Bug #54983: mgr/dashboard: RGW users and buckets tables are empty if the selected gateway is down - Dashboard - Ceph

Actions

Copy link

Bug #54983

closed

mgr/dashboard: RGW users and buckets tables are empty if the selected gateway is down

Added by Volker Theile about 2 years ago. Updated almost 2 years ago.

Status:

Resolved

Priority:

Normal

Assignee:

Volker Theile

Category:

Component - RGW

Target version:

Ceph - v16.2.8

% Done:

Source:

Community (dev)

Tags:

Backport:

pacific, quincy

Regression:

Severity:

3 - minor

Reviewed:

Affected Versions:

Ceph - v16.2.8

ceph-qa-suite:

Pull request ID:

45584

Crash signature (v1):

Crash signature (v2):

Description

It seems the Dashboard or backend is storing the previously selected node somewhere to retrieve the data via the RGW AdminOps API. In a cluster with multiple RGW gateway, if the selected gateway is down, no data is visible in the Dashboard users/bucket pages anymore, even if the data could be displayed if the other still active gateway would be selected.
The problem could be solved by letting the user select the still active gateway or switching the previosly selected (and somewhere stored gateway) automatically to the still running gateway.

How to reproduce:

Setup a cluster with multiple nodes and at least 2 RGW gateways. Go to the RGW daemons page, select one of the gateways displayed in the dropdown. Remember the node where the gateway is deployed. Visit users and buckets page to validate all is working.
Now shutdown the node of the previously selected RGW gateway. The users and bucket page will not show any content now. The daemons page shows only one gateway.

Bootup the halted node again, all pages are showing content again. Now shutdown the node of the gateway that is not selected in the dropdown on the daemons page. Visit users and bucket page. All tables are displaying the data correctly.

Related issues 2 (0 open — 2 closed)