Emil Kleszcz created ZOOKEEPER-4334:
---------------------------------------
Summary: SASL authentication fails when using host aliases
Key: ZOOKEEPER-4334
URL: https://issues.apache.org/jira/browse/ZOOKEEPER-4334
Project: ZooKeeper
Issue Type: Bug
Affects Versions: 3.6.1
Reporter: Emil Kleszcz
I faced an issue while trying to use alternative alises with Zookeeper quorum
when SASL is enabled. The errors I get in zookeeper log are the following:
```
2021-07-12 21:04:46,437 [myid:3] - WARN
[NIOWorkerThread-3:ZooKeeperServer@1661] - Client /<IP addr>:37368 failed to
SASL authenticate: {}
javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException:
Failure unspecified at GSS-API level (Mechanism level: Checksum failed)]
at
com.sun.security.sasl.gsskerb.GssKrb5Server.evaluateResponse(GssKrb5Server.java:199)
at
org.apache.zookeeper.server.ZooKeeperSaslServer.evaluateResponse(ZooKeeperSaslServer.java:49)
at
org.apache.zookeeper.server.ZooKeeperServer.processSasl(ZooKeeperServer.java:1650)
at
org.apache.zookeeper.server.ZooKeeperServer.processPacket(ZooKeeperServer.java:1599)
at
org.apache.zookeeper.server.NIOServerCnxn.readRequest(NIOServerCnxn.java:379)
at
org.apache.zookeeper.server.NIOServerCnxn.readPayload(NIOServerCnxn.java:182)
at org.apache.zookeeper.server.NIOServerCnxn.doIO(NIOServerCnxn.java:339)
at
org.apache.zookeeper.server.NIOServerCnxnFactory$IOWorkRequest.doWork(NIOServerCnxnFactory.java:522)
at
org.apache.zookeeper.server.WorkerService$ScheduledWorkRequest.run(WorkerService.java:154)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Caused by: GSSException: Failure unspecified at GSS-API level (Mechanism level:
Checksum failed)
at sun.security.jgss.krb5.Krb5Context.acceptSecContext(Krb5Context.java:856)
at sun.security.jgss.GSSContextImpl.acceptSecContext(GSSContextImpl.java:342)
at sun.security.jgss.GSSContextImpl.acceptSecContext(GSSContextImpl.java:285)
at
com.sun.security.sasl.gsskerb.GssKrb5Server.evaluateResponse(GssKrb5Server.java:167)
... 11 more
Caused by: KrbException: Checksum failed
at
sun.security.krb5.internal.crypto.Aes256CtsHmacSha1EType.decrypt(Aes256CtsHmacSha1EType.java:102)
at
sun.security.krb5.internal.crypto.Aes256CtsHmacSha1EType.decrypt(Aes256CtsHmacSha1EType.java:94)
at sun.security.krb5.EncryptedData.decrypt(EncryptedData.java:175)
at sun.security.krb5.KrbApReq.authenticate(KrbApReq.java:281)
at sun.security.krb5.KrbApReq.<init>(KrbApReq.java:149)
at
sun.security.jgss.krb5.InitSecContextToken.<init>(InitSecContextToken.java:108)
at sun.security.jgss.krb5.Krb5Context.acceptSecContext(Krb5Context.java:829)
... 14 more
Caused by: java.security.GeneralSecurityException: Checksum failed
at
sun.security.krb5.internal.crypto.dk.AesDkCrypto.decryptCTS(AesDkCrypto.java:451)
at
sun.security.krb5.internal.crypto.dk.AesDkCrypto.decrypt(AesDkCrypto.java:272)
at sun.security.krb5.internal.crypto.Aes256.decrypt(Aes256.java:76)
at
sun.security.krb5.internal.crypto.Aes256CtsHmacSha1EType.decrypt(Aes256CtsHmacSha1EType.java:100)
... 20 more
```
What did I do?
1) created host aliases for each quorum node (a,b,c): zk1, zk2, zk3
2) Changed in zoo.cfg:
changed from
server.1=a
server.2=b
server.3=c
to:
server.1=zk1
server.2=zk2
server.3=zk3
(at this stage after restarting the ensemble all works as expected.
3) Generate new keytab with alias-based principals and host-based principals in
zookeeper.keytab
4) Change jaas.conf (server) definition from:
Server {
com.sun.security.auth.module.Krb5LoginModule required
useKeyTab=true
keyTab="/etc/zookeeper/conf/zookeeper.keytab"
storeKey=true
useTicketCache=false
principal="zookeeper/a.com@COM";
};
to
Server {
com.sun.security.auth.module.Krb5LoginModule required
useKeyTab=true
keyTab="/etc/zookeeper/conf/zookeeper.keytab"
storeKey=true
useTicketCache=false
principal="zookeeper/zk1.com@COM";
};
>From that moment, after restarting quorum members, I get the above error.
