',
  PRIMARY KEY (`alarm_id`,`start_time`),
  UNIQUE KEY `uniq_alarm_id` (`alarm_id`,`start_time`),
  KEY `idx_start_time` (`start_time`)
) PARTITION BY RANGE(floor(UNIX_TIMESTAMP(start_time))) (
    PARTITION p1 VALUES LESS THAN (UNIX_TIMESTAMP('2022-01-01 00:00:00')),
    PARTITION p2 VALUES LESS THAN (UNIX_TIMESTAMP('2022-01-10 00:00:00')),
    PARTITION p3 VALUES LESS THAN (UNIX_TIMESTAMP('2022-01-20 00:00:00')),
    PARTITION p4 VALUES LESS THAN (UNIX_TIMESTAMP('2022-01-31 00:00:00')),
    PARTITION p5 VALUES LESS THAN (UNIX_TIMESTAMP('2022-02-01 00:00:00')),
    PARTITION p6 VALUES LESS THAN (UNIX_TIMESTAMP('2022-02-10 00:00:00')),
    PARTITION p7 VALUES LESS THAN (UNIX_TIMESTAMP('2022-02-20 00:00:00')),
    PARTITION p8 VALUES LESS THAN (UNIX_TIMESTAMP('2022-02-28 00:00:00')),
    PARTITION p9 VALUES LESS THAN (UNIX_TIMESTAMP('2022-03-01 00:00:00')),
    PARTITION p10 VALUES LESS THAN (UNIX_TIMESTAMP('2022-03-10 00:00:00')),
    PARTITION p11 VALUES LESS THAN (UNIX_TIMESTAMP('2022-03-20 00:00:00')),
    PARTITION p12 VALUES LESS THAN (UNIX_TIMESTAMP('2022-03-31 00:00:00')),
    PARTITION p13 VALUES LESS THAN (UNIX_TIMESTAMP('2022-04-01 00:00:00')),
    PARTITION p14 VALUES LESS THAN (UNIX_TIMESTAMP('2022-04-10 00:00:00')),
    PARTITION p15 VALUES LESS THAN (UNIX_TIMESTAMP('2022-04-20 00:00:00')),
    PARTITION p16 VALUES LESS THAN (UNIX_TIMESTAMP('2022-04-30 00:00:00')),
    PARTITION p17 VALUES LESS THAN (UNIX_TIMESTAMP('2022-05-01 00:00:00')),
    PARTITION p18 VALUES LESS THAN (UNIX_TIMESTAMP('2022-05-10 00:00:00')),
    PARTITION p19 VALUES LESS THAN (UNIX_TIMESTAMP('2022-05-20 00:00:00')),
    PARTITION p20 VALUES LESS THAN (UNIX_TIMESTAMP('2022-05-31 00:00:00')),
    PARTITION p21 VALUES LESS THAN (UNIX_TIMESTAMP('2022-06-01 00:00:00')),
    PARTITION p22 VALUES LESS THAN (UNIX_TIMESTAMP('2022-06-10 00:00:00')),
    PARTITION p23 VALUES LESS THAN (UNIX_TIMESTAMP('2022-06-20 00:00:00')),
    PARTITION p24 VALUES LESS THAN (UNIX_TIMESTAMP('2022-06-30 00:00:00')),
    PARTITION p25 VALUES LESS THAN (UNIX_TIMESTAMP('2022-07-01 00:00:00')),
    PARTITION p26 VALUES LESS THAN (UNIX_TIMESTAMP('2022-07-10 00:00:00')),
    PARTITION p27 VALUES LESS THAN (UNIX_TIMESTAMP('2022-07-20 00:00:00')),
    PARTITION p28 VALUES LESS THAN (UNIX_TIMESTAMP('2022-07-31 00:00:00')),
    PARTITION p29 VALUES LESS THAN (UNIX_TIMESTAMP('2022-08-01 00:00:00')),
    PARTITION p30 VALUES LESS THAN (UNIX_TIMESTAMP('2022-08-10 00:00:00')),
    PARTITION p31 VALUES LESS THAN (UNIX_TIMESTAMP('2022-08-20 00:00:00')),
