Hi Percona community,
I’m looking for guidance on the best way to use pt-table-checksum in a vendor-managed MySQL environment.
Environment:
- Oracle Unified Assurance application
- Bundled MySQL Enterprise Edition as a third-party/vendor-managed component
- Two-node MySQL replication setup
- Using Percona Toolkit from internally approved RPMs
- Goal: monitor replication/table drift and alert when checksum differences are detected
pt-table-checksum is working technically when using the default replicate table:
--replicate=percona.checksums
--create-replicate-table
However, we have a concern that creating a new schema named percona inside the vendor-managed MySQL instance may not be allowed under our product/license/support boundaries.
My questions:
- Is it fully supported to use
pt-table-checksumwith an existing application or monitoring schema instead, for example:
--replicate=Assure1.pt_table_checksums
or:
--replicate=ua_monitoring.pt_table_checksums
-
Does
pt-table-checksumrequire the replicate table to be in a separate schema, or can it safely live in any replicated schema/table as long as both source and replica can access it? -
For environments where creating any new schema/table is not allowed, is there a recommended Percona Toolkit alternative for recurring drift detection?
-
Would
pt-table-sync --printbetween two DSNs be a reasonable read-only alternative for alerting, or is that considered too heavy/risky for periodic monitoring? -
Are there any best practices for using
pt-table-checksumin vendor-managed or restricted MySQL environments where DB object creation must be minimized?
Our current requirement is alerting only. We don’t need automatic repair. We just need reliable detection of drift with minimal impact and minimal changes to the vendor-managed database.
Thanks in advance for any recommendations or real-world experience.
Regards, Gokul