[CONFSERVER-73508] Confluence index size calculation not working for manual reindex from scratch after split index upgrade task

Type: Bug
Resolution: Fixed
Priority: Low
Fix Version/s: 9.2.0, 8.5.18
Affects Version/s: 7.9.0, 7.13.0, 8.7.2
Component/s: Search - Indexing
Labels:

Support reference count:
12
Symptom Severity:
Severity 3 - Minor
UIS:
2
Bug Fix Policy:
View Atlassian Server bug fix policy

Problem

The logic to calculate the index size for one of the analytics events (finished_indexing) is not working since the index was split into content and change segments.

Environment

Confluence cluster with two or more nodes on version 7.9+ (tested with 7.13)

Steps to Reproduce

Stop Confluence on all nodes
Add the following system property as you would do for a manual rebuild from scratch:
```
confluence.cluster.index.recovery.num.attempts=0
```

Do not delete the index folder, just start the nodes
Go to General Configuration > Content Indexing and click Rebuild

Expected Results

Index rebuild and propagation finishes without errors or warning messages. An analytics event (finished_indexing) is captured which contains index type and size.

Actual Results

Index rebuild and propagation completes, but the analytics event cannot be captured and the following warning is seen in the atlassian-confluence-index.log:

2021-09-27 20:59:15,518 WARN [lucene-interactive-reindexing-thread] [internal.index.lucene.LuceneIndexHelper] sizeOfIndexMB Cannot determine size of index located at change
 -- referer: http://10.125.90.153:8090/plugins/servlet/rebuildindex | url: /rest/prototype/latest/index/reindex | traceId: 5ad7ec4c52bda48a | userName: admin
java.nio.file.NoSuchFileException: change
	at java.base/sun.nio.fs.UnixException.translateToIOException(UnixException.java:92)
	at java.base/sun.nio.fs.UnixException.rethrowAsIOException(UnixException.java:111)
	at java.base/sun.nio.fs.UnixException.rethrowAsIOException(UnixException.java:116)
	at java.base/sun.nio.fs.UnixFileSystemProvider.newDirectoryStream(UnixFileSystemProvider.java:432)
	at java.base/java.nio.file.Files.newDirectoryStream(Files.java:472)
	at java.base/java.nio.file.Files.list(Files.java:3699)
	at com.atlassian.confluence.internal.index.lucene.LuceneIndexHelper.listRegularFilesNonRecursive(LuceneIndexHelper.java:91)
	at com.atlassian.confluence.internal.index.lucene.LuceneIndexHelper.sizeOfIndex(LuceneIndexHelper.java:37)
	at com.atlassian.confluence.internal.index.lucene.LuceneIndexHelper.sizeOfIndexMB(LuceneIndexHelper.java:52)
	at com.atlassian.confluence.impl.index.DefaultIndexRecoveryService.onReIndexFinishedEvent(DefaultIndexRecoveryService.java:436)

mentioned in: Page Failed to load; Page Failed to load

Arun Antony made changes - 05/May/2025 9:04 AM

QA Demo Status	Original: Done [ 14331 ]	New: Not Needed [ 14332 ]
QA Kickoff Status	Original: Done [ 14235 ]	New: Not Needed [ 14236 ]
Resolution		New: Fixed [ 1 ]
Status	Original: Waiting for Release [ 12075 ]	New: Closed [ 6 ]

Arun Antony made changes - 05/May/2025 9:04 AM

Fix Version/s		New: 9.2.0 [ 110293 ]
Fix Version/s		New: 8.5.18 [ 110396 ]

Arun Antony made changes - 25/Nov/2024 9:10 AM

QA Demo Status	Original: Not Done [ 14330 ]	New: Done [ 14331 ]
QA Kickoff Status	Original: Not Done [ 14234 ]	New: Done [ 14235 ]
Testing Notes		New: Did test and verified the event from Analytics page
Status	Original: In Progress [ 3 ]	New: Waiting for Release [ 12075 ]

Sumedha Bajpai made changes - 14/Nov/2024 11:32 AM

Remote Link

New: This issue links to "EXPSBAJPAI A1 Server. Master Baseline › Main unit and integration tests › issue-~~CONFSERVER-73508~~-fix-build-failure (server-syd-bamboo)" [ 964389 ]