    PARTITION p32 VALUES LESS THAN (UNIX_TIMESTAMP('2022-08-31 00:00:00')),
    PARTITION p33 VALUES LESS THAN (UNIX_TIMESTAMP('2022-09-01 00:00:00')),
    PARTITION p34 VALUES LESS THAN (UNIX_TIMESTAMP('2022-09-10 00:00:00')),
    PARTITION p35 VALUES LESS THAN (UNIX_TIMESTAMP('2022-09-20 00:00:00')),
    PARTITION p36 VALUES LESS THAN (UNIX_TIMESTAMP('2022-09-30 00:00:00')),
    PARTITION p37 VALUES LESS THAN (UNIX_TIMESTAMP('2022-10-01 00:00:00')),
    PARTITION p38 VALUES LESS THAN (UNIX_TIMESTAMP('2022-10-10 00:00:00')),
    PARTITION p39 VALUES LESS THAN (UNIX_TIMESTAMP('2022-10-20 00:00:00')),
    PARTITION p40 VALUES LESS THAN (UNIX_TIMESTAMP('2022-10-31 00:00:00')),
    PARTITION p41 VALUES LESS THAN (UNIX_TIMESTAMP('2022-11-01 00:00:00')),  
    PARTITION p42 VALUES LESS THAN (UNIX_TIMESTAMP('2022-11-10 00:00:00')),
    PARTITION p43 VALUES LESS THAN (UNIX_TIMESTAMP('2022-11-20 00:00:00')),
    PARTITION p44 VALUES LESS THAN (UNIX_TIMESTAMP('2022-11-30 00:00:00')),
    PARTITION p45 VALUES LESS THAN (UNIX_TIMESTAMP('2022-12-01 00:00:00')),
    PARTITION p46 VALUES LESS THAN (UNIX_TIMESTAMP('2022-12-10 00:00:00')),
    PARTITION p47 VALUES LESS THAN (UNIX_TIMESTAMP('2022-12-20 00:00:00')),
    PARTITION p48 VALUES LESS THAN (UNIX_TIMESTAMP('2022-12-31 00:00:00')),
     PARTITION p49 VALUES LESS THAN (UNIX_TIMESTAMP('2023-01-01 00:00:00')),
     PARTITION p50 VALUES LESS THAN (UNIX_TIMESTAMP('2023-01-10 00:00:00')),
     PARTITION p51 VALUES LESS THAN (UNIX_TIMESTAMP('2023-01-20 00:00:00')),
     PARTITION p52 VALUES LESS THAN (UNIX_TIMESTAMP('2023-01-31 00:00:00')),
     PARTITION p53 VALUES LESS THAN (UNIX_TIMESTAMP('2023-02-01 00:00:00')),
     PARTITION p54 VALUES LESS THAN (UNIX_TIMESTAMP('2023-02-10 00:00:00')),
     PARTITION p55 VALUES LESS THAN (UNIX_TIMESTAMP('2023-02-20 00:00:00')),
     PARTITION p56 VALUES LESS THAN (UNIX_TIMESTAMP('2023-02-28 00:00:00')),
     PARTITION p57 VALUES LESS THAN (UNIX_TIMESTAMP('2023-03-01 00:00:00')),
     PARTITION p58 VALUES LESS THAN (UNIX_TIMESTAMP('2023-03-10 00:00:00')),
     PARTITION p59 VALUES LESS THAN (UNIX_TIMESTAMP('2023-03-20 00:00:00')),
     PARTITION p60 VALUES LESS THAN (UNIX_TIMESTAMP('2023-03-31 00:00:00')),
     PARTITION p61 VALUES LESS THAN (UNIX_TIMESTAMP('2023-04-01 00:00:00')),
     PARTITION p62 VALUES LESS THAN (UNIX_TIMESTAMP('2023-04-10 00:00:00')),
     PARTITION p63 VALUES LESS THAN (UNIX_TIMESTAMP('2023-04-20 00:00:00')),
     PARTITION p64 VALUES LESS THAN (UNIX_TIMESTAMP('2023-04-30 00:00:00')),
