Motivation

The #21816 introduces the pulsar_storage_backlog_age_seconds metrics, and refactors logic for checkTimeBacklogExceeded. It has two problems:

1. Even if the consumers under the topic have consumed all the data, pulsar_storage_backlog_age_seconds will not become 0 and will continue to increase over time. This is because its calculation: current time - publish time of the last message.
2. If the user has configured BacklogQuota.message_age, it will also affect the ability of producers to continue sending messages, even if there is currently no backlog under this topic.

Modifications

1. In PersistentTopic.checkTimeBacklogExceeded method, if the topic has no backlog, it should return false directly.
2. In PersistentTopic.asyncGetStats method, if the topic has no backlog, it should be set oldestBacklogMessageAgeSeconds = 0

Verifying this change

• Add backlogsStatsPreciseWithNoBacklog unit test to cover this case with configuration: preciseTimeBasedBacklogQuotaCheck=true, exposePreciseBacklogInPrometheus=true
• Add backlogsStatsNotPreciseWithNoBacklog unit test to cover this case with configuration: preciseTimeBasedBacklogQuotaCheck=false, exposePreciseBacklogInPrometheus=false

Does this pull request potentially affect one of the following parts:

If the box was checked, please highlight the changes

• Dependencies (add or upgrade a dependency)
• The public API
• The schema
• The default values of configurations
• The threading model
• The binary protocol
• The REST endpoints
• The admin CLI options
• The metrics
• Anything that affects deployment

Documentation

• doc
• doc-required
• doc-not-needed
• doc-complete

Matching PR in forked repository

PR in forked repository: