accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Keith Turner (JIRA)" <>
Subject [jira] [Commented] (ACCUMULO-2217) fall back to reliable compression settings on failure
Date Fri, 17 Jan 2014 21:38:22 GMT


Keith Turner commented on ACCUMULO-2217:

If falling back to uncompressed, could unexpectedly fill up disk.   Gzip will not fill up
disk, but will use more CPU and take longer.   Taking a table offline seems like a bad thing
to do if one node is misconfigured, in some ways the current behavior does this (it can take
tablets offline for write).

Falling back to gzip seems like a simple thing to do .  Its better in than the current behavior.
 If needed we could make the fallback action configurable.   

There is also the possibility of the master detecting which tservers do not have snappy. 
Then tablets that use snappy would not be assigned there.  However this is a much more complex
solution that would probably just introduce more problems.

> fall back to reliable compression settings on failure
> -----------------------------------------------------
>                 Key: ACCUMULO-2217
>                 URL:
>             Project: Accumulo
>          Issue Type: Bug
>    Affects Versions: 1.4.4, 1.5.0
>            Reporter: Adam Fuchs
> Bad compression configuration keeps tables online:
> {code}
> createtable test
> config -t test -s table.file.compression.type=snappy
> flush
> droptable test
> {code}
> The flush results in an unsatisfied link error when Snappy is not properly configured.
This keeps the tablet online. Dropping the table when the tablet is stuck online puts a FATE
operation in the queue, which backs everything up.
> Can we fall back to something that is in the java code base (i.e. gz) in the event of
failures to load other compression code?

This message was sent by Atlassian JIRA

View raw message