All Products
Search

This Product

    Hologres:Diagnose table metadata

    Document Center

    Hologres:Diagnose table metadata

    Last Updated:Feb 04, 2026

    Data inconsistency between the table metadata in the Storage Master and the frontend (FE) nodes of a Hologres database can cause Data Definition Language (DDL) operations to fail or affect your costs. Hologres provides a table metadata diagnosis feature to check for data inconsistency in your instance. The diagnosis results are updated weekly. You can use the provided solutions to resolve metadata issues, which improves the availability and stability of your instance.

    Procedure

    You can view the table metadata diagnosis results in HoloWeb.

    1. Go to the HoloWeb developer page. For more information, see Connect to HoloWeb and execute a query.

    2. In the top navigation bar, click Diagnostics and Optimization.

    3. In the navigation pane on the left, choose Instance Diagnosis > Table Meta Diagnosis.

    Diagnosis list

    The diagnosis results are updated weekly. The following table describes the issues, their impacts, and the solutions.

    Problem

    Impact

    Solution

    Notes

    The distribution key of a child table is inconsistent with its parent table.

    If this issue is not resolved, queries that hit the corresponding child table will fail.

    Recreate the partitioned parent table and all its child tables.

    Execute this operation during off-peak hours.

    The distribution key of a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The clustering key of a child table is inconsistent with its parent table.

    If this issue is not resolved, queries that hit the corresponding child table will fail.

    Recreate the partitioned parent table and all its child tables.

    Execute this operation during off-peak hours.

    The clustering key of a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The segment key of a child table is inconsistent with its parent table.

    If this issue is not resolved, queries that hit the corresponding child table will fail.

    Recreate the partitioned parent table and all its child tables.

    Execute this operation during off-peak hours.

    The segment key of a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The encryption property of a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The storage mode of a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    To resolve this issue, run the following command: call set_table_property('<schema_name>.<table_name>', 'table_storage_mode', 'hot/cold');. Set the storage mode to hot or cold as required.

    Execute this operation during off-peak hours. Migrating data between hot and cold storage consumes system resources.

    The storage format of a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    To resolve this issue, run the following command to convert the table to the ORC format: call set_table_property('<schema_name>.<table_name>', 'storage_format', 'orc');.

    Execute this operation during off-peak hours. The ORC format conversion process consumes system resources.

    The Proxima vectors of a table are inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The Proxima vectors of a table contain invalid values.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The Time to Live (TTL) of a table is inconsistent with its metadata.

    If this issue is not resolved, the data lifecycle may be affected, which can lead to accidental data deletion.

    Run the following SQL statement to resolve this issue: call set_table_property('<schema_name>.<table_name>', 'time_to_live_in_seconds', '<target_value>');. Set <target_value> to a large value.

    Execute this operation during off-peak hours.

    The binary logging (binlog) TTL of a table is inconsistent with its metadata.

    If this issue is not resolved, the binlog lifecycle may be affected, which can lead to accidental deletion of binlogs.

    Run the following SQL statement to resolve this issue: call set_table_property('<schema_name>.<table_name>', 'binlog.ttl', '<target_value>');. Set <target_value> to a large value.

    Execute this operation during off-peak hours.

    The binlog property of a table is inconsistent with its metadata.

    Binary logging may be disabled for the table but enabled in the metadata. This can increase the table's storage usage.

    Modify the binlog property. For example, you can run the following SQL statement to disable binary logging: call set_table_property('<parent_schema_name>.<parent_table_name>', 'binlog.level','none');. You can also enable binary logging as needed.

    Enable or disable binary logging based on your requirements. Otherwise, related binlog tasks may be affected.

    The binlog property of a child table is inconsistent with its parent table.

    Binary logging may be disabled for the parent table but enabled for a child table, which causes inconsistent binlog properties for the partitioned table. In this case, queries that hit the child table may fail and the table's storage usage may increase.

    Modify the binlog property of the parent table. For example, you can run the following SQL statement to disable binary logging for the parent table: call set_table_property('<parent_schema_name>.<parent_table_name>', 'binlog.level', 'none');. You can also enable binary logging as needed.

    Enable or disable binary logging based on your requirements. Otherwise, related binlog tasks may be affected.

    The number of indexes for a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The index property of a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The readonly property of a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    To resolve this issue, run the following command to set the readonly property of the table to false: call set_table_property('<schema_name>.<table_name>', 'readonly', 'false');.

    Execute this operation during off-peak hours.

    The number of fields in a table is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The table name is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The nullable property of a table field is inconsistent with its metadata.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    A table field name contains unsupported characters.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The table has no fields.

    This is an invalid table. You should delete it to avoid consuming metadata memory.

    Delete the table.

    Execute this operation during off-peak hours.

    The clustering key of a table contains an unsupported data type.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The segment key of a table contains an unsupported data type.

    If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.

    The table metadata has a critical error and cannot be accessed.

    The table has a critical metadata issue. The automatic diagnosis cannot identify the specific cause. If this issue is not resolved, queries will fail.

    Recreate the table.

    Execute this operation during off-peak hours.