hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Anoop Sam John (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-17112) Prevent setting timestamp of delta operations being same as previous value's
Date Thu, 17 Nov 2016 05:53:59 GMT

    [ https://issues.apache.org/jira/browse/HBASE-17112?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15672846#comment-15672846

Anoop Sam John commented on HBASE-17112:

Ya that is what I also see.. We dont consider ts in Cells within Increment/Append.  So this
issue is possible iff time at RS going back.
Another rare case I can think of is this..  2 increments are happening with same TS. Cur TS
 = old cell.. So as per the logic we will put oldTS + 1..  (= now +1)..  Now say with same
TS a delete also happening for the row.  But this delete can not mask the increment put as
its ts> delete ts.  (Even if mvcc of delete op is greater)..  I know it is rare.. Just
saying..   Or am I going wrong some where?

> Prevent setting timestamp of delta operations being same as previous value's
> ----------------------------------------------------------------------------
>                 Key: HBASE-17112
>                 URL: https://issues.apache.org/jira/browse/HBASE-17112
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 1.1.7, 0.98.23, 1.2.4
>            Reporter: Phil Yang
>            Assignee: Phil Yang
>         Attachments: HBASE-17112-branch-1-v1.patch, HBASE-17112-branch-1.1-v1.patch,
HBASE-17112-v1.patch, HBASE-17112-v2.patch
> In delta operations, Increment and Append. We will read current value first and then
write the new whole result into WAL as the type of Put with current timestamp. If the previous
ts is larger than current ts, we will use the previous ts.
> If we have two Puts with same TS, we will ignore the Put with lower sequence id. It is
not friendly with versioning. And for replication we will drop sequence id  while writing
to peer cluster so in the slave we don't know what the order they are being written. If the
pushing is disordered, the result will be wrong.
> We can set the new ts to previous+1 if the previous is not less than now.

This message was sent by Atlassian JIRA

View raw message