I run them every 1h, but this is frequency we chose in mutual talk with DBAs.  We actually backup all our database archive logs (Oracle and SAP) and transaction logs (SQL) every 1h.  I believe such approach may be overkill so most likely as we implement smarter approach this may change (eg. backup tiering where production and development get 1h backups while test every 4). Frequency you control via interval (by default is 24:00 so you change it to 01:00).

As for different retentions, I'm not big fan of them nor I believe in those.  Example, if there is jusified reason to have archive logs every 1h, then this means DBAs/business have to ensure PiT for any of those within backup retention. Period. No discussion. So, I use same retention for both db and tlog - which is similar and close to yours (9 days).

I've done this in our environment.

We use 2 different groups.

1 group does the fulls at a set time and the other runs the incrimentals hourly.

I also added a probe that checks to see if a backup is already running so the incrimentals dont try to start when the full is running.

Travis.Baker wrote: I also added a probe that checks to see if a backup is already running so the incrimentals dont try to start when the full is running.

As of 5.2.2 this is no longer needed (as they fixed the issue inside with concurrent TLOG and DB runs).

Good to know thanks!

Sent from my iPhone