Lindsay Mathieson
2016-Apr-17 04:12 UTC
[Gluster-users] Continual heals happening on cluster
gluster 3.7.10 Proxmox (debian jessie) I'm finding the following more than a little concerning. I've created a datastore with the following settings: Volume Name: datastore4 Type: Replicate Volume ID: 0ba131ef-311d-4bb1-be46-596e83b2f6ce Status: Started Number of Bricks: 1 x 3 = 3 Transport-type: tcp Bricks: Brick1: vnb.proxmox.softlog:/tank/vmdata/datastore4 Brick2: vng.proxmox.softlog:/tank/vmdata/datastore4 Brick3: vna.proxmox.softlog:/tank/vmdata/datastore4 Options Reconfigured: features.shard-block-size: 64MB network.remote-dio: enable cluster.eager-lock: enable performance.io-cache: off performance.read-ahead: off performance.quick-read: off performance.stat-prefetch: on performance.strict-write-ordering: off nfs.enable-ino32: off nfs.addr-namelookup: off nfs.disable: on cluster.server-quorum-type: server cluster.quorum-type: auto features.shard: on cluster.data-self-heal: on cluster.self-heal-window-size: 1024 transport.address-family: inet performance.readdir-ahead: on I've transferred 12 Windows VM's to it (gfapi) and am running them all, spread across three nodes. "gluster volume heal datastore3 statistics heal-count" shows zero heals on all nodes. but "gluster volume heal datastore4 info" shows heals occurring on mutliple shards on all nodes, different shards each time its called. gluster volume heal datastore4 info Brick vnb.proxmox.softlog:/tank/vmdata/datastore4 /.shard/d297f8d6-e263-4af3-9384-6492614dc115.221 /.shard/744c5059-303d-4e82-b5be-0a5f53b1aeff.1362 /.shard/bbdff876-290a-4e5e-93ef-a95276d57220.942 /.shard/eaeb41ec-9c0d-4fed-984f-cf832d8d33e0.1032 /.shard/f8ce4b49-14d0-46ef-9a95-456884f34fd4.623 /.shard/e9a39d2e-a1b7-4ea0-9d8c-b55370048d03.483 /.shard/f8ce4b49-14d0-46ef-9a95-456884f34fd4.47 /.shard/eaeb41ec-9c0d-4fed-984f-cf832d8d33e0.160 Status: Connected Number of entries: 8 Brick vng.proxmox.softlog:/tank/vmdata/datastore4 /.shard/bd493985-2ee6-43f1-b8d5-5f0d5d3aa6f5.33 /.shard/d297f8d6-e263-4af3-9384-6492614dc115.48 /.shard/744c5059-303d-4e82-b5be-0a5f53b1aeff.1304 /.shard/d297f8d6-e263-4af3-9384-6492614dc115.47 /.shard/719041d0-d755-4bc6-a5fc-6b59071fac17.142 Status: Connected Number of entries: 5 Brick vna.proxmox.softlog:/tank/vmdata/datastore4 /.shard/d297f8d6-e263-4af3-9384-6492614dc115.357 /.shard/bbdff876-290a-4e5e-93ef-a95276d57220.996 /.shard/d297f8d6-e263-4af3-9384-6492614dc115.679 /.shard/d297f8d6-e263-4af3-9384-6492614dc115.496 /.shard/eaeb41ec-9c0d-4fed-984f-cf832d8d33e0.160 /.shard/719041d0-d755-4bc6-a5fc-6b59071fac17.954 /.shard/d297f8d6-e263-4af3-9384-6492614dc115.678 /.shard/719041d0-d755-4bc6-a5fc-6b59071fac17.852 /.shard/bbdff876-290a-4e5e-93ef-a95276d57220.1544 Status: Connected Number of entries: 9 -- Lindsay Mathieson -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160417/203cdefb/attachment.html>
Lindsay Mathieson
2016-Apr-17 09:09 UTC
[Gluster-users] Continual heals happening on cluster
I shutdown all the vms, waited for heals to finish, then stopped tall gluster processes. set the following to off: performance.write-behind datastore4 performance.flush-behind datastore4 cluster.data-self-heal and restarted everything - same issue - assuming it actually is an "issue" -- Lindsay Mathieson -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160417/80371ba9/attachment.html>