helm upgrade: update_database script fails

Summary

I get an error running the upgrade_database script to upgrade to v4.0.0

Steps to reproduce

curl -s https://gitlab.com/gitlab-org/charts/gitlab/raw/${GITLAB_RELEASE}/scripts/database-upgrade | bash -s -- -r gitlab pre
Error from server (NotFound): pods "gitlab-task-runner-69db984cbf-5s5lh\ngitlab-task-runner-backup-1590027420-xhqlr\ngitlab-task-runner-backup-1590113820-v6x8z" not found

Configuration used

(Please provide a sanitized version of the configuration used wrapped in a code block (```yaml))

(Paste sanitized configuration here)

Current behavior

The problem is presumably with $(selector task-runner). My pods are currently:

% kc get pods
NAME                                          READY   STATUS      RESTARTS   AGE
gitlab-gitaly-0                               1/1     Running     0          36h
gitlab-gitlab-exporter-846cc94d86-q9gfl       1/1     Running     0          36h
gitlab-gitlab-runner-658cb7984b-t6b7w         1/1     Running     0          25h
gitlab-gitlab-shell-85b6c4857b-5fxpn          1/1     Running     0          3d20h
gitlab-gitlab-shell-85b6c4857b-xw27h          1/1     Running     0          36h
gitlab-grafana-9cbdf479c-4k259                2/2     Running     0          24h
gitlab-postgresql-0                           2/2     Running     0          36h
gitlab-prometheus-server-86bbc4c747-zhgfv     2/2     Running     0          36h
gitlab-redis-master-0                         2/2     Running     0          36h
gitlab-registry-785fd9b54d-wc2th              1/1     Running     0          25h
gitlab-registry-785fd9b54d-xb4gz              1/1     Running     0          24h
gitlab-sidekiq-all-in-1-v1-568cb6b77b-hwz77   1/1     Running     0          25h
gitlab-task-runner-69db984cbf-5s5lh           1/1     Running     0          36h
gitlab-task-runner-backup-1590027420-xhqlr    0/1     Completed   0          36h
gitlab-task-runner-backup-1590113820-v6x8z    0/1     Completed   0          12h
gitlab-unicorn-6f4bb4f466-l5ww2               2/2     Running     0          24h
gitlab-unicorn-6f4bb4f466-znr5b               2/2     Running     0          36h

NB -- note the backup scripts run.

Expected behavior

Script to run without fiddling around.

Versions

• Chart: gitlab-3.3.5 (upgrading to 4.0.0)
• Platform:
  • Cloud: EKS
• Kubernetes: (kubectl version)
  • Client: 1.18.1
  • Server: 1.15.11
• Helm: (helm version)
  • Client: v3.0.1
  • Server: --

Relevant logs

(Please provide any relevate log snippets you have collected, using code blocks (```) to format)