Hello everybody, since auto sharding does not work on replicated clusters (we only share the user account and metadata and not the actual data) I would like to implement it on my own. But when I reshard a bucket from 53 to 101 (yep, we have two buckets with around 8m files in it) it takes a long time. So my question is: does this somehow affect customer workload, or do I put their data in danger, when I reshard and they upload files? And how do you approach this problem? Do you have a very high default for all buckets, or do you just ignore the large omap objects message? Cheers Boris _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |