hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Andrew Wang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-11402) HDFS Snapshots should capture point-in-time copies of OPEN files
Date Fri, 21 Apr 2017 02:33:04 GMT

    [ https://issues.apache.org/jira/browse/HDFS-11402?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15977967#comment-15977967

Andrew Wang commented on HDFS-11402:

Hi Manoj, took another look, mostly just code cleanup-type comments. Would also be interested
to know how you chose 512 inodes/thread as it seems like a magic number, was this based on

bq. the byte[][] is already available getINodesAndPaths() along with INode[]. Wouldn't it
be a lot slower to have one more helper method to construct the same byte[][] ?

It seems equivalent if we have two functions as follows:

  private static INode[] getINodes(final INode inode) {
    int depth = 0, index;
    INode tmp = inode;
    while (tmp != null) {
      tmp = tmp.getParent();
    INode[] inodes = new INode[depth];
    tmp = inode;
    index = depth;
    while (tmp != null) {
      inodes[index] = tmp;
      tmp = tmp.getParent();
    return inodes;

  private static byte[][] getPaths(final INode[] inodes) {
    byte paths[][] = new byte[inodes.length][];
    for (int i=0; i<inodes.length; i++) {
      paths[i] = inodes[i].getKey();
    return paths;

Was this earlier comment addressed?

bq. isDescendant, if we're an IIP, can we simply look in our array of INodes for the specified
ancestor? This method looks expensive right now.

hasReadLock checks hasWriteLock, so the additional hasWriteLock check is unnecessary:

    assert fsnamesystem.hasReadLock() || fsnamesystem.hasWriteLock();

* We could use a stride increment to simplify the work partitioning logic (and make work distribution
more even):

for (int idx = workerIdx; idx < inodeCount; idx += workerCount) {
  // do work

We aren't using workerIdx when getting the future results, so can simplify with foreach:

    for (Future<List<INodesInPath>> f: futureList) {
      try {
      } catch (Exception e) {
        LOG.warn("INode filter task encountered exception: ", e);

Mind filing a JIRA to migrate LeaseManager over to SLF4J as well? We're adding a pretty gnarly
new log that would be more readable with SLF4j.

> HDFS Snapshots should capture point-in-time copies of OPEN files
> ----------------------------------------------------------------
>                 Key: HDFS-11402
>                 URL: https://issues.apache.org/jira/browse/HDFS-11402
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: hdfs
>    Affects Versions: 2.6.0
>            Reporter: Manoj Govindassamy
>            Assignee: Manoj Govindassamy
>         Attachments: HDFS-11402.01.patch, HDFS-11402.02.patch, HDFS-11402.03.patch, HDFS-11402.04.patch,
HDFS-11402.05.patch, HDFS-11402.06.patch
> *Problem:*
> 1. When there are files being written and when HDFS Snapshots are taken in parallel,
Snapshots do capture all these files, but these being written files in Snapshots do not have
the point-in-time file length captured. That is, these open files are not frozen in HDFS Snapshots.
These open files grow/shrink in length, just like the original file, even after the snapshot
> 2. At the time of File close or any other meta data modification operation on these files,
HDFS reconciles the file length and records the modification in the last taken Snapshot. All
the previously taken Snapshots continue to have those open Files with no modification recorded.
So, all those previous snapshots end up using the final modification record in the last snapshot.
Thus after the file close, file lengths in all those snapshots will end up same.
> Assume File1 is opened for write and a total of 1MB written to it. While the writes are
happening, snapshots are taken in parallel.
> {noformat}
> |---Time---T1-----------T2-------------T3----------------T4------>
> |-----------------------Snap1----------Snap2-------------Snap3--->
> |---File1.open---write---------write-----------close------------->
> {noformat}
> Then at time,
> T2:
> Snap1.File1.length = 0
> T3:
> Snap1.File1.length = 0
> Snap2.File1.length = 0
> <File1 write completed and closed>
> T4:
> Snap1.File1.length = 1MB
> Snap2.File1.length = 1MB
> Snap3.File1.length = 1MB
> *Proposal*
> 1. At the time of taking Snapshot, {{SnapshotManager#createSnapshot}} can optionally
request {{DirectorySnapshottableFeature#addSnapshot}} to freeze open files. 
> 2. {{DirectorySnapshottableFeature#addSnapshot}} can consult with {{LeaseManager}} and
get a list INodesInPath for all open files under the snapshot dir. 
> 3. {{DirectorySnapshottableFeature#addSnapshot}} after the Snapshot creation, Diff creation
and updating modification time, can invoke {{INodeFile#recordModification}} for each of the
open files. This way, the Snapshot just taken will have a {{FileDiff}} with {{fileSize}} captured
for each of the open files. 
> 4. Above model follows the current Snapshot and Diff protocols and doesn't introduce
any any disk formats. So, I don't think we will be needing any new FSImage Loader/Saver changes
for Snapshots.
> 5. One of the design goals of HDFS Snapshot was ability to take any number of snapshots
in O(1) time. LeaseManager though has all the open files with leases in-memory map, an iteration
is still needed to prune the needed open files and then run recordModification on each of
them. So, it will not be a strict O(1) with the above proposal. But, its going be a marginal
increase only as the new order will be of O(open_files_under_snap_dir). In order to avoid
HDFS Snapshots change in behavior for open files and avoid change in time complexity, this
improvement can be made under a new config {{"dfs.namenode.snapshot.freeze.openfiles"}} which
by default can be {{false}}.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org

View raw message