[HADOOP-15609] Retry KMS calls when SSLHandshakeException occurs

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 3.1.0
• Fix Version/s: 3.2.0, 3.0.4, 3.1.2
• Component/s: common, kms
• Target Version/s: 3.2.0
• Hadoop Flags: Reviewed
• Assignee: Kitti Nanasi
• Reporter: Kitti Nanasi
• Created: 16/Jul/18 11:19
• Updated: 02/Oct/19 17:15
• Resolved: 25/Jul/18 04:47

Description

KMS call should retry when javax.net.ssl.SSLHandshakeException occurs and FailoverOnNetworkExceptionRetry policy is used.

For example in the following stack trace, we can see that the KMS Provider's connection is lost, an SSLHandshakeException is thrown and the operation is not retried:

W0711 18:19:50.213472  1508 LoadBalancingKMSClientProvider.java:132] KMS provider at [https://example.com:16000/kms/v1/] threw an IOException:
Java exception follows:
javax.net.ssl.SSLHandshakeException: Remote host closed connection during handshake
        at sun.security.ssl.SSLSocketImpl.readRecord(SSLSocketImpl.java:1002)
        at sun.security.ssl.SSLSocketImpl.performInitialHandshake(SSLSocketImpl.java:1385)
        at sun.security.ssl.SSLSocketImpl.startHandshake(SSLSocketImpl.java:1413)
        at sun.security.ssl.SSLSocketImpl.startHandshake(SSLSocketImpl.java:1397)
        at sun.net.www.protocol.https.HttpsClient.afterConnect(HttpsClient.java:559)
        at sun.net.www.protocol.https.AbstractDelegateHttpsURLConnection.connect(AbstractDelegateHttpsURLConnection.java:185)
        at sun.net.www.protocol.http.HttpURLConnection.getOutputStream0(HttpURLConnection.java:1316)
        at sun.net.www.protocol.http.HttpURLConnection.getOutputStream(HttpURLConnection.java:1291)
        at sun.net.www.protocol.https.HttpsURLConnectionImpl.getOutputStream(HttpsURLConnectionImpl.java:250)
        at org.apache.hadoop.crypto.key.kms.KMSClientProvider.call(KMSClientProvider.java:512)
        at org.apache.hadoop.crypto.key.kms.KMSClientProvider.call(KMSClientProvider.java:502)
        at org.apache.hadoop.crypto.key.kms.KMSClientProvider.decryptEncryptedKey(KMSClientProvider.java:791)
        at org.apache.hadoop.crypto.key.kms.LoadBalancingKMSClientProvider$5.call(LoadBalancingKMSClientProvider.java:288)
        at org.apache.hadoop.crypto.key.kms.LoadBalancingKMSClientProvider$5.call(LoadBalancingKMSClientProvider.java:284)
        at org.apache.hadoop.crypto.key.kms.LoadBalancingKMSClientProvider.doOp(LoadBalancingKMSClientProvider.java:124)
        at org.apache.hadoop.crypto.key.kms.LoadBalancingKMSClientProvider.decryptEncryptedKey(LoadBalancingKMSClientProvider.java:284)
        at org.apache.hadoop.crypto.key.KeyProviderCryptoExtension.decryptEncryptedKey(KeyProviderCryptoExtension.java:532)
        at org.apache.hadoop.hdfs.DFSClient.decryptEncryptedDataEncryptionKey(DFSClient.java:927)
        at org.apache.hadoop.hdfs.DFSClient.createWrappedInputStream(DFSClient.java:946)
        at org.apache.hadoop.hdfs.DistributedFileSystem$4.doCall(DistributedFileSystem.java:316)
        at org.apache.hadoop.hdfs.DistributedFileSystem$4.doCall(DistributedFileSystem.java:311)
        at org.apache.hadoop.fs.FileSystemLinkResolver.resolve(FileSystemLinkResolver.java:81)
        at org.apache.hadoop.hdfs.DistributedFileSystem.open(DistributedFileSystem.java:323)
Caused by: java.io.EOFException: SSL peer shut down incorrectly
        at sun.security.ssl.InputRecord.read(InputRecord.java:505)
        at sun.security.ssl.SSLSocketImpl.readRecord(SSLSocketImpl.java:983)
        ... 22 more
W0711 18:19:50.239328  1508 LoadBalancingKMSClientProvider.java:149] Aborting since the Request has failed with all KMS providers(depending on hadoop.security.kms.client.failover.max.retries=1 setting and numProviders=1) in the group OR the exception is not recoverable