[Bug] 无法插入数据
- Dominant language
- Java
- Stars
- 6.4k
- Forks
- 1.2k
- Avg merge
- 1d 23h
- Merged PRs (30d)
- 115
Description
### Search before asking
- [X] I searched in the [issues](https://github.com/apache/iotdb/issues) and found nothing similar.
### Version
1.1.2
### Describe the bug and provide the minimal reproduce step
本地环境:
系统:win11,
cpu:i5 13400
系统总内存:32G
iotdb版本:1.1.2
iotdb server配置:
confignode:
MAX_HEAP_SIZE=3072M
cn_internal_address=192.168.1.131
cn_target_config_node_list=192.168.1.131:10710
datanode:
MAX_HEAP_SIZE=3072M
dn_rpc_address=192.168.1.131
dn_internal_address=192.168.1.131
dn_target_config_node_list=192.168.1.131:10710
iotdb-common:
admin_name=xxx
admin_password=xxx
enable_auto_create_schema=false
enable_last_cache=false
query_timeout_threshold=120000
ip_white_list=127.0.0.1/32,192.168.1.131
enable_discard_out_of_order_data=true
timestamp_precision=us
time_partition_interval=31536000000
8月15日下午我使用java原生接口方式执行SessionPool#insertAlignedRecordsOfOneDevice方法把mysql数据循环插入到我本地windows版本的iotdb 1.1.2中,每次循环插入10000条record,期间运行了几个小时,我下午18:30看时还是正常插入的,等8月16日早上9:30看时插入方法已经执行完毕,接着我用SELECT count(imei) from root.bms.pi.pi_2022查询的是3563534(imei在mysql中是非null字段,一定有值,可以统计到),这个时候我发现不对劲,因为按照我的java代码来说应该要插入2000万多一点的record才对,但是实际只插入了3563534条record,然后我把8月15日的插入方法在运行了一段时间,预计应该再插入5w条数据,然后我用SELECT count(imei) from root.bms.pi.pi_2022查询发现record的条数还是3563534(time时间戳是微秒,总共16位,最后6位是随机生成的,不可能全部和已有的时间戳一样,可以排除掉时间戳相同会覆盖的情况)。最后,我把iotdb server端重启,先查询time为1637953692485879的数据不存在,接着手动插入一条time为1637953692485879的record,插入完毕后再查询一次time为1637953692485879的数据,发现还是没有,也就是没有插入成功,log_datanode_all.log的日志从8月15日我开始插入数据后到8月16日都没有一个ERROR的日志。
DBeaver客户端操作如下:



原生cli客户端操作如下:

### What did you expect to see?
希望能找出是什么问题导致的
### What did you see instead?
如上图
### Anything else?
_No response_
### Are you willing to submit a PR?
- [X] I'm willing to submit a PR!
Contributor guide
Research direction
Start with the Java SessionPool#insertAlignedRecordsOfOneDevice call and reproduce the issue against IoTDB 1.1.2 using the supplied Windows configuration and batch size. Compare insert results with SELECT count(imei) and direct CLI/DBeaver inserts, then inspect log_datanode_all.log; done means identifying a reproducible cause or a clear failing response.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- java
- Domain
- databases
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100