     PARTITION p65 VALUES LESS THAN (UNIX_TIMESTAMP('2023-05-01 00:00:00')),
     PARTITION p66 VALUES LESS THAN (UNIX_TIMESTAMP('2023-05-10 00:00:00')),
     PARTITION p67 VALUES LESS THAN (UNIX_TIMESTAMP('2023-05-20 00:00:00')),
     PARTITION p68 VALUES LESS THAN (UNIX_TIMESTAMP('2023-05-31 00:00:00')),
     PARTITION p69 VALUES LESS THAN (UNIX_TIMESTAMP('2023-06-01 00:00:00')),
     PARTITION p70 VALUES LESS THAN (UNIX_TIMESTAMP('2023-06-10 00:00:00')),
     PARTITION p71 VALUES LESS THAN (UNIX_TIMESTAMP('2023-06-20 00:00:00')),
     PARTITION p72 VALUES LESS THAN (UNIX_TIMESTAMP('2023-06-30 00:00:00')),
     PARTITION p73 VALUES LESS THAN (UNIX_TIMESTAMP('2023-07-01 00:00:00')),
     PARTITION p74 VALUES LESS THAN (UNIX_TIMESTAMP('2023-07-10 00:00:00')),
     PARTITION p75 VALUES LESS THAN (UNIX_TIMESTAMP('2023-07-20 00:00:00')),
     PARTITION p76 VALUES LESS THAN (UNIX_TIMESTAMP('2023-07-31 00:00:00')),
     PARTITION p77 VALUES LESS THAN (UNIX_TIMESTAMP('2023-08-01 00:00:00')),
     PARTITION p78 VALUES LESS THAN (UNIX_TIMESTAMP('2023-08-10 00:00:00')),
     PARTITION p79 VALUES LESS THAN (UNIX_TIMESTAMP('2023-08-20 00:00:00')),
     PARTITION p80 VALUES LESS THAN (UNIX_TIMESTAMP('2023-08-31 00:00:00')),
     PARTITION p81 VALUES LESS THAN (UNIX_TIMESTAMP('2023-09-01 00:00:00')),
     PARTITION p82 VALUES LESS THAN (UNIX_TIMESTAMP('2023-09-10 00:00:00')),
     PARTITION p83 VALUES LESS THAN (UNIX_TIMESTAMP('2023-09-20 00:00:00')),
     PARTITION p84 VALUES LESS THAN (UNIX_TIMESTAMP('2023-09-30 00:00:00')),
     PARTITION p85 VALUES LESS THAN (UNIX_TIMESTAMP('2023-10-01 00:00:00')),
     PARTITION p86 VALUES LESS THAN (UNIX_TIMESTAMP('2023-10-10 00:00:00')),
     PARTITION p87 VALUES LESS THAN (UNIX_TIMESTAMP('2023-10-20 00:00:00')),
     PARTITION p88 VALUES LESS THAN (UNIX_TIMESTAMP('2023-10-31 00:00:00')),
     PARTITION p90 VALUES LESS THAN (UNIX_TIMESTAMP('2023-11-01 00:00:00')),
     PARTITION p91 VALUES LESS THAN (UNIX_TIMESTAMP('2023-11-10 00:00:00')),
     PARTITION p92 VALUES LESS THAN (UNIX_TIMESTAMP('2023-11-20 00:00:00')),
     PARTITION p93 VALUES LESS THAN (UNIX_TIMESTAMP('2023-11-30 00:00:00')),
     PARTITION p94 VALUES LESS THAN (UNIX_TIMESTAMP('2023-12-01 00:00:00')),
     PARTITION p95 VALUES LESS THAN (UNIX_TIMESTAMP('2023-12-10 00:00:00')),
     PARTITION p96 VALUES LESS THAN (UNIX_TIMESTAMP('2023-12-20 00:00:00')),
     PARTITION p97 VALUES LESS THAN (UNIX_TIMESTAMP('2023-12-31 00:00:00')),
     PARTITION p1024 VALUES LESS THAN MAXVALUE
);

