Versions Compared

Old Version 1

changes.mady.by.user Nicholas Telford

Saved on

compared with

New Version 2

changes.mady.by.user Nicholas Telford

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Update JIRA

...

Discussion thread: here [Change the link from the KIP proposal email archive to your own email thread]

JIRA:here [Change the link from KAFKA-1 to your own ticket] 

Jira
serverASF JIRA
columnIdsissuekey,summary,issuetype,created,updated,duedate,assignee,reporter,priority,status,resolution
columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
serverId5aa69414-a9e9-3523-82ec-879b028fb15b
keyKAFKA-14412

Please keep the discussion on the mailing list rather than commenting on the wiki (wiki discussions get unwieldy fast).

...

When determining the partition assignment, StreamsPartitionsAssignor considers the checkpoint offsets on-disk for all StateStores on each instance, even for Tasks they have not (yet) been assigned. This is done via TaskManager#getTaskOffsetSums(), which directly reads from the per-Task .checkpoint file.

With our new API, this file will not contain the offsets for stores that manage changelog offsets themselves. So we need to modify the TaskManager to read not from the.checkpoint file, but from each StateStore by calling getCommittedOffset. For this reason, getCommittedOffset is required to provide the committed offset even if the store has not been initialized via init. StateStore  implementations MUST ensure they can provide offsets when the store is closed/not yet initialized without incurring a significant penalty (e.g. by having to open then close the store).

...