Ceph

I have a cluster with an erasure-code pool as the backend storage and a replicated pool as the cache tier of the erasure-code pool. Can I increase the pg_num and pgp_num of the replicated pool (the cache-tier pool) when the pool is still serving I/O requests? Is there any risk on this operation?

When I submit the following command on my production cluster:
"ceph osd pool set reppool pg_num 128",
an error message is printed on the console: "Error EPERM: splits in cache pools must be followed by scrubs and leave sufficient free space to avoid overfilling. use --yes-i-really-mean-it to force."

How should I do in this case?

Thanks a lot.