Strahil
2019-Apr-22 18:00 UTC
[Gluster-users] Gluster 5.6 slow read despite fast local brick
I've set 'cluster.choose-local: on' and the sequential read is aprox 550MB/s , but this is far below the 1.3G I have observed with gluster v5.5 . Should I consider it a bug, or some options need to be changed ? What about rolling back? I've tried to roll back one of my nodes, but it never came back until I have upgraded to 5.6 . Maybe a full offline downgrade could work... Best Regards, Strahil NikolovOn Apr 22, 2019 17:18, Strahil Nikolov <hunter86_bg at yahoo.com> wrote:> > As I had the option to rebuild the volume - I did it and it still reads quite slower than before 5.6 upgrade. > > I have set cluster.choose-local to 'on' but still the same performance. > > Volume Name: data_fast > Type: Replicate > Volume ID: 888a32ea-9b5c-4001-a9c5-8bc7ee0bddce > Status: Started > Snapshot Count: 0 > Number of Bricks: 1 x (2 + 1) = 3 > Transport-type: tcp > Bricks: > Brick1: ovirt1:/gluster_bricks/data_fast/data_fast > Brick2: ovirt2:/gluster_bricks/data_fast/data_fast > Brick3: ovirt3:/gluster_bricks/data_fast/data_fast (arbiter) > Options Reconfigured: > cluster.choose-local: on > network.ping-timeout: 30 > cluster.granular-entry-heal: enable > performance.strict-o-direct: on > storage.owner-gid: 36 > storage.owner-uid: 36 > user.cifs: off > features.shard: on > cluster.shd-wait-qlength: 10000 > cluster.shd-max-threads: 8 > cluster.locking-scheme: granular > cluster.data-self-heal-algorithm: full > cluster.server-quorum-type: server > cluster.quorum-type: auto > cluster.eager-lock: enable > network.remote-dio: off > performance.low-prio-threads: 32 > performance.io-cache: off > performance.read-ahead: off > performance.quick-read: off > transport.address-family: inet > nfs.disable: on > performance.client-io-threads: off > cluster.enable-shared-storage: enable > > Any issues expected when downgrading the version ? > > Best Regards, > Strahil Nikolov > > > ? ??????????, 22 ????? 2019 ?., 0:26:51 ?. ???????-4, Strahil <hunter86_bg at yahoo.com> ??????: > > > Hello Community, > > I have been left with the impression that FUSE mounts will read from both local and remote bricks , is that right? > > I'm using oVirt as a hyperconverged setup and despite my slow network (currently 1 gbit/s, will be expanded soon), I was expecting that at least the reads from the local brick will be fast, yet I can't reach more than 250 MB/s while the 2 data bricks are NVME with much higher capabilities. > > Is there something I can do about that ? > Maybe change cluster.choose-local, as I don't see it on my other volumes ? > What are the risks associated with that? > > Volume Name: data_fast > Type: Replicate > Volume ID: b78aa52a-4c49-407d-bfd8-fdffb2a3610a > Status: Started > Snapshot Count: 0 > Number of Bricks: 1 x (2 + 1) = 3 > Transport-type: tcp > Bricks: > Brick1: ovirt1:/gluster_bricks/data_fast/data_fast > Brick2: ovirt2:/gluster_bricks/data_fast/data_fast > Brick3: ovirt3:/gluster_bricks/data_fast/data_fast (arbiter) > Options Reconfigured: > performance.client-io-threads: off > nfs.disable: on > transport.address-family: inet > performance.quick-read: off > performance.read-ahead: off > performance.io-cache: off > performance.low-prio-threads: 32 > network.remote-dio: off > cluster.eager-lock: enable > cluster.quorum-type: auto > cluster.server-quorum-type: server > cluster.data-self-heal-algorithm: full > cluster.locking-scheme: granular > cluster.shd-max-threads: 8 > cluster.shd-wait-qlength: 10000 > features.shard: on > user.cifs: off > cluster.choose-local: off > storage.owner-uid: 36 > storage.owner-gid: 36 > performance.strict-o-direct: on > cluster.granular-entry-heal: enable > network.ping-timeout: 30 > cluster.enable-shared-storage: enable > > Best Regards, > Strahil Nikolov > > Hello Community, > > I have been left with the impression that FUSE mounts will read from both local and remote bricks , is that right? > > I'm using oVirt as a hyperconverged setup and despite my slow network (currently 1 gbit/s, will be expanded soon), I was expecting that at least the reads from the local brick will be fast, yet I can't reach more than 250 MB/s while the-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20190422/3a159f1a/attachment.html>