You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Francis Liu (JIRA)" <ji...@apache.org> on 2018/03/12 23:50:00 UTC

[jira] [Updated] (HBASE-19502) Make sure we have closed all StoreFileScanners if we fail to open any StoreFileScanners

     [ https://issues.apache.org/jira/browse/HBASE-19502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Francis Liu updated HBASE-19502:
--------------------------------
    Fix Version/s:     (was: 1.3.3)
                   1.3.2

> Make sure we have closed all StoreFileScanners if we fail to open any StoreFileScanners
> ---------------------------------------------------------------------------------------
>
>                 Key: HBASE-19502
>                 URL: https://issues.apache.org/jira/browse/HBASE-19502
>             Project: HBase
>          Issue Type: Sub-task
>          Components: regionserver, Scanners
>            Reporter: Chia-Ping Tsai
>            Assignee: Chia-Ping Tsai
>            Priority: Major
>             Fix For: 2.0.0, 1.3.2, 1.4.1, 1.2.7
>
>         Attachments: HBASE-19502.branch-1.4.patch, HBASE-19502.v0.patch
>
>
> {code:title=StoreFileScanner.java}
>   public static List<StoreFileScanner> getScannersForStoreFiles(Collection<StoreFile> files,
>       boolean cacheBlocks, boolean usePread, boolean isCompaction, boolean canUseDrop,
>       ScanQueryMatcher matcher, long readPt, boolean isPrimaryReplica) throws IOException {
>     List<StoreFileScanner> scanners = new ArrayList<StoreFileScanner>(files.size());
>     List<StoreFile> sorted_files = new ArrayList<>(files);
>     Collections.sort(sorted_files, StoreFile.Comparators.SEQ_ID);
>     for (int i = 0; i < sorted_files.size(); i++) {
>       StoreFile.Reader r = sorted_files.get(i).createReader(canUseDrop);
>       r.setReplicaStoreFile(isPrimaryReplica);
>       StoreFileScanner scanner = r.getStoreFileScanner(cacheBlocks, usePread, isCompaction, readPt,
>         i, matcher != null ? !matcher.hasNullColumnInQuery() : false);
>       scanners.add(scanner);
>     }
>     return scanners;
>   }
> {code}
> The missed decrement of ref count will obstruct the cleanup of compacted files.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)