Changing RETENTION_DAYS with nodestore

[kodebach]

Self-Hosted Version

25.10.0

CPU Architecture

x86_64

Docker Version

28.3.2

Docker Compose Version

28.3.2

Machine Specification

• My system meets the minimum system requirements of Sentry

Steps to Reproduce

1. Run ./install.sh
2. Change SENTRY_EVENT_RETENTION_DAYS in .env
3. Re-run ./install.sh

Expected Result

The new SENTRY_EVENT_RETENTION_DAYS set in .env is used.

Actual Result

AFAICT with the new seaweedfs container the lifecycle policy is only set during initial setup in install/bootstrap-s3-nodestore.sh

The setup for the lifecycle policy is behind this if

self-hosted/install/bootstrap-s3-nodestore.sh

Lines 10 to 12 in 976e065

	if [[ $(echo "$bucket_list" | tail -1 | awk '{print $3}') != 's3://nodestore' ]]; then
	apply_config_changes_nodestore=0
	# Only touch if no existing nodestore config is found

and the else branch is just a no-op

self-hosted/install/bootstrap-s3-nodestore.sh

Lines 66 to 89 in 976e065

	# XXX(aldy505): Should we refactor this?
	lifecycle_policy=$(
	cat <<EOF
	<?xml version="1.0" encoding="UTF-8"?>
	<LifecycleConfiguration>
	<Rule>
	<ID>Sentry-Nodestore-Rule</ID>
	<Status>Enabled</Status>
	<Filter></Filter>
	<Expiration>
	<Days>$SENTRY_EVENT_RETENTION_DAYS</Days>
	</Expiration>
	</Rule>
	</LifecycleConfiguration>
	EOF
	)
	$dc exec seaweedfs sh -c "printf '%s' '$lifecycle_policy' > /tmp/nodestore-lifecycle-policy.xml"
	$s3cmd --access_key=sentry --secret_key=sentry --no-ssl --region=us-east-1 --host=localhost:8333 --host-bucket='localhost:8333/%(bucket)' setlifecycle /tmp/nodestore-lifecycle-policy.xml s3://nodestore
	echo "Making sure the bucket lifecycle policy is all set up correctly..."
	$s3cmd --access_key=sentry --secret_key=sentry --no-ssl --region=us-east-1 --host=localhost:8333 --host-bucket='localhost:8333/%(bucket)' getlifecycle s3://nodestore
	else
	echo "Node store already exists, skipping..."
	fi

This is probably a bit of an edge case, and some additional documentation about changing the retention period might be enough. I just noticed the issue while looking at the script, while manually upgrading to 25.10.0 after skipping the nodestore change for 25.9.0.

Event ID

No response

[aldy505]

Do you want to write the fix for it?

[kodebach]

See #4040 for a baseline fix, but it probably needs a bit more work to cover all cases

[aldy505]

@kodebach FYI I'm not confident to include the fix on 25.11.0, we'll test some more for 25.12.0