I see the original intent was to archive data of a master table to another (non replication) server. This explains the unusual where clause. Maybe pt-archiver is better suited for this since it also deletes old data from master, and can deal with deadlocks (--retries option).
Since the future version of pt-table-sync will be restricted to normal replication, and will work on top of pt-table-checksum results (which is better prepared to deal with deadlocks), this issue can be considered moot.
Although this can be considered a true bug, for the above reasons I'm marking this as won't fix.
Valerii, feel free to reopen or comment if you have other considerations.
Hi Valerii,
Revisiting this old "heisenbug" ...
I see the original intent was to archive data of a master table to another (non replication) server. This explains the unusual where clause. Maybe pt-archiver is better suited for this since it also deletes old data from master, and can deal with deadlocks (--retries option).
Since the future version of pt-table-sync will be restricted to normal replication, and will work on top of pt-table-checksum results (which is better prepared to deal with deadlocks), this issue can be considered moot.
Although this can be considered a true bug, for the above reasons I'm marking this as won't fix.
Valerii, feel free to reopen or comment if you have other considerations.