influxd inspect build-tsi
influxd inspect build-tsi command rebuilds the TSI index and, if necessary,
the series file.
influxd inspect build-tsi [flags]
InfluxDB builds the index by reading all Time-Structured Merge tree (TSM) indexes and Write Ahead Log (WAL) entries in the TSM and WAL data directories. If the series file directory is missing, it rebuilds the series file. If the TSI index directory already exists, the command will fail.
Use the following options to adjust the performance of the indexing process:
--max-log-file-size determines how much of an index to store in memory before
compacting it into memory-mappable index files.
If you find the memory requirements of your TSI index are too high, consider
decreasing this setting.
--max-cache-size defines the maximum cache size.
The indexing process replays WAL files into a
If the maximum cache size is too low, the indexing process will fail.
--max-cache-size to account for the size of your WAL files.
--batch-size defines the size of the batches written into the index.
Altering the batch size can improve performance but may result in significantly
higher memory usage.
|Size of the batches to write to the index. Defaults to
10000. See above.
|Bucket ID (required if
--shard-id is present).
|Compact existing series file. Does not rebuilt index.
|Number of workers to dedicate to shard index building. Defaults to
GOMAXPROCS (8 by default).
|Path to the TSM data directory. Default is
|Help for the
|Maximum cache size. Defaults to
1073741824. See above.
|Maximum log file size. Defaults to
1048576. See above .
|Shard ID (requires a
|Enable verbose output.
|Path to the WAL data directory. Defaults to
Was this page helpful?
Thank you for your feedback!
Support and feedback
Thank you for being part of our community! We welcome and encourage your feedback and bug reports for InfluxDB and this documentation. To find support, use the following resources:
Customers with an annual or support contract can contact InfluxData Support.