[jira] [Created] (IGNITE-10219) MVCC: TX: Backup node update may fails after lost tx rollback.

[Anton Vinogradov (Jira)]

Andrew Mashenkov created IGNITE-10219:
-----------------------------------------

             Summary: MVCC: TX: Backup node update may fails after lost tx rollback.
                 Key: IGNITE-10219
                 URL: https://issues.apache.org/jira/browse/IGNITE-10219
             Project: Ignite
          Issue Type: Bug
          Components: mvcc
            Reporter: Andrew Mashenkov


The use case:
 # Start Tx and update entry.
 # Kill backup.
 # Rollback Tx (backup miss this tx state change due to outage)
 # Start backup.
 # Update same entry may fails with unknown tx state for latest entry version.

Backup won't rebalance partition for the key as rollback doesn't increment partition counter and can't found active transaction for latest entry version as TxLog contains neither commit nor rollback record.

Also Tx can't be detected as rolled back as mvcc coordinator version hasn't changed during backup node outage.

 

Possible solutions are
 # Increment mvcc coordinator version on every node join event that need to be carefully tested.
 # Scan cache to cleanup such entries on node startup (right after recovery from WAL) which is inefficient.
 # Or may be log Tx start in TxLog and rollback all active Tx on node startup.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)