[IGNITE-13877] Error restarting the node with switching from disabled WAL archiving to enabled

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 2.11
• Component/s: persistence
• Release Note: Safe restart of node with switching from disabled to enabled WAL archiving.
• Ignite Flags: Release Notes Required
• Assignee: Kirill Tkalenko
• Reporter: Kirill Tkalenko
• Reviewer: Sergey Chugunov
• Created: 21/Dec/20 05:35
• Updated: 10/Feb/21 07:27
• Resolved: 10/Feb/21 07:22

Description

If a user starts a node with WAL archiving disabled, and then poured data there and there were more than DataStorageConfiguration#walSegments and then wants to restart a node with WAL archiving enabled, they will fail due to the following error:

SEVERE: Critical system error detected. Will be handled accordingly to configured handler [hnd=StopNodeOrHaltFailureHandler [tryStop=false, timeout=0, super=AbstractFailureHandler [ignoredFailureTypes=UnmodifiableSet [SYSTEM_WORKER_BLOCKED, SYSTEM_CRITICAL_OPERATION_TIMEOUT]]], failureCtx=FailureContext [type=CRITICAL_ERROR, err=class o.a.i.i.processors.cache.persistence.StorageException: Failed to read checkpoint record from WAL, persistence consistency cannot be guaranteed. Make sure configuration points to correct WAL folders and WAL folder is properly mounted [ptr=FileWALPointer [idx=11, fileOff=15864934, len=21409], walPath=db/wal, walArchive=db/wal/archive]]]
class org.apache.ignite.internal.processors.cache.persistence.StorageException: Failed to read checkpoint record from WAL, persistence consistency cannot be guaranteed. Make sure configuration points to correct WAL folders and WAL folder is properly mounted [ptr=FileWALPointer [idx=11, fileOff=15864934, len=21409], walPath=db/wal, walArchive=db/wal/archive]
	at org.apache.ignite.internal.processors.cache.persistence.GridCacheDatabaseSharedManager.performBinaryMemoryRestore(GridCacheDatabaseSharedManager.java:2324)
	at org.apache.ignite.internal.processors.cache.persistence.GridCacheDatabaseSharedManager.readMetastore(GridCacheDatabaseSharedManager.java:799)
	at org.apache.ignite.internal.processors.cache.persistence.GridCacheDatabaseSharedManager.notifyMetaStorageSubscribersOnReadyForRead(GridCacheDatabaseSharedManager.java:3523)
	at org.apache.ignite.internal.IgniteKernal.start(IgniteKernal.java:1206)
	at org.apache.ignite.internal.IgnitionEx$IgniteNamedInstance.start0(IgnitionEx.java:2089)
	at org.apache.ignite.internal.IgnitionEx$IgniteNamedInstance.start(IgnitionEx.java:1758)
	at org.apache.ignite.internal.IgnitionEx.start0(IgnitionEx.java:1147)
	at org.apache.ignite.internal.IgnitionEx.startConfigurations(IgnitionEx.java:1065)
	at org.apache.ignite.internal.IgnitionEx.start(IgnitionEx.java:951)
	at org.apache.ignite.internal.IgnitionEx.start(IgnitionEx.java:850)
	at org.apache.ignite.internal.IgnitionEx.start(IgnitionEx.java:720)
	at org.apache.ignite.internal.IgnitionEx.start(IgnitionEx.java:689)
	at org.apache.ignite.Ignition.start(Ignition.java:344)

At this point, the user can be offered the following workaround:
Move all segments to WAL archive directory (include consistentId directory) as they are except the last one. Last one rename as index % DataStorageConfiguration#walSegments.

Described workaround should be done automatically without user intervention.

Issue Links

• links to: GitHub Pull Request #8681