Now, why do I do this?
To allow other services such as zkfc,hbase,hdfs,yarn to connect to the quorum
using aliases. Interestingly, without changing the zookeeper principal, hbase
works perfectly, but the other 3 services fail with:
```
<2021-07-12T20:45:19.491+0200> <INFO> <org.apache.zookeeper.ZooKeeper>:
<Initiating client connection,
connectString=zk01.com:2181,zk02.com:2181,zk03.com:2181 sessionTimeout=10000
watcher=org.apache.hadoop.ha.ActiveStandbyElector$WatcherWithClientRef@3246fb96>
<2021-07-12T20:45:19.519+0200> <INFO> <org.apache.zookeeper.Login>: <Client
successfully logged in.>
<2021-07-12T20:45:19.521+0200> <INFO> <org.apache.zookeeper.Login>: <TGT
refresh thread started.>
<2021-07-12T20:45:19.524+0200> <INFO> <org.apache.zookeeper.Login>: <TGT valid
starting at: Mon Jul 12 20:45:19 CEST 2021>
<2021-07-12T20:45:19.524+0200> <INFO> <org.apache.zookeeper.Login>: <TGT
expires: Tue Jul 13 21:45:19 CEST 2021>
<2021-07-12T20:45:19.524+0200> <INFO> <org.apache.zookeeper.Login>: <TGT
refresh sleeping until: Tue Jul 13 17:05:16 CEST 2021>
<2021-07-12T20:45:19.524+0200> <INFO>
<org.apache.zookeeper.client.ZooKeeperSaslClient>: <Client will use GSSAPI as
SASL mechanism.>
<2021-07-12T20:45:19.530+0200> <INFO> <org.apache.zookeeper.ClientCnxn>:
<Opening socket connection to server zk02.com/<ip addr>:2181. Will attempt to
SASL-authenticate using Login Context section 'Client'>
<2021-07-12T20:45:19.535+0200> <INFO> <org.apache.zookeeper.ClientCnxn>:
<Socket connection established to zk02.com/<ip addr>:2181, initiating session>
<2021-07-12T20:45:19.543+0200> <INFO> <org.apache.zookeeper.ClientCnxn>:
<Session establishment complete on server zk02.com/<ip addr>:2181, sessionid =
0x200247870fb0007, negotiated timeout = 10000>
<2021-07-12T20:45:19.561+0200> <ERROR>
<org.apache.zookeeper.client.ZooKeeperSaslClient>: <SASL authentication failed
using login context 'Client' with exception: {}>
javax.security.sasl.SaslException: Error in authenticating with a Zookeeper
Quorum member: the quorum member's saslToken is null.
at
org.apache.zookeeper.client.ZooKeeperSaslClient.createSaslToken(ZooKeeperSaslClient.java:279)
at
org.apache.zookeeper.client.ZooKeeperSaslClient.respondToServer(ZooKeeperSaslClient.java:242)
at org.apache.zookeeper.ClientCnxn$SendThread.readResponse(ClientCnxn.java:805)
at org.apache.zookeeper.ClientCnxnSocketNIO.doIO(ClientCnxnSocketNIO.java:94)
at
org.apache.zookeeper.ClientCnxnSocketNIO.doTransport(ClientCnxnSocketNIO.java:366)
at org.apache.zookeeper.ClientCnxn$SendThread.run(ClientCnxn.java:1145)
<2021-07-12T20:45:19.564+0200> <INFO> <org.apache.zookeeper.ClientCnxn>:
<Unable to read additional data from server sessionid 0x200247870fb0007, likely
server has closed socket, closing socket connection and attempting reconnect>
<2021-07-12T20:45:19.671+0200> <INFO>
<org.apache.hadoop.ha.ActiveStandbyElector>: <Session connected.>
<2021-07-12T20:45:19.672+0200> <ERROR>
<org.apache.hadoop.hdfs.tools.DFSZKFailoverController>:
<DFSZKFailOverController exiting due to earlier exception java.io.IOException:
Couldn't determine existence of znode
```
When I change the principle of zookeeper hbase starts failing with this error
and other services except for the zookeeper itself is somehow working fine.
After that, I cannot connect manually to the zk quorum using zkCli and
zookeeper-client with all possible combinations of principals.
I wonder if that may have something to do with the "Server
environment:host.name=" pointing to the canonical name (and not the alias)
during the startup.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
