Strange. I am getting these tablet issues:

2025-10-13 11:36:27.780 CEST
Starrocks FE 1
host:starrocks-fe-1 platform:StarRocks 2025-10-13 09:36:25.816Z ERROR (stats-cache-refresher-0|1218) [OlapScanNode.addScanRangeLocations():567] Failed to get scan range, no queryable replica found in tablet=10057 replica=10219:1019306/-1/1019306/1019295:BAD:ALIVE, schema_hash=430336460 version=1019306

2025-10-13 11:36:31.873 CEST
Starrocks FE 1
host:starrocks-fe-1 platform:StarRocks 2025-10-13 09:36:30.609Z ERROR (stats-cache-refresher-1|1219) [OlapScanNode.addScanRangeLocations():567] Failed to get scan range, no queryable replica found in tablet=10057 replica=10219:1019306/-1/1019306/1019295:BAD:ALIVE, schema_hash=430336460 version=1019306

2025-10-13 11:36:31.873 CEST
Starrocks FE 1
host:starrocks-fe-1 platform:StarRocks Caused by: com.starrocks.common.UserException: Failed to get scan range, no queryable replica found in tablet=10057 replica=10219:1019306/-1/1019306/1019295:BAD:ALIVE, schema_hash=430336460 version=1019306

2025-10-13 11:36:33.008 CEST
Starrocks FE 3
host:starrocks-fe-3 platform:StarRocks 2025-10-13 09:36:30.608Z ERROR (stats-cache-refresher-7|217) [OlapScanNode.addScanRangeLocations():567] Failed to get scan range, no queryable replica found in tablet=10065 replica=10185:1019397/-1/1019397/1019397:NORMAL:ALIVE, schema_hash=430336460 version=1019306

2025-10-13 11:36:33.008 CEST
Starrocks FE 3
host:starrocks-fe-3 platform:StarRocks Caused by: com.starrocks.common.UserException: Failed to get scan range, no queryable replica found in tablet=10065 replica=10185:1019397/-1/1019397/1019397:NORMAL:ALIVE, schema_hash=430336460 version=1019306

When I show tablet on those ID's, it reveals that we are having problems with the statistics.column_statistics table. This is further confirmed when I run:

select * from _statistics_.column_statistics;

- giving me:

Error: ER_PARSE_ERROR: Build Exec OlapScanNode fail, scan info is invalid
[Done] Finished MySQL query.

How do I solve this problem? Starrocks will not allow me to repair the table, saying that it is not an OLAP table. It seems that this problem is causing quite a lot of dead CPU cycles on my BE nodes.

can force drop

_statistics_

database, the db and tables will be rebuilt by FE service.

I force dropped it, now I just get spammed with errors that the database is missing 😕

wait for a few minutes, it will be recreated.