I've added my vote.
I've increased the chunk size-from 512 entries (default) to 10'000
entries, as from the xsd: "The number of cache entries to batch in each
transfer." I thought that an increased chunk-size helps to populate the
cache faster? Or is it the opposite? One entry is very small, just
I'm not 100% sure, but this sounds like you're hitting ISPN-5141 [1].
Please add your vote, it might make it into 7.2.0.
Lowering the state transfer chunkSize may also help, especially if your
values are large. The chunkSize is the number of entries in a single chunk,
you should aim at making the
Thanks. The cache is a distributed cache with 3 nodes and replication
factor 2. It contains ~40 million entries. As soon as a node has loaded
all entries from disk, new entries are added to this node with high
frequency (several thousand/s).
The other nodes might not be up yet. It takes 'forever'
