Hi Matt,
We upgraded our cluster to 13.2.8 yesterday . After restarting radosgw, gc
process successfully cleaned up those objects and omap.
thanks again !
By the way, for other users , in our case the backlog had been increased
to more than 3 million. the cleanup after upgrading runs really fast and thus
cpu on gc nodes costs much than usual. That may caused by our setting
"rgw_gc_max_concurrent_io = 40 , rgw_gc_max_trim_chunk = 1024". The client IO
of cluster is less than 300MB/s at that time, so the performance hasn't been
affected much. However in our other test , speeding up gc for large amount of
objects can cause severe performance degradation, so it's better to consider
the current client IO level before upgrading.
_______________________________________________
ceph-users mailing list -- ceph-users@ceph.io
To unsubscribe send an email to ceph-users-le...@ceph.io
