On 03/07/18 13:46, John Spray wrote: > To directly address that warning rather than silencing it, you'd > increase the number of PGs in your primary data pool. Since the number of PGs per OSD is limited (or, at least, a recommended limit), I rather prefer to invest them in my datapools. Since I am using erasure coding in my datapools, the PG explosion is important (My EC is 8+2). Wasting PGs in an empty datapool (the empty root datapool is not using EC, it is using size:3), when that is irreversible, is... a waste. My resources are scarce. > There's a conflict here between pools with lots of data (where the MB > per PG might be the main concern, not the object size), vs. > metadata-ish pools (where the object counter per PG is the main > concern). Maybe it doesn't really make sense to group them all > together when calculating the average object-per-pg count that's used > in this health warning -- I'll bring that up over on ceph-devel in a > moment. This is a good point. Some weeks ago I asked about what is worse for OSD: number of PGs or number of objects stored. As a programmer I would say that number of objects is worse (you need to track each object), so better to have 1000 PGs with 100 objects than 10 PGs with a million objects. Nevertheless the answer is the list was "objects don't matter, PGs do". I don't understand the reason for it. -- Jesús Cea Avión _/_/ _/_/_/ _/_/_/ jcea@xxxxxxx - http://www.jcea.es/ _/_/ _/_/ _/_/ _/_/ _/_/ Twitter: @jcea _/_/ _/_/ _/_/_/_/_/ jabber / xmpp:jcea@xxxxxxxxxx _/_/ _/_/ _/_/ _/_/ _/_/ "Things are not so easy" _/_/ _/_/ _/_/ _/_/ _/_/ _/_/ "My name is Dump, Core Dump" _/_/_/ _/_/_/ _/_/ _/_/ "El amor es poner tu felicidad en la felicidad de otro" - Leibniz
Attachment:
signature.asc
Description: OpenPGP digital signature
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com