Search before asking

• I had searched in the issues and found no similar issues.

Version

v1.2.680-p2

What's Wrong?

Hi

We have a few table which seem to exist in a purgatory state.

When we try to run UNDROP TABLE gold.illicit_activity_summary; it results in the following message.
SQL Error: Failover Retry Error executing query after 0 failover retry: SQL: UNDROP TABLE gold.illicit_activity_summary Query failed: QueryErrors{code=2308, message=Undrop Table 'illicit_activity_summary' already exists} cause: null

The above exception also occurs when trying to create the table via normal CREATE TABLE and CTAS DDL statements.

Then when we try to run DROP TABLE gold.illicit_activity_summary; it results in this message.
SQL Error: Failover Retry Error executing query after 0 failover retry: SQL: DROP TABLE gold.illicit_activity_summary Query failed: QueryErrors{code=1025, message=Unknown table 'illicit_activity_summary'} cause: null

We currently have 3 tables where this is an issue.

How to Reproduce?

We have not been successful in trying to reproduce this error on demand.
So far it seems to be affecting only 3 tables.

When looking at the system.tables_with_history the affected tables are not present.

We are running 3 meta nodes (4 vcpu, 16 GB memory, ARM64 architecture).
7 query nodes (16 vCpu, 256GB memory, ARM64 architecture)

I have attached a snippet of the debug logs of a node where we tried to run the UNDROP statement.

databend_debug.log

What else can I send to assist in the diagnosis of an issue like this?

Are you willing to submit PR?

• Yes I am willing to submit a PR!