Rebalance some storage shards

Production Change - Criticality 2 C2

Rebalance some storage shards Prevent our storage servers from running out of space
Change Type C2
Services Impacted GitLab.com
Change Team Members @skarbek
Change Severity How critical is the change
Buddy check A colleague will review the change
Tested in staging Kinda...
Schedule of the change TBD
Duration of the change Time to execute the change ( including a possible rollback )
Detailed steps for the change. Each step must include: See below

Procedure

  • scp the storage script gitlab-com/runbooks!878 (merged) into a path where the git user has access
    • scp storage_rebalance.rb console-01-sv-gprd.c.gitlab-production.internal:/tmp/
  • Start by logging into the console server
  • This is a repetitive procedure that will repeat PER server using the below table as substitutes

Steps

  1. sudo gitlab-rails runner /tmp/storage_rebalance.rb --current-file-server <SERVERNAME> --target-file-server <SERVERNAME> --move-amount <MOVE_AMOUNT> --dry-run false --wait 600 | tee <current-file-server-name>.log

Movement Table

Current File Server Target File Server Move Amount Run
nfs-file13 nfs-file25 1600 ✅
nfs-file14 nfs-file26 2080 ✅
nfs-file15 nfs-file27 1280 ✅
nfs-file16 nfs-file28 2240 ✅
nfs-file17 nfs-file25 1760 ✅
nfs-file18 nfs-file26 2400 ✅
nfs-file19 nfs-file27 1600 ✅
nfs-file20 nfs-file28 1760 ✅
nfs-file21 nfs-file25 1920 ❗
nfs-file22 nfs-file26 3520 ✅
nfs-file23 nfs-file27 2560 ✅
nfs-file24 nfs-file28 3040 ✅
  • nfs-file25 will have approximately 44% used space
  • nfs-file26 will have approximately 61% used space
  • nfs-file27 will have approximately 45% used space
  • nfs-file28 will have approximately 55% used space

/label C2 ~S2 change

Edited Feb 07, 2019 by John Skarbek
Assignee Loading
Time tracking Loading