[
https://issues.apache.org/jira/browse/BOOKKEEPER-716?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13931833#comment-13931833
]
Ivan Kelly commented on BOOKKEEPER-716:
---------------------------------------
{quote}for 1), i don't mind adding the flag. but I am not very sure if we
really need the flag. since in the past release we also bump journal version,
which were also breaking changes as you said. and more concerned that shall we
need to do each time we bump version on either disk format? {quote}
The last change was between the 4.1.0 the 4.2.0 release. The changes in 4.3.0
are riskier and there is a lot more production usage of bookkeeper, so
stability is even more important. I'm not just talking about this change, but
the other changes in the code related to checkpointing etc. I think it would be
best to make this feature (and therefore the incompatibility) optional until
the rest of the code has been proven for a few months.
That said, I think this change can be made without breaking BC. Instead of
having a [mask][len][bytes...] which is 8bytes min, you could have
[len][ledgerid][entryid] where ledgerid = -1, which is 20bytes min, but which
will be simply discarded when the scanner gets it on recovery, because the
ledger doesn't exist anymore. This would be backwards compat.
{quote}
for 2), as the batching logic is in journal side, I would prefer adding the
padding logic in journal side, rather than journalchannel.
for 3), the padding there is just for easy to track if it is a padding or a
normal record. it doesn't affect any correctness or performance issue. you
could still let reads start from 512 boundary.
{quote}
ok.
> padding writes for bookie journal
> ---------------------------------
>
> Key: BOOKKEEPER-716
> URL: https://issues.apache.org/jira/browse/BOOKKEEPER-716
> Project: Bookkeeper
> Issue Type: Sub-task
> Components: bookkeeper-server
> Reporter: Sijie Guo
> Assignee: Sijie Guo
> Fix For: 4.3.0
>
> Attachments: BOOKKEEPER-716.diff, BOOKKEEPER-716_715.diff
>
>
> it would be better to pad journal writes to align sector size, which to avoid
> second syncing corrupt an already synced sector/page.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