写入

insert1:   insert into history_wti_alarm set id = 2,start_time = "2022-11-10 00:58:59",alarm_id = "123";
insert2:   insert into history_wti_alarm set id = 2,start_time = "2022-11-10 07:58:59",alarm_id = "123";
insert3:   insert into history_wti_alarm set id = 2,start_time = "2022-11-10 08:58:59",alarm_id = "123";

查询

select * from history_wti_alarm where start_time > "2022-11-10 00:00:00";

【遇到的问题】：
1 查询之后只能看到insert3的数据，insert1 insert2的数据看不到。理论上这个三个数据都应该落入到p43分区，但是insert1 insert2 落到了分区p42上。
2 新搭建集群6.5.0未发现该问题，于是将历史的集群升级到6.5.0但该问题仍然存在

【资源配置】

[vcdog]

我这边刚好，有一套集群，最近也是刚从v5.4.0版本在线升级到v6.50版本。直接原样使用你的测试sql语句，创建表+插入数据+查看执行计划。查看分区落入区间一切正常。

mysql> explain select * from history_wti_alarm where start_time > "2022-11-10 00:00:00" and start_time < "2022-11-12
00:
+-------------------------+---------+-----------+----------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------+
| id                      | estRows | task      | access object                          | operator info                                                                                                                            |
+-------------------------+---------+-----------+----------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------+
| TableReader_9           | 0.07    | root      |                                        | data:Selection_8                                                                                                                         |
| └─Selection_8           | 0.07    | cop[tikv] |                                        | gt(test.history_wti_alarm.start_time, 2022-11-10 00:00:00.000000), lt(test.history_wti_alarm.start_time, 2022-11-12 00:00:00.000000) |
|   └─TableFullScan_7     | 3.00    | cop[tikv] | table:history_wti_alarm, partition:p43 | keep order:false, stats:pseudo                                                                                                           |
+-------------------------+---------+-----------+----------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------+
3 rows in set (0.01 sec)

mysql> 

mysql> explain select * from history_wti_alarm where start_time >="2022-11-10 00:00:00" and start_time<="2022-11-12 00:
+-------------------------+---------+-----------+----------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------+
| id                      | estRows | task      | access object                          | operator info                                                                                                                            |
+-------------------------+---------+-----------+----------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------+
| TableReader_9           | 0.07    | root      |                                        | data:Selection_8                                                                                                                         |
| └─Selection_8           | 0.07    | cop[tikv] |                                        | ge(test.history_wti_alarm.start_time, 2022-11-10 00:00:00.000000), le(test.history_wti_alarm.start_time, 2022-11-12 00:00:00.000000) |
|   └─TableFullScan_7     | 3.00    | cop[tikv] | table:history_wti_alarm, partition:p43 | keep order:false, stats:pseudo                                                                                                           |
+-------------------------+---------+-----------+----------------------------------------+------------------------------------------------------------------------------------------------------------------------------------------+
3 rows in set (0.00 sec)

[dba-kit (张天师)]

方便输出一下system_time_zone和time_zone的值么

MySQL []> show variables like '%zone%';
+------------------+---------------+
| Variable_name    | Value         |
+------------------+---------------+
| system_time_zone | Asia/Shanghai |
| time_zone        | SYSTEM        |
+------------------+---------------+
2 rows in set (0.001 sec)

[Billdi表弟]

方便把时区搞成utc吗

[Billdi表弟]

张真人，结果是这样的

[Billdi表弟]

@dba-kit

show variables like '%zone%';

tidb可能有些潜在bug，我在建集群的使用用了默认的上海时区，因为涉及海外数据，搞了分区表在insert的时候tidb计算分区的时候用的是上海时区，但是到了建表和查询的时候用的都是utc时区，导致分区裁剪部分数据查询不出来。修改了system_zone值(文档说不能修改在mysql.tidb里面强制更新)，最后集群重启，就完美解决了。

[Billdi表弟]

Proposal: Infer the System Timezone of a TiDB cluster via TZ environment variable

• Author(s): [Zhexuan Yang](file:/Users/ji.wu/Downloads/tidb-5.4.0/docs/design/www.github.com/zhexuany)
• Last updated: 2018/09/09
• Discussion at: Not applicable

Abstract

When it comes to time-related calculation, it is hard for the distributed system. This proposal tries to resolve two problems: 1. timezone may be inconsistent across multiple TiDB instances, 2. performance degradation caused by pushing System down to TiKV . The impact of this proposal is changing the way of TiDB inferring system’s timezone name. Before this proposal, the default timezone name pushed down to TiKV is System when session’s timezone is not set. After this, TiDB evaluates system’s timezone name via TZ environment variable and the path of the soft link of /etc/localtime . If both of them are failed, TiDB then push UTC to TiKV .

