June 9, 20197 yr It think it would be super useful to be able to limit the speed for long running I/O jobs using cgroups. For example, I want to preclear a disk that is shared in a jbod with my main array disks, however, the preclearing consumes almost all bandwidth leaving me with a very slow / unreliable experience. If I could limit the preclear IO to something like 50MB/s then it will leave me with a good headroom. Another use case is when checking for parity or rebuilding a disk. Thanks.
Archived
This topic is now archived and is closed to further replies.