/// <summary>
/// Set up the distributed cache by localizing the resources, and updating
/// the configuration with references to the localized resources.
/// </summary>
/// <param name="conf"/>
/// <exception cref="System.IO.IOException"/>
public virtual void Setup(JobConf conf)
{
FilePath workDir = new FilePath(Runtime.GetProperty("user.dir"));
// Generate YARN local resources objects corresponding to the distributed
// cache configuration
IDictionary <string, LocalResource> localResources = new LinkedHashMap <string, LocalResource
>();
MRApps.SetupDistributedCache(conf, localResources);
// Generating unique numbers for FSDownload.
AtomicLong uniqueNumberGenerator = new AtomicLong(Runtime.CurrentTimeMillis());
// Find which resources are to be put on the local classpath
IDictionary <string, Path> classpaths = new Dictionary <string, Path>();
Path[] archiveClassPaths = DistributedCache.GetArchiveClassPaths(conf);
if (archiveClassPaths != null)
{
foreach (Path p in archiveClassPaths)
{
classpaths[p.ToUri().GetPath().ToString()] = p;
}
}
Path[] fileClassPaths = DistributedCache.GetFileClassPaths(conf);
if (fileClassPaths != null)
{
foreach (Path p in fileClassPaths)
{
classpaths[p.ToUri().GetPath().ToString()] = p;
}
}
// Localize the resources
LocalDirAllocator localDirAllocator = new LocalDirAllocator(MRConfig.LocalDir);
FileContext localFSFileContext = FileContext.GetLocalFSFileContext();
UserGroupInformation ugi = UserGroupInformation.GetCurrentUser();
ExecutorService exec = null;
try
{
ThreadFactory tf = new ThreadFactoryBuilder().SetNameFormat("LocalDistributedCacheManager Downloader #%d"
).Build();
exec = Executors.NewCachedThreadPool(tf);
Path destPath = localDirAllocator.GetLocalPathForWrite(".", conf);
IDictionary <LocalResource, Future <Path> > resourcesToPaths = Maps.NewHashMap();
foreach (LocalResource resource in localResources.Values)
{
Callable <Path> download = new FSDownload(localFSFileContext, ugi, conf, new Path(
destPath, System.Convert.ToString(uniqueNumberGenerator.IncrementAndGet())), resource
);
Future <Path> future = exec.Submit(download);
resourcesToPaths[resource] = future;
}
foreach (KeyValuePair <string, LocalResource> entry in localResources)
{
LocalResource resource_1 = entry.Value;
Path path;
try
{
path = resourcesToPaths[resource_1].Get();
}
catch (Exception e)
{
throw new IOException(e);
}
catch (ExecutionException e)
{
throw new IOException(e);
}
string pathString = path.ToUri().ToString();
string link = entry.Key;
string target = new FilePath(path.ToUri()).GetPath();
Symlink(workDir, target, link);
if (resource_1.GetType() == LocalResourceType.Archive)
{
localArchives.AddItem(pathString);
}
else
{
if (resource_1.GetType() == LocalResourceType.File)
{
localFiles.AddItem(pathString);
}
else
{
if (resource_1.GetType() == LocalResourceType.Pattern)
{
//PATTERN is not currently used in local mode
throw new ArgumentException("Resource type PATTERN is not " + "implemented yet. "
+ resource_1.GetResource());
}
}
}
Path resourcePath;
try
{
resourcePath = ConverterUtils.GetPathFromYarnURL(resource_1.GetResource());
}
catch (URISyntaxException e)
{
throw new IOException(e);
}
Log.Info(string.Format("Localized %s as %s", resourcePath, path));
string cp = resourcePath.ToUri().GetPath();
if (classpaths.Keys.Contains(cp))
{
localClasspaths.AddItem(path.ToUri().GetPath().ToString());
}
}
}
finally
{
if (exec != null)
{
exec.Shutdown();
}
}
// Update the configuration object with localized data.
if (!localArchives.IsEmpty())
{
conf.Set(MRJobConfig.CacheLocalarchives, StringUtils.ArrayToString(Sharpen.Collections.ToArray
(localArchives, new string[localArchives.Count])));
}
if (!localFiles.IsEmpty())
{
conf.Set(MRJobConfig.CacheLocalfiles, StringUtils.ArrayToString(Sharpen.Collections.ToArray
(localFiles, new string[localArchives.Count])));
}
setupCalled = true;
}
internal virtual bool FileIsPublic(Path remotePath, FileSystem remoteFs, FileStatus
status)
{
return(FSDownload.IsPublic(remoteFs, remotePath, status, null));
}
