[Bug] The bookie is occupying a large amount of disk space without releasing it.

[coulsonm]

Search before reporting

• I searched in the issues and found nothing similar.

Read release policy

• I understand that unsupported versions don't get bug fixes. I will attempt to reproduce the issue on a supported version of Pulsar client and Pulsar broker.

User environment

broker version:2.11.0
bookie version:2.10.1
we deploy with docker using the official image

Issue Description

In this Pulsar cluster, a Bookie node's disk usage grows rapidly while others remain balanced.The ledger files dating back 180 days on one node but only 10 days on others. I setted TTL and retention policies for topics. Nothing found about error info in log files. Why and how can i fix it? Thx

Error messages

Reproducing the issue

my cluster run long time. it may not reproducd...

Additional information

No response

Are you willing to submit a PR?

• I'm willing to submit a PR!

[lhotari]

broker version:2.11.0
bookie version:2.10.1
we deploy with docker using the official image

This is an unsupported version with known security vulnerabilities. Please upgrade to one of the supported versions.

In this Pulsar cluster, a Bookie node's disk usage grows rapidly while others remain balanced.The ledger files dating back 180 days on one node but only 10 days on others. I setted TTL and retention policies for topics. Nothing found about error info in log files. Why and how can i fix it? Thx

If these are rocksdb files, it could be related to apache/bookkeeper#4554 / apache/bookkeeper#4555 . The workaround is described in apache/pulsar-helm-chart#633 (comment) .