hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Andrew Purtell (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-14054) Acknowledged writes may get lost if regionserver clock is set backwards
Date Mon, 13 Jul 2015 21:34:05 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14054?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14625392#comment-14625392

Andrew Purtell commented on HBASE-14054:

I also read the HLC paper with interest and agree it is one technique that could be explored
for avoiding some time skew issues. One nice thing about it is its timestamps can stand in
for the system clock without any changes like HBASE-8927 (although that's not an argument
not to do that, perhaps before or after, anyway, they are orthogonal).

Here is the HLC paper: http://www.cse.buffalo.edu/tech-reports/2014-04.pdf and a companion
blog post: http://muratbuffalo.blogspot.com/2014/07/hybrid-logical-clocks.html

> Acknowledged writes may get lost if regionserver clock is set backwards
> -----------------------------------------------------------------------
>                 Key: HBASE-14054
>                 URL: https://issues.apache.org/jira/browse/HBASE-14054
>             Project: HBase
>          Issue Type: Bug
>          Components: regionserver
>    Affects Versions: 0.98.6
>         Environment: Linux
>            Reporter: Tobi Vollebregt
> We experience a small amount of lost acknowledged writes in production on July 1st (~700
identified so far).
> What happened was that we had NTP turned off since June 29th to prevent issues due to
the leap second on June 30th. NTP was turned back on July 1st.
> The next day, we noticed we were missing writes to a few of our higher throughput aggregation
> We found that this is caused by HBase taking the current time using System.currentTimeMillis,
which may be set backwards by NTP, and using this without any checks to populate the timestamp
of rows for which the client didn't supply a timestamp.
> Our application uses a read-modify-write pattern using get+checkAndPut to perform aggregation
as follows:
> 1. read version 1
> 2. mutate
> 3. write version 2
> 4. read version 2
> 5. mutate
> 6. write version 3
> The application retries the full read-modify-write if the checkAndPut fails.
> What must have happened on July 1st, after we started NTP back up, was this (timestamps
> 1. read version 1 (timestamp 10)
> 2. mutate
> 3. write version 2 (HBase-assigned timestamp 11)
> 4. read version 2 (timestamp 11)
> 5. mutate
> 6. write version 3 (HBase-assigned timestamp 10)
> Hence, the last write was eclipsed by the first write, and hence, an acknowledged write
was lost.
> While this seems to match documented behavior (paraphrasing: "if timestamp is not specified
HBase will assign a timestamp using System.currentTimeMillis" "the row with the highest timestamp
will be returned by get"), I think it is very unintuitive and needs at least a big warning
in the documentation, along the lines of "Acknowledged writes may not be visible unless the
timestamp is explicitly specified and equal to or larger than the highest timestamp for that
> I would also like to use this ticket to start a discussion on if we can make the behavior
> Could HBase assign a timestamp of {{max(max timestamp for the row, System.currentTimeMillis())}}
in the checkAndPut write path, instead of blindly taking {{System.currentTimeMillis()}}, similar
to what has been done in HBASE-12449 for increment and append?
> Thoughts?

This message was sent by Atlassian JIRA

View raw message