copy-cluster consistency

Pietro_La_Torre

Hi, while doing copy-cluster from Cluster A to Cluster B may I use Cluster A for querying? What happens if an object is created or dropped or modified in some way? Is it guaranteed consistency of data copied to Cluster B? Do I have to force database shutdown on Cluster A or does copy-cluster with vbr.py manage database shutdown? thanks, Pietro

Daniel_Leybovic

Hi! >> may I use Cluster A for querying? Yes. >> What happens if an object is created or dropped or modified in some way? You will not see this changes in Cluster B. >> Is it guaranteed consistency of data copied to Cluster B? Data of Cluster B will be identically to data of Cluster A at snapshot creation. All changes on data after backup snapshot creation that done on Cluster A you will not see.

colin_loghin

To your last question, Cluster A should be UP and cluster B should be down for a successful cluster copy. You can orchestrate the entire process ( shutdown cluster B, copy cluster , restart of cluster using admintools command line + shell scripting + vbr.py utility. Good luck.

Pietro_La_Torre

Thank you!!

Daniel_Leybovic

FYI (if you have 2 Vertica servers): may be next algorithm will be a better choice then incremental backup: * CONNECT (Connects to another Vertica database to enable data import ) * COPY FROM VERTICA (Copies data from another Vertica database. ) https://my.vertica.com/docs/6.1.x/HTML/index.htm#15572.htm https://my.vertica.com/docs/6.1.x/HTML/index.htm#15591.htm All this you can automate by scheduled script (or External Procedure) that will update Cluster B. There are one big advantage in this algorithm: no down time for restoring backup