Background

After we solved the daylight saving time issue, we found the performance degradation of TiKV side. Thanks for the investigation done by engineers from TiKV. The root cause of such performance degradation is that TiKV infers System timezone name via a third party lib, which calls a syscall and costs a lot. In our internal benchmark system, after this PR, our codebase is 1000 times slower than before. We have to address this.

Another problem needs also to be addressed is the potentially incosistent timezone name across multiple TiDB instances. TiDB instances may reside at different timezone which could cause incorrect calculation when it comes to time-related calculation. Just getting TiDB’s system timezone could be broken. We need find a way to ensure the uniqueness of global timezone name across multiple TiDB’s timezone name and also to leverage to resolve the performance degradation.

Proposal

Firstly, we need to introduce the TZ environment. In POSIX system, the value of TZ variable can be one of the following three formats. A detailed description can be found in this link

* std offset
* std offset dst [offset], start[/time], end[/time]
* :characters

The std means the IANA timezone name; the offset means timezone offset; the dst indicates the leading timezone having daylight saving time.

In our case, which means both TiDB and TiKV , we need care the first and third formats. For answering why we do not need the second format, we need to review how Golang evaluates timezone. In time package, the method LoadLocation reads tzData from pre-specified sources(directories may contain tzData) and then builds time.Location from such tzData which already contains daylight saving time information.

In this proposal, we suggest setting TZ to a valid IANA timezone name which can be read from TiDB later. If TiDB can’t get TZ or the supply of TZ is invalid, TiDB just falls back to evaluate the path of the soft link of /etc/localtime . In addition, a warning message telling the user you should set TZ properly will be printed. Setting TZ can be done in our tidb-ansible project, it is also can be done at user side by export TZ="Asia/Shanghai" . If both of them are failed, TiDB will use UTC as timezone name.

The positive side of this change is resolving performance degradation issue and ensuring the uniqueness of global timezone name in multiple TiDB instances.

The negative side is just adding a config item which is a very small matter and the user probably does not care it if we can take care of it and more importantly guarantee the correctness.

Rationale

We tried to read system timezone name by checking the path of the soft link of /etc/localtime but, sadly, failed at a corner case. The failed case is docker. In docker image, it copies the real timezone file and links to /usr/share/zoneinfo/utc . The timezone data is correct but the path is not. Regarding of UTC , Golang just returns UTC instance and will not further read tzdata from sources. This leads to a fallback solution. When we cannot evaluate from the path, we fall back to UTC .

Compatibility

It does not have compatibility issue as long as the user deploys by tidb-ansible . We may mention this in our release-node and the message printed before tidb quits, which must be easy to understand.

The upgrading process need to be handled in particular. TZ environment variable has to be set before we start new TiDB binary. In this way, the following bootstrap process can benefit from this and avoid any hazard happening.

Implementation

The implementation is relatively easy. We just get TZ environment from system and check whether it is valid or not. If it is invalid, TiDB evaluates the path of soft link of /etc/localtime . In addition, a warning message needs to be printed indicating user has to set TZ variable properly. For example, if /etc/localtime links to /usr/share/zoneinfo/Asia/Shanghai , then timezone name TiDB gets should be Asia/Shanghai .

In order to ensure the uniqueness of global timezone across multiple TiDB instances, we need to write timezone name into variable_value with variable name system_tz in mysql.tidb . This cached value can be read once TiDB finishes its bootstrap stage. A method loadLocalStr can do this job.

Open issues (if applicable)

PR of this proposal: https://github.com/pingcap/tidb/pull/7638/files PR of change TZ loading logic of golang: https://github.com/golang/go/pull/27570