LizardFS Metrics

This plugin/script for Telegraf will collect the metrics from LizardFS and stores it into InfluxDB, then you can view your metrics in Grafana on a templated dashboard.

Full description at https://blog.kruyt.org/lizardfs-in-grafana/

Pre install

Make sure you have installed InfluxDB as the time-series database Telegraf as the collector first.

Install

Get all the files, at My GitHub page.

• Put lizardfs.conf and lizardfs.sh in your /etc/telegraf/telegraf.d directory.
• Edit the file lizardfs.sh and change LFSADMIN variable as appropriate for your lizardfs-admin binary.
• Edit the file lizardfs.sh and change the LIZARDHOSTS to your lizardmaster, you can add multple liardclusters here as needed.

Test

test with:

telegraf --test --config /etc/telegraf/telegraf.d/lizardfs.conf

It should produce some thing like this:

* Plugin: inputs.exec, Collection 1
> lizard_info,lizardmaster=192.168.0.1,host=lxh-sysa-005 FS_objects=470118,Directories=36961,Files=432583,Memory_usage=198946816,Total_space=37159571947520,Available_space=22595288821760,Trash_files=0,Chunk_copies=643179,Regular_copies_deprecated=643179,Trash_space=0,Reserved_space=0,Reserved_files=0,Chunks=214393 1502747150000000000
> lizard_ready-chunkservers-count,lizardmaster=192.168.0.1,host=lxh-sysa-005 chunkservers-count=10 1502747150000000000
> lizard_chunks-health,state=AVA,goal=1,host=lxh-sysa-005,lizardmaster=192.168.0.1 safe=0,unsafe=0,lost=0 1502747150000000000
> lizard_chunks-health,goal=2,host=lxh-sysa-005,lizardmaster=192.168.0.1,state=AVA unsafe=0,lost=0,safe=0 1502747150000000000
> lizard_chunks-health,lizardmaster=192.168.0.1,state=AVA,goal=3,host=lxh-sysa-005 unsafe=0,lost=0,safe=214393 1502747150000000000

***snip***

> lizard_disks,lizardmaster=192.168.0.1,chunkserver=192.168.174.63:9533,path=/data/1/data/3.10.4/,host=lxh-sysa-005 max_fsync_time=0,used_space=705392361472,chunks=32392,read_bytes=0,max_write_time=0,read_ops=0,fsync_ops=0,to_delete="no",damaged="no",scanning="no",written_bytes=0,max_read_time=0,last_error="0",total_space=1857978597376,write_ops=0 1502747150000000000

Restart telegraf

execute:

service telegraf restart

After this it should log the statistics of LizardFS in InfluxDB. Now you can import the dashboard Grafana LizardFS Metrics.json for LizardFS in Grafana.