You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am using DBMaker.memoryShardedHashMap() to create a memory map and DBMaker.fileDB() to create a disk map to be used as expireOverflow() for the memory map.
Occasionally (not every time), when data starts getting evicted from memory and put on disk, an attempt to remove entries fails because the storage is corrupted. The stack traces vary.
This seems to be the case only when expireAfterCreate() is set on the disk map, even if not setting any expiration strategy, i.e. no expireStoreSize(), etc. There is no expireExecutor(), hence expiration is synchronous. Would appreciate some insight on the same:
java.lang.ArrayIndexOutOfBoundsException: Index 4 out of bounds for length 4
at org.mapdb.DataInput2$ByteArray.unpackInt(DataInput2.java:162)
at org.mapdb.DataInput2.readUTF(DataInput2.java:491)
at org.mapdb.serializer.SerializerString.deserialize(SerializerString.java:20)
at org.mapdb.serializer.SerializerString.deserialize(SerializerString.java:11)
at org.mapdb.HTreeMap$leafValueExternalSerializer$1.deserialize(HTreeMap.kt:228)
at org.mapdb.HTreeMap$leafValueExternalSerializer$1.deserialize(HTreeMap.kt:214)
at org.mapdb.Serializer.deserializeFromLong(Serializer.java:621)
at org.mapdb.StoreDirect.get(StoreDirect.kt:543)
at org.mapdb.HTreeMap.expireEvictEntry(HTreeMap.kt:903)
at org.mapdb.HTreeMap$expireEvictSegment$1.take(HTreeMap.kt:881)
at org.mapdb.QueueLong.takeUntil(QueueLong.kt:153)
at org.mapdb.HTreeMap.expireEvictSegment(HTreeMap.kt:859)
at org.mapdb.HTreeMap.remove(HTreeMap.kt:462)
org.mapdb.DBException$GetVoid: Record does not exist, recid=29
at org.mapdb.StoreDirect.get(StoreDirect.kt:533)
at org.mapdb.HTreeMap.expireEvictEntry(HTreeMap.kt:903)
at org.mapdb.HTreeMap$expireEvictSegment$1.take(HTreeMap.kt:881)
at org.mapdb.QueueLong.takeUntil(QueueLong.kt:153)
at org.mapdb.HTreeMap.expireEvictSegment(HTreeMap.kt:859)
at org.mapdb.HTreeMap.remove(HTreeMap.kt:462)
org.mapdb.DBException$DataCorruption: nodeRecid not found in this leaf
at org.mapdb.HTreeMap.expireEvictEntry(HTreeMap.kt:921)
at org.mapdb.HTreeMap$expireEvictSegment$1.take(HTreeMap.kt:881)
at org.mapdb.QueueLong.takeUntil(QueueLong.kt:153)
at org.mapdb.HTreeMap.expireEvictSegment(HTreeMap.kt:859)
at org.mapdb.HTreeMap.remove(HTreeMap.kt:462)
kotlin.KotlinNullPointerException: null
at org.mapdb.QueueLong.remove(QueueLong.kt:165)
at org.mapdb.HTreeMap.removeprotected(HTreeMap.kt:493)
at org.mapdb.HTreeMap.remove(HTreeMap.kt:464)
at org.mapdb.DB$HashMapMaker$verify$listener$1.modify(DB.kt:736)
at org.mapdb.HTreeMap.listenerNotify(HTreeMap.kt:1189)
at org.mapdb.HTreeMap.removeprotected(HTreeMap.kt:513)
at org.mapdb.HTreeMap.remove(HTreeMap.kt:464)
org.mapdb.DBException$PointerChecksumBroken: Broken bit parity
at org.mapdb.DataIO.parity1Get(DataIO.java:440)
at org.mapdb.DataIO.unpackRecid(DataIO.java:172)
at org.mapdb.serializer.SerializerRecid.deserialize(SerializerRecid.java:22)
at org.mapdb.serializer.SerializerRecid.deserialize(SerializerRecid.java:13)
at org.mapdb.QueueLong$Node$SERIALIZER.deserialize(QueueLong.kt:50)
at org.mapdb.QueueLong$Node$SERIALIZER.deserialize(QueueLong.kt:39)
at org.mapdb.StoreDirectAbstract.deserialize(StoreDirectAbstract.kt:231)
at org.mapdb.StoreDirect.get(StoreDirect.kt:547)
at org.mapdb.QueueLong.remove(QueueLong.kt:170)
at org.mapdb.HTreeMap.removeprotected(HTreeMap.kt:493)
at org.mapdb.HTreeMap.remove(HTreeMap.kt:464)
The text was updated successfully, but these errors were encountered:
I am using
DBMaker.memoryShardedHashMap()
to create a memory map andDBMaker.fileDB()
to create a disk map to be used asexpireOverflow()
for the memory map.Occasionally (not every time), when data starts getting evicted from memory and put on disk, an attempt to remove entries fails because the storage is corrupted. The stack traces vary.
This seems to be the case only when
expireAfterCreate()
is set on the disk map, even if not setting any expiration strategy, i.e. noexpireStoreSize()
, etc. There is noexpireExecutor()
, hence expiration is synchronous. Would appreciate some insight on the same:The text was updated successfully, but these errors were encountered: