backup: CREATE SCHEDULE for table backup does not save qualified table name

[dt]

CREATE DATABASE mydb; USE mydb; CREATE TABLE mytbl (id int primary key); CREATE SCHEDULE FOR BACKUP mytbl INTO 'userfile:///a' RECURRING '@hourly'

Schedule creation does a dry run backup to see if it will work, which succeeds in this case. But then the backup fails saying the table doesn't exist. This is because the BACKUP is of mytbl but in the background, scheduled job session, that does not resolve to mydb.mytbl.

We should replace the backup that we run with the fully qualified mydb.mytbl name in the serialized backup command.

Epic CRDB-8816

[blathers-crl]

Hi @dt, please add a C-ategory label to your issue. Check out the label system docs.

While you're here, please consider adding an A- label to help keep our repository tidy.

_{🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is otan.}

[adityamaru]

A result of this bug was that scheduled backups with table targets might fail in older releases. Adding backport labels and a GA Blocker tag to get this into 21.2.

[blathers-crl]

Hi @adityamaru, please add branch-* labels to identify which branch(es) this release-blocker affects.

_{🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is otan.}