Package: postgresql-8.2 Version: 8.2.3-1 Severity: wishlist If remaining disk capacity is too small, pg_upgradecluster fails (gracefully) during attempted creation of the new cluster. As an early stage, it could instead estimate the available capacity (perhaps with df) and compare it to the approximate disk usage of the old cluster (maybe with du on the data directory).
Martin suggest refusing to continue if the available capacity is less than current usage, and continuing only after a warning if it is less than 130% of current usage. This scheme will not catch all cases, but should catch the majority of them. Missed cases will remain with the default behavior. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
