[MAGNOLIA-9214] Instances randomly fail to start up on fast machines Created: 06/Dec/23  Updated: 30/Jan/24  Resolved: 30/Jan/24

Status: Closed
Project: Magnolia
Component/s: None
Affects Version/s: 6.2.38
Fix Version/s: 6.3.0

Type: Bug Priority: Neutral
Reporter: Maxime Michel Assignee: Mikaël Geljić
Resolution: Fixed Votes: 0
Labels: 6.3-foundation, VN-Testing, dx-core-6.3
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: PNG File Screenshot 2023-12-05 at 10.50.37.png    
Issue Links:
Cloners
clones MAGNOLIA-9097 Instances randomly fail to start up o... Closed
Problem/Incident
Relates
relates to MAGNOLIA-6593 resource-loader module throws NPE upo... Closed
relates to MGNLDIFF-146 Revise permissions to diff with modul... Closed
relation
Template:
Acceptance criteria:
Empty
Task DoD:
[X]* Doc/release notes changes? Comment present?
[X]* Downstream builds green?
[X]* Solution information and context easily available?
[X]* Tests
[X]* FixVersion filled and not yet released
[ ]  Architecture Decision Record (ADR)
Bug DoR:
[X]* Steps to reproduce, expected, and actual results filled
[X]* Affected version filled
Release notes required:
Yes
Date of First Response:
Story Points: 3
T-Shirt Size: Medium
Team: Foundation
Work Started:
Approved:
Yes

 Description   

USer Story: users are experiencing sporadic failures during the startup of Magnolia instances, specifically on faster machines. The issue is related to a delay in applying server configuration or order of applying certain part of the configuration. This started to appear after the upgrade to Magnolia version 6.2.38, although previous versions also can be affected

 

As described here: https://magnolia-cms.slack.com/archives/CC1GVEGFP/p1694701640514539

TL;DR (apchelintcev)

as I see it - the problem is that core brings in the correct server config, but there’s a 1 second gap before it is applied and new beans are produced on its basis, which may become visible on the newer/faster machines

re: fix - guess we could slash the delay during the install time to smth really small, and/or manually re-trigger the sensitive beans directly from core module installation phase

Dev note:

  • to reproduce the issue steadily change this delay to 5000.

Generated at Mon Feb 12 04:39:38 CET 2024 using Jira 9.4.2#940002-sha1:46d1a51de284217efdcb32434eab47a99af2938b.