HDFS-4383. Document the lease limits. Contributed by Arshad Mohammad.

This commit is contained in:
Akira Ajisaka 2015-05-21 17:30:43 +09:00
parent fb6b38d67d
commit dc8434ab2b
2 changed files with 22 additions and 3 deletions

View File

@ -575,6 +575,8 @@ Release 2.8.0 - UNRELEASED
HDFS-4185. Add a metric for number of active leases (Rakesh R via raviprak)
HDFS-4383. Document the lease limits. (Arshad Mohammad via aajisaka)
OPTIMIZATIONS
HDFS-8026. Trace FSOutputSummer#writeChecksumChunks rather than

View File

@ -42,10 +42,27 @@ import org.apache.hadoop.util.StringUtils;
@InterfaceAudience.Private
public interface HdfsServerConstants {
int MIN_BLOCKS_FOR_WRITE = 1;
//
// Timeouts, constants
//
/**
* For a HDFS client to write to a file, a lease is granted; During the lease
* period, no other client can write to the file. The writing client can
* periodically renew the lease. When the file is closed, the lease is
* revoked. The lease duration is bound by this soft limit and a
* {@link HdfsServerConstants#LEASE_HARDLIMIT_PERIOD hard limit}. Until the
* soft limit expires, the writer has sole write access to the file. If the
* soft limit expires and the client fails to close the file or renew the
* lease, another client can preempt the lease.
*/
long LEASE_SOFTLIMIT_PERIOD = 60 * 1000;
/**
* For a HDFS client to write to a file, a lease is granted; During the lease
* period, no other client can write to the file. The writing client can
* periodically renew the lease. When the file is closed, the lease is
* revoked. The lease duration is bound by a
* {@link HdfsServerConstants#LEASE_SOFTLIMIT_PERIOD soft limit} and this hard
* limit. If after the hard limit expires and the client has failed to renew
* the lease, HDFS assumes that the client has quit and will automatically
* close the file on behalf of the writer, and recover the lease.
*/
long LEASE_HARDLIMIT_PERIOD = 60 * LEASE_SOFTLIMIT_PERIOD;
long LEASE_RECOVER_PERIOD = 10 * 1000; // in ms
// We need to limit the length and depth of a path in the filesystem.