You are currently reviewing an older revision of this page.
KDB file numbering for one or more engines moves into the negatives.
For example, the .kdb file on disk is pe-2147483644.kdb and the negative DB ID is referenced in the corresponding db_ log.In the gateway log we can see that the standard, positive integer numbering is being referenced and then in the next checkpoint, the negative numbering for the DB ID is called.
I believe this issue can occur after improperly resetting a particular engine's KDBs.
To correct it, I think shutting down the engine and then renaming the kdbs to positive numbers (making sure the most recent is highest of course) is sufficient.
Also, if you're currently running on a negative numbered kdb, you should immediately make sure that there are no positive numbered kdbs in the gw directory. The usual "highest number kdb" rules still apply, so we want to make sure any restart would select the kdb you're currently writing to.
We just need to add a boiler plate warning saying that renaming .kdb files is not supported beyond what is documented in this article.
Last Reviewed: November 2017