[
https://issues.apache.org/jira/browse/IOTDB-1136?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17274190#comment-17274190
]
Chao Wang commented on IOTDB-1136:
----------------------------------
1. yes, our code have some deal， which data have some missing... but do not
deal Out-of-order.
For example:
1200，1100, 1000, 2200....
this data will think it is error data, which are not suite to Regular
encoding.. if have this data, we can delete it, which is not affect other...
current encoding algorithm:
Calculates the difference between the first value and the last value, and then
the ratio of the minimum difference between adjacent values in the data...
If the minimum difference between adjacent values is negative, but the
difference between the first and last values is positive, then the problem
occurs.
To store such data with large deviations, need rewrite the algorithm, including
the storage mode in chunk. but It's against the original purpose of the
algorithm.
> Improved reliability in flush error
> -----------------------------------
>
> Key: IOTDB-1136
> URL: https://issues.apache.org/jira/browse/IOTDB-1136
> Project: Apache IoTDB
> Issue Type: Improvement
> Components: Core/Engine
> Reporter: Chao Wang
> Assignee: Chao Wang
> Priority: Critical
>
> current, when flush data error ,
> IoTDBDescriptor.getInstance().getConfig().setReadOnly(true), which will lead
> to MultiFileLogNodeManager#forceTask useless.
> But do not have a mechanism， which recover forceTask... which this will lead
> to write failed....
> so I will do this recover... Automatic fault recovery.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
