/// <summary>
/// Sweeps and processes the shared cache area to clean up stale and orphaned
/// files.
/// </summary>
internal virtual void Process()
{
// mark the beginning of the run in the metrics
metrics.ReportCleaningStart();
try
{
// now traverse individual directories and process them
// the directory structure is specified by the nested level parameter
// (e.g. 9/c/d/<checksum>)
string pattern = SharedCacheUtil.GetCacheEntryGlobPattern(nestedLevel);
FileStatus[] resources = fs.GlobStatus(new Path(root, pattern));
int numResources = resources == null ? 0 : resources.Length;
Log.Info("Processing " + numResources + " resources in the shared cache");
long beginMs = Runtime.CurrentTimeMillis();
if (resources != null)
{
foreach (FileStatus resource in resources)
{
// check for interruption so it can abort in a timely manner in case
// of shutdown
if (Sharpen.Thread.CurrentThread().IsInterrupted())
{
Log.Warn("The cleaner task was interrupted. Aborting.");
break;
}
if (resource.IsDirectory())
{
ProcessSingleResource(resource);
}
else
{
Log.Warn("Invalid file at path " + resource.GetPath().ToString() + " when a directory was expected"
);
}
// add sleep time between cleaning each directory if it is non-zero
if (sleepTime > 0)
{
Sharpen.Thread.Sleep(sleepTime);
}
}
}
long endMs = Runtime.CurrentTimeMillis();
long durationMs = endMs - beginMs;
Log.Info("Processed " + numResources + " resource(s) in " + durationMs + " ms.");
}
catch (IOException e1)
{
Log.Error("Unable to complete the cleaner task", e1);
}
catch (Exception)
{
Sharpen.Thread.CurrentThread().Interrupt();
}
}
internal virtual IDictionary <string, string> GetInitialCachedResources(FileSystem
fs, Configuration conf)
{
// get the root directory for the shared cache
string location = conf.Get(YarnConfiguration.SharedCacheRoot, YarnConfiguration.DefaultSharedCacheRoot
);
Path root = new Path(location);
if (!fs.Exists(root))
{
string message = "The shared cache root directory " + location + " was not found";
Log.Error(message);
throw new IOException(message);
}
int nestedLevel = SharedCacheUtil.GetCacheDepth(conf);
// now traverse individual directories and process them
// the directory structure is specified by the nested level parameter
// (e.g. 9/c/d/<checksum>/file)
string pattern = SharedCacheUtil.GetCacheEntryGlobPattern(nestedLevel + 1);
Log.Info("Querying for all individual cached resource files");
FileStatus[] entries = fs.GlobStatus(new Path(root, pattern));
int numEntries = entries == null ? 0 : entries.Length;
Log.Info("Found " + numEntries + " files: processing for one resource per " + "key"
);
IDictionary <string, string> initialCachedEntries = new Dictionary <string, string>
();
if (entries != null)
{
foreach (FileStatus entry in entries)
{
Path file = entry.GetPath();
string fileName = file.GetName();
if (entry.IsFile())
{
// get the parent to get the checksum
Path parent = file.GetParent();
if (parent != null)
{
// the name of the immediate parent directory is the checksum
string key = parent.GetName();
// make sure we insert only one file per checksum whichever comes
// first
if (initialCachedEntries.Contains(key))
{
Log.Warn("Key " + key + " is already mapped to file " + initialCachedEntries[key]
+ "; file " + fileName + " will not be added");
}
else
{
initialCachedEntries[key] = fileName;
}
}
}
}
}
Log.Info("A total of " + initialCachedEntries.Count + " files are now mapped");
return(initialCachedEntries);
}
