For a staging instance of KoBo, I would need to fully understand how the workflow on S3 works. To do so, let me share the relevant settings I have configured on my KoBo instance:
AWS Access Key : XXXXXX AWS Secret Key : YYYYYY AWS Bucket name : test-bucket Would you like to validate your AWS credentials? Yes AWS credentials successfully validated Do you want to activate backups? Yes Do you want to use WAL-E for continuous archiving of PostgreSQL backups? Yes PostgreSQL backup schedule? [0 2 * * 0]: 0 21 * * * MongoDB backup schedule? [0 1 * * 0]: 0 21 * * * Redis backup schedule? [0 3 * * 0]: 0 21 * * * AWS Backups bucket name : test-backup-bucket How many yearly backups to keep? : 5 How many monthly backups to keep? : 36 How many weekly backups to keep? : 52 How many daily backups to keep? : 180 PostgresSQL backup minimum size (in MB)? Files below this size will be ignored when rotating backups. : 2 MongoDB backup minimum size (in MB)? Files below this size will be ignored when rotating backups. : 1 Redis backup minimum size (in MB)? Files below this size will be ignored when rotating backups. : 1 Chunk size of multipart uploads (in MB)? : 15 Use AWS LifeCycle deletion rule? No
What type of information is stored in the bucket specified under
AWS Bucket name(
test-bucket)? And in which frequency?
Is it safe to use the same bucket for
AWS Bucket nameand
AWS Backups bucket name?
With WAL-E enabled, do I still restore PostgreSQL backups with:
pg_restore -U kobo -d kobotoolbox -c "<path/to/postgres.pg_dump>"?
Regarding the backup minimum size, does it mean that once a backup task is trigerred and the zip/dump file created, if its size is lower than the specified value, it won’t be sent to the backup bucket on S3?
I noticed that specially the backups from MongoDB and Redis are less than 1MB in the beginning, but they should still be sent to the bucket. Can I simply say zero MB to force sending all backups to S3?
Thanks and Regards