The previously asked question (SRM - Does DataMiner perform automatic clean up? - DataMiner Dojo) resolved how we can configure DataMiner to automatically SRM information.
I have a question on the procedure of how DataMiner performs the cleaning up. If you for example configure something like this:
<TTL type="creservationinstance">
<Indexing>1D</Indexing>
</TTL>
- Will DataMiner clean up all reservation instances that are created more than one dag ago? Or is it only going to remove reservation instances that are completed?
- Can we remove automatically stale 'Partial' bookings that are forgotten or should we manage this ourselves?
Hi Jarno, It unfortunately seems that changing this XML is not sufficient to enable TTL on the SRM data. We currently believe this was only compatible for the Reservation Instances on Cassandra. Additional changes are required to DataMiner to support this for Reservation Instances on Elasticsearch. I can however already give you an idea of how the TTL logic currently works. (Since this is already available for DOM)
- Will DataMiner clean up all reservation instances that are created more than one day ago? Or is it only going to remove reservation instances that are completed?
- It will remove all reservation instances that have not been modified for more than the provided TTL time. It does not check or consider any state or field of the reservation instances. They are checked every 30 minutes.
- Can we remove automatically stale ‘Partial’ bookings that are forgotten or should we manage this ourselves?
- You should indeed manage this yourself. The TTL logic has no knowledge of partial data. There is also currently no way of flagging an instance depending on configurable conditions.