Is it normal for the StorageGRID decommission process to take a long time?

Applies to

Answer

Decommissioning could take days, weeks, or even months to complete. If needed, you can pause the decommissioning procedure during specific stages.

To see the progress of the decommissioning task go to: 

• Grid Topology > Primary Admin node > CMN > Grid Tasks and look at the Decommission completion. When the decommission is nearing completion, progress can level off at around +/- 95%. This is normal behavior.
• Grid Topology > Storage Node > Reports > Used Storage Capacity and verify if the used storage capacity is decreasing.

Additional Information

• Decommissioning procedure for grid nodes
• StorageGrid: Error on decommissioning task
• When EC data migration is finished, the chart of Acitve: % complete will go to 100% once but replicated data migration will get started automatically.
  So the chart of Acitve: % complete will be reset and starts from 0%. This is normal and expected behavior.

• From the chart on StorageGRID Grid Manager -> SUPPORT -> Metrics -> Node (Internal Use) -> Time Spent on IO per Device can be observed that the two disk devices named sdr and sdaj always keep 100% disk utilization.
  This is because this chart is based on iostat Linux command of StorageGRID's base OS and these two disk devices are used for metadata store (StorageGRID-obj-00) which keeps busy doing metadata lookup while data migrating during the decommissioning period.
  root@AMS-SG5712-8T-1:~ # ssh 0 -p 8022
Linux SG 4.19.0-18-amd64 #1 SMP Debian 4.19.208-1+ntapB (2021-10-11) x86_64
-----------------------------------------------------------------------------
This private computer system is for authorized use only and contains software
that is provided under license. Please consult the End User License Agreement
for terms and conditions.
-----------------------------------------------------------------------------
Last login: Wed Feb  8 19:13:14 2023 from 127.0.0.1
root@SG:~ #  multipath -ll
.......
StorageGRID-obj-00 (3600a098000d5e5fc000000b55bb6f1bb) dm-18 NETAPP,INF-01-00
size=3.8T features='4 pg_init_retries 50 queue_mode mq' hwhandler='1 alua' wp=rw
`-+- policy='service-time 0' prio=50 status=active
|- 0:0:0:253 sdr 65:16 active ready running　
`- 7:0:0:253 sdaj 66:48 active ready running　
.......