[Gluster-users] ls slow performance

Paladox thomasmulhall410 at yahoo.com
Mon Sep 23 21:57:32 UTC 2019

Hi, we are seeing slow performance with using ls. it is also causing our wikis to go down due to the mount taking a long time to respond (which causes our health checks to depool).
We use MediaWiki.
Any idea what config's i should adjust (and to what) to try and alleviate this as much as possible? (the slowness happens once every while).
Our config:

cluster.use-compound-fops: on

performance.client-io-threads: on

client.tcp-user-timeout: 42

cluster.rebal-throttle: lazy

cluster.nufa: off

cluster.server-quorum-type: none

performance.parallel-readdir: on

server.event-threads: 8

client.event-threads: 8

network.ping-timeout: 10

server.ssl: on

client.ssl: on

features.inode-quota: off

features.quota: off

features.shard: on

transport.address-family: inet

nfs.disable: on

performance.io-thread-count: 2

performance.stat-prefetch: on

performance.write-behind-window-size: 256MB

features.cache-invalidation: on

features.cache-invalidation-timeout: 600

performance.cache-invalidation: on

performance.md-cache-timeout: 600

network.inode-lru-limit: 200000

performance.high-prio-threads: 64

performance.normal-prio-threads: 64

performance.low-prio-threads: 64

storage.health-check-timeout: 60

storage.health-check-interval: 60

storage.owner-uid: -1

storage.owner-gid: -1

performance.cache-size: 256MB

dht.force-readdirp: off

performance.force-readdirp: off

performance.cache-max-file-size: 128MB

performance.cache-refresh-timeout: 60

server.outstanding-rpc-limit: 2056
disperse.eager-lock: off
We use HHD.
We really would like to move from LizardFS seeing as gluster has ssl/tls.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20190923/baf29d59/attachment.html>

More information about the Gluster-users mailing list