[
https://issues.apache.org/jira/browse/BOOKKEEPER-246?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13401203#comment-13401203
]
Rakesh R commented on BOOKKEEPER-246:
-------------------------------------
Hi Ivan, patch looks great. I have few suggestions and thoughts. Also, I think
need to rebase the Auditor BOOKKEEPER-272 patch accordingly.
- Can we have factory for the LedgerUnderreplicationManager?
- We are creating underreplica znode like "%s/urL%08d"
urL00000009
But LedgerManager is creating znode like "%010d"
L_0000000009
I feel would be good to follow same pattern for maintainability.
(Presently there is no functional problem as replication logic is separate.)
- In the getLedgerToRereplicate(), changedLatch.countDown() should be only on
EventType.NodeChildrenChanged like as follows, just to minimize the chance of
unwanted exceptions.
{code}
public void process(WatchedEvent e) {
if (event.getType() == EventType.NodeChildrenChanged)
changedLatch.countDown();
}
{code}
- It would be good to have few info or debug logs for debuggability. I have
seen many places we are silently continuing without logs
markLedgerUnderreplicated
markLedgerComplete
getLedgerToRereplicate
releaseLedger
- Also, one good pattern of handling the InterruptedException.
When you catch and swallow InterruptedException, you should call
Thread.currentThread().interrupt() afterward, so that the interrupt status
isn't lost
{code}
} catch (InterruptedException ie) {
Thread.currentThread().interrupt()
// logging or throw exception back
}
{code}
- getLedgerToRereplicate:
{code}
List<String> children = zkc.getChildren(urLedgerPath, w);
zkc.getChildren(urLockPath, w);
Collections.shuffle(children);
while (children.size() > 0) {
{code}
Instead can we filter the children which doesn't have any lock, like:
{code}
List<String> children = zkc.getChildren(urLedgerPath, w);
List<String> lockedChildren = zkc.getChildren(urLockPath, w);
Collection<String> nolockChildren = CollectionUtils.subtract(children,
lockedChildren);
while (nolockChildren.size() > 0) {
Collections.shuffle(nolockChildren);
{code}
-Rakesh
> Recording of underreplication of ledger entries
> -----------------------------------------------
>
> Key: BOOKKEEPER-246
> URL: https://issues.apache.org/jira/browse/BOOKKEEPER-246
> Project: Bookkeeper
> Issue Type: Sub-task
> Components: bookkeeper-client, bookkeeper-server
> Reporter: Ivan Kelly
> Assignee: Ivan Kelly
> Fix For: 4.2.0
>
> Attachments: BOOKKEEPER-246.diff, BOOKKEEPER-246.diff
>
>
> This JIRA is to decide how to record that entries in a ledger are
> underreplicated.
> I think there is a common understanding (correct me if im wrong), that
> rereplication can be broken into two logically distinct phases. A) Detection
> of entry underreplication & B) Rereplication.
> This subtask is to handle the interaction between these two stages. Stage B
> needs to know what to rereplicate; how should Stage A inform it?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
