Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.Original topic: 如何查看titandb里的Blob内容都是属于那些表的
|
username: TiDB_C罗
[TiDB Usage Environment] Production Environment / Testing / PoC
[TiDB Version]
[Reproduction Path] What operations were performed when the issue occurred
[Encountered Issue: Problem Phenomenon and Impact]
[Resource Configuration] Enter TiDB Dashboard - Cluster Info - Hosts and take a screenshot of this page
[Attachments: Screenshots/Logs/Monitoring]
An error occurred during a normal business DDL change (only creating an index). The issue is similar to this one Error 1105 (HY000): DDL job rollback, error reported when executing create index statement after PD scaling. After communicating with the user, we performed a reload operation during a low business period.
When executing tiup cluster reload cluster_name, the following error occurred:
Error: init config failed: xxx.xxx.xxx.xxx:20160: executor.ssh.execute_failed: Failed to execute command over SSH for 'tidb@xxx.xxx.xxx.xxx:22' {ssh_stderr: /etc/profile.d/history.sh: line 1: HISTSIZE: readonly variable
/etc/profile.d/history.sh: line 2: HISTTIMEFORMAT: readonly variable
invalid configuration: check: titandb-data-dir-empty; err: "data-dir.empty.check: the number of file with extension blob in directory /work/tidb-data/tikv-20160/db/titandb is non-zero, got 1587, expect 0."; hint: You have disabled titan when its data directory is not empty. To properly shutdown titan, please enter fallback blob-run-mode and wait till titandb files are all safely ingested.
, ssh_stdout: , ssh_command: export LANG=C; PATH=$PATH:/bin:/sbin:/usr/bin:/usr/sbin /work/tidb-deploy/tikv-20160/bin/tikv-server --config-check --config=/work/tidb-deploy/tikv-20160/conf/tikv.toml --pd "" --data-dir "/work/tidb-data/tikv-20160"}, cause: Process exited with status 1: check config failed
This node previously had [rocksdb.titan] enabled = true. Later, we found that the disk space was unbalanced (TiKV存储目录不均衡 - TiDB 的问答社区), so we disabled titan and restarted this TiKV node. Monitoring showed a decrease, but it was very slow. How can we check which tables these blob data belong to and perform a compact operation on a single table?
