hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Wei-Chiu Chuang (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HDFS-9908) Datanode should tolerate disk scan failure during NN handshake
Date Thu, 24 Mar 2016 16:16:25 GMT

     [ https://issues.apache.org/jira/browse/HDFS-9908?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Wei-Chiu Chuang updated HDFS-9908:
----------------------------------
    Attachment: HDFS-9908.004.patch

Thanks for Eddy's comments:
Attached rev04 to address Eddy's comments.

{quote}
I think that not all IOE}}s are DU related? Throwing a {{DiskUsageException here might be
confused.
{quote}
What about if I change the exception name to AddBlockPoolException? It sounds more accurate
to me.
{quote}
About handleDiskUsageError(), what if there are IOE that are not from DU? Should it throw
these exceptions?
{quote}
This is to keep the existing behavior: pop up exceptions to callers, whether or not they come
from DU. The idea is  only process the exception when it happens to be during handshake, and
to reduce the side effect to other callers of volume scanner. I am not sure if people prefer
to keep the it as is, or do something differently.

{quote}
If an IOE is thrown on this volume, is the metadata of the blocks on this volume still in
memory? If so, can you add some comments.
{quote}
I'm not quite sure about this question. Do you mean the metadata of the blocks at NameNode?
I think the answer is yes, because this is unrelated to NameNode. But a du input/output error
can mean any files in the volume, and this includes the block, finalized, rbw and maybe others
as well.

{quote}
If possible, it'd be better to not use static member for tests. If there is anything happened
before you reset the flag, other tests will mistakenly see this flag as enabled.
{quote}
I've updated the code to pass the test info in a Configuration object. Thanks a lot for the
advice!

> Datanode should tolerate disk scan failure during NN handshake
> --------------------------------------------------------------
>
>                 Key: HDFS-9908
>                 URL: https://issues.apache.org/jira/browse/HDFS-9908
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: datanode
>    Affects Versions: 2.5.0
>         Environment: CDH5.3.3
>            Reporter: Wei-Chiu Chuang
>            Assignee: Wei-Chiu Chuang
>         Attachments: HDFS-9908.001.patch, HDFS-9908.002.patch, HDFS-9908.003.patch, HDFS-9908.004.patch
>
>
> DN may treat a disk scan failure exception as an NN handshake exception, and this can
prevent a DN to join a cluster even if most of its disks are healthy.
> During NN handshake, DN initializes block pools. It will create a lock files per disk,
and then scan the volumes. However, if the scanning throws exceptions due to disk failure,
DN will think it's an exception because NN is inconsistent with the local storage (see {{DataNode#initBlockPool}}.
As a result, it will attempt to reconnect to NN again.
> However, at this point, DN has not deleted its lock files on the disks. If it reconnects
to NN again, it will think the same disks are already being used, and then it will fail handshake
again because all disks can not be used (due to locking), and repeatedly. This will happen
even if the DN has multiple disks, and only one of them fails. The DN will not be able to
connect to NN despite just one failing disk. Note that it is possible to successfully create
a lock file on a disk, and then has error scanning the disk.
> We saw this on a CDH 5.3.3 cluster (which is based on Apache Hadoop 2.5.0, and we still
see the same bug in 3.0.0 trunk branch). The root cause is that DN treats an internal error
(single disk failure) as an external one (NN handshake failure) and we should fix it.
> {code:title=DataNode.java}
> /**
>    * One of the Block Pools has successfully connected to its NN.
>    * This initializes the local storage for that block pool,
>    * checks consistency of the NN's cluster ID, etc.
>    * 
>    * If this is the first block pool to register, this also initializes
>    * the datanode-scoped storage.
>    * 
>    * @param bpos Block pool offer service
>    * @throws IOException if the NN is inconsistent with the local storage.
>    */
>   void initBlockPool(BPOfferService bpos) throws IOException {
>     NamespaceInfo nsInfo = bpos.getNamespaceInfo();
>     if (nsInfo == null) {
>       throw new IOException("NamespaceInfo not found: Block pool " + bpos
>           + " should have retrieved namespace info before initBlockPool.");
>     }
>     
>     setClusterId(nsInfo.clusterID, nsInfo.getBlockPoolID());
>     // Register the new block pool with the BP manager.
>     blockPoolManager.addBlockPool(bpos);
>     
>     // In the case that this is the first block pool to connect, initialize
>     // the dataset, block scanners, etc.
>     initStorage(nsInfo);
>     // Exclude failed disks before initializing the block pools to avoid startup
>     // failures.
>     checkDiskError();
>     data.addBlockPool(nsInfo.getBlockPoolID(), conf);  <----- this line throws disk
error exception
>     blockScanner.enableBlockPoolId(bpos.getBlockPoolId());
>     initDirectoryScanner(conf);
>   }
> {code}
> {{FsVolumeList#addBlockPool}} is the source of exception.
> {code:title=FsVolumeList.java}
>   void addBlockPool(final String bpid, final Configuration conf) throws IOException {
>     long totalStartTime = Time.monotonicNow();
>     
>     final List<IOException> exceptions = Collections.synchronizedList(
>         new ArrayList<IOException>());
>     List<Thread> blockPoolAddingThreads = new ArrayList<Thread>();
>     for (final FsVolumeImpl v : volumes) {
>       Thread t = new Thread() {
>         public void run() {
>           try (FsVolumeReference ref = v.obtainReference()) {
>             FsDatasetImpl.LOG.info("Scanning block pool " + bpid +
>                 " on volume " + v + "...");
>             long startTime = Time.monotonicNow();
>             v.addBlockPool(bpid, conf);
>             long timeTaken = Time.monotonicNow() - startTime;
>             FsDatasetImpl.LOG.info("Time taken to scan block pool " + bpid +
>                 " on " + v + ": " + timeTaken + "ms");
>           } catch (ClosedChannelException e) {
>             // ignore.
>           } catch (IOException ioe) {
>             FsDatasetImpl.LOG.info("Caught exception while scanning " + v +
>                 ". Will throw later.", ioe);
>             exceptions.add(ioe);
>           }
>         }
>       };
>       blockPoolAddingThreads.add(t);
>       t.start();
>     }
>     for (Thread t : blockPoolAddingThreads) {
>       try {
>         t.join();
>       } catch (InterruptedException ie) {
>         throw new IOException(ie);
>       }
>     }
>     if (!exceptions.isEmpty()) {
>       throw exceptions.get(0); <----- here's the original of exception
>     }
>     
>     long totalTimeTaken = Time.monotonicNow() - totalStartTime;
>     FsDatasetImpl.LOG.info("Total time to scan all replicas for block pool " +
>         bpid + ": " + totalTimeTaken + "ms");
>   }
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message