amazon rds - Unable to restore SQL Server bak file from S3, says file too large -

i'm trying run restore query bak file stored in s3 bucket rds sql server web edition, , kept getting error:

[2017-09-13 20:30:22.227] aborted task because of task failure or concurrent restore_db request. [2017-09-13 20:30:22.287] there not enough space on disk perform restore database operaton.

the bak file 77 gb , db has 2tb, how come still not enough?

this query aws docs:

exec msdb.dbo.rds_restore_database          @restore_db_name='database_name',          @s3_arn_to_restore_from='arn:aws:s3:::bucket_name/file_name_and_extension'; 

source: http://docs.aws.amazon.com/amazonrds/latest/userguide/sqlserver.procedural.importing.html#sqlserver.procedural.importing.native.using

there not enough space on disk perform restore database operaton.

...

the bak file 77 gb , db has 2tb, how come still not enough?

you need 2tb of space able restore backup.

the fact restore operation reconstruct original database 2tb.

backup backing data, not empty space. if backup 77gb , not compressed means your original database has 77gb of data (or less, because backup contains amout of log well).

any database consists of data file(s) , log file(s), , if db 2tb 77gb of data, means has enormous log file. think it's in full recovery model , not take regular log backups (or did not take log backup @ all!!)

so should take @ original db, change recovery model simple if don't need point in time recovery , take no log backups, or, if need full recovery model, should backup log more frequently.

taking regular log backups or changing recovery model simple permit shrink log reasonable size, moment not need 2gb of space restore anymore