Hello,
We are in a multi-site indexer cluster environment, and we are going to upgrade our infrastructure from 3 Indexer to 6 Indexer.
Basically we will add 6 new indexers, and decommission the 3 old indexers.
Do you know which is the speed in MBytes/s that will be reached once the data will start to be copied from the 3 old Indexers to the 6 new indexers?
We did some test in our development environment, where we have simulated a similar scenario, and it seems it started copying very fast with 170 MBytes/s peak (we have checked it with an nmon session at the source Indexer machine). In order to start the copying process we had:
- added the new Indexers to the Master Node
- switched the Splunk HFs to forward data to the new Indexers
- run, one by one on each old Indexer, the command splunk offline --enforce-counts
Before starting the test we have read the following documentation:
https://docs.splunk.com/Documentation/Splunk/latest/Indexer/Takeapeeroffline#Estimate_the_cluster_re...
In the official documentation it is mentioned:
10GB (rawdata and/or index files) from one peer to another across a LAN takes about 5-10 minutes
therefore copying speed should go from 136 Mbit/s to 272 Mbit/s.
If the speed we have observed is correct (more than 1Gbit/s), do you know if there is any way to limit the output bandwidth?
Thanks a lot,
Edoardo
