I know this is not a good test, but when I dd to a rbd image like this [@ ~]# dd if=/dev/zero of=/dev/rbd0 bs=1M count=100 status=progress 100+0 records in 100+0 records out 104857600 bytes (105 MB, 100 MiB) copied, 0.399629 s, 262 MB/s It is a big cache difference doing this via the tcmu-runner [@~]# dd if=/dev/zero of=/dev/sdac bs=1M count=100 status=progress 100+0 records in 100+0 records out 104857600 bytes (105 MB, 100 MiB) copied, 10.3571 s, 10.1 MB/s Is there anything I can change so the default config of tcmu-runner behaves like the rbdmap? _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |