[
https://issues.apache.org/jira/browse/FLINK-41435?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
LuChengxiang updated FLINK-41435:
---------------------------------
Description:
*Background*
Flink periodically obtains HDFS delegation tokens and distributes them through
HadoopDelegationTokenReceiver. The receiver currently adds the new credentials
to UserGroupInformation.getCurrentUser() on the refresh thread. Long-lived
Hadoop clients, such as checkpoint storage and the job-result store, capture a
UGI when their DFSClient is created and use tokens from that UGI for later RPC
connections.
If the refresh thread's current UGI is not the UGI captured by the client, the
client continues using the startup token even after a new token was created
successfully. The job later fails with SecretManager$InvalidToken at the 24h
deadline.
*Observed failure*
This was observed in a Kubernetes application-mode job. The primary failure was
in checkpoint storage:
CheckpointCoordinator.initializeCheckpointLocation
-> FsCheckpointStorageAccess.initializeLocationForCheckpoint
-> HadoopFileSystem.mkdirs
-> DFSClient.mkdirs
The checkpoint storage client reported InvalidToken for the startup token after
the refresh had created a new HDFS token and the receiver reported success. The
same JVM also showed a FileSystemJobResultStore.hasCleanJobResultEntryInternal
stack using the expired token during cleanup. The job-result store is therefore
a secondary long-lived client with the same failure mode, not the primary
trigger.
An earlier test build that skipped the second security installation (`Security
configuration is already installed`) still hit the 24h deadline; install
idempotency alone is therefore not sufficient.
FLINK-27204 (1.19+) moved FileSystemJobResultStore I/O onto the shared
cluster-io executor. DefaultDelegationTokenManager uses the same executor for
periodic token refresh. The provider and receiver are invoked sequentially on
the same refresh thread. The provider's freshUGI.doAs() scope ends before the
receiver runs, so the receiver's getCurrentUser() reflects the Subject
inherited by the cluster-io worker, not the provider's short-lived UGI.
HadoopSecurityContext is created by HadoopSecurityContextFactory from
UserGroupInformation.getLoginUser() during security installation. It stores
that particular UGI/Subject, and clients created inside
SecurityContext.runSecured() capture it. The token refresh is different:
DefaultDelegationTokenManager dispatches the refresh with
scheduledExecutor.schedule(() -> ioExecutor.execute(this::startTokensUpdate),
...). A worker in that fixed thread pool captures its AccessControlContext when
the worker is created. If the first cluster-io task is submitted from a
different Subject lifecycle, later refresh executions on that worker see that
inherited Subject through UserGroupInformation.getCurrentUser(). The provider's
freshUGI.doAs() only wraps the provider call and ends before the receiver runs;
the receiver therefore uses the worker's inherited Subject, not the
context/login UGI. The shared cluster-io executor is what makes that
worker/Subject mismatch reachable in this path; checkpoint storage and the
job-result store are both long-lived clients whose DFSClient can remain pinned
to a UGI that does not receive the refreshed credentials.
The 1.18 path had synchronous FileSystemJobResultStore I/O, so this
worker/Subject interaction was not triggered in the deployments observed. 1.15
did not have the unified delegation-token manager path. These version
differences explain the observed frequency; they do not prove that those
versions are immune.
YARN with ZooKeeper HA uses ZooKeeperLeaderElectionHaServices, which constructs
FileSystemJobResultStore.fromConfiguration(configuration, executor);
ClusterEntrypoint passes the same ioExecutor to the token manager. The
executor-sharing code path therefore exists on YARN as well. The startup order
differs: YarnApplicationClusterEntrypoint does not install the security context
before getPackagedProgram(); the only installation is in
ClusterEntrypoint.startCluster(), before runCluster() starts the cluster
services. In the normal YARN application-mode path, the first cluster-io tasks
are submitted from threads already running under the login/security context, so
the worker's inherited Subject is normally the context/login UGI. This is why a
frequent YARN failure would be surprising and why YARN is not listed as a
confirmed affected deployment type, but it is not a platform-level exemption:
the decisive condition remains the Subject inherited by the first worker.
*Root cause*
HadoopDelegationTokenReceiver targets getCurrentUser() of the refresh thread.
That target is not guaranteed to be the UGI held by HadoopSecurityContext or
the UGI captured by a long-lived checkpoint storage or job-result-store
DFSClient.
*Proposed change*
This issue addresses the case where a token is obtained successfully but the
refreshed credentials do not reach long-lived Hadoop clients, primarily
checkpoint storage.
* Add an accessor for the UGI held by HadoopSecurityContext.
* In HadoopDelegationTokenReceiver, add refreshed credentials to the
security-context UGI and login UGI, with the current UGI as a fallback,
deduplicated by object identity.
* Add a test where the refresh thread's current user differs from the
security-context user, and verify that the security-context/login UGI receives
the credentials.
Out of scope for this issue:
* The separate case where HadoopFSDelegationTokenProvider evaluates
delegation-token requirements against the wrong current UGI and does not obtain
a token at all.
* Closing FileSystem instances created under the short-lived fresh UGI; this is
separate lifecycle cleanup.
* SecurityUtils.install idempotency.
*Testing*
* A JVM probe confirmed that a fixed thread pool created under Subject A but
first submitted under Subject B runs subsequent tasks under Subject B.
* Receiver test as described above.
* A deployed application-mode job should survive the 24h token deadline and
continue checkpointing without SecretManager$InvalidToken. Verify that the
checkpoint storage DFSClient receives the new token; the job-result-store
client should receive the same credentials because it uses the same target UGI.
*Acceptance criteria*
* A token refresh updates the UGI used by checkpoint storage and other
long-lived Hadoop clients.
* Refresh does not depend on the Subject inherited by the shared cluster-io
worker.
* A long-running application-mode job survives the 24h HDFS token deadline.
was:
*Background*
Flink periodically obtains HDFS delegation tokens and distributes them through
HadoopDelegationTokenReceiver. The receiver currently adds the new credentials
to UserGroupInformation.getCurrentUser() on the refresh thread. Long-lived
Hadoop clients, such as checkpoint storage and the job-result store, capture a
UGI when their DFSClient is created and use tokens from that UGI for later RPC
connections.
If the refresh thread's current UGI is not the UGI captured by the client, the
client continues using the startup token even after a new token was created
successfully. The job later fails with SecretManager$InvalidToken at the 24h
deadline.
*Observed failure*
This was observed in a Kubernetes application-mode job. The primary failure was
in checkpoint storage:
CheckpointCoordinator.initializeCheckpointLocation
-> FsCheckpointStorageAccess.initializeLocationForCheckpoint
-> HadoopFileSystem.mkdirs
-> DFSClient.mkdirs
The checkpoint storage client reported InvalidToken for the startup token after
the refresh had created a new HDFS token and the receiver reported success. The
same JVM also showed a FileSystemJobResultStore.hasCleanJobResultEntryInternal
stack using the expired token during cleanup. The job-result store is therefore
a secondary long-lived client with the same failure mode, not the primary
trigger.
An earlier test build that skipped the second security installation (`Security
configuration is already installed`) still hit the 24h deadline; install
idempotency alone is therefore not sufficient.
FLINK-27204 (1.19+) moved FileSystemJobResultStore I/O onto the shared
cluster-io executor. DefaultDelegationTokenManager uses the same executor for
periodic token refresh. The provider and receiver are invoked sequentially on
the same refresh thread. The provider's freshUGI.doAs() scope ends before the
receiver runs, so the receiver's getCurrentUser() reflects the Subject
inherited by the cluster-io worker, not the provider's short-lived UGI.
HadoopSecurityContext is created by HadoopSecurityContextFactory from
UserGroupInformation.getLoginUser() during security installation. It stores
that particular UGI/Subject, and clients created inside
SecurityContext.runSecured() capture it. The token refresh is different:
DefaultDelegationTokenManager dispatches the refresh with
scheduledExecutor.schedule(() -> ioExecutor.execute(this::startTokensUpdate),
...). A worker in that fixed thread pool captures its AccessControlContext when
the worker is created. If the first cluster-io task is submitted from a
different Subject lifecycle, later refresh executions on that worker see that
inherited Subject through UserGroupInformation.getCurrentUser(). The provider's
freshUGI.doAs() only wraps the provider call and ends before the receiver runs;
the receiver therefore uses the worker's inherited Subject, not the
context/login UGI. FileSystemJobResultStore I/O using the same cluster-io
executor is what makes that worker/Subject mismatch reachable in this path.
The 1.18 path had synchronous FileSystemJobResultStore I/O, so this
worker/Subject interaction was not triggered in the deployments observed. 1.15
did not have the unified delegation-token manager path. These version
differences explain the observed frequency; they do not prove that those
versions are immune.
YARN with ZooKeeper HA uses ZooKeeperLeaderElectionHaServices, which constructs
FileSystemJobResultStore.fromConfiguration(configuration, executor);
ClusterEntrypoint passes the same ioExecutor to the token manager. The
executor-sharing code path therefore exists on YARN as well. The startup order
differs: YarnApplicationClusterEntrypoint does not install the security context
before getPackagedProgram(); the only installation is in
ClusterEntrypoint.startCluster(), before runCluster() starts the cluster
services. In the normal YARN application-mode path, the first cluster-io tasks
are submitted from threads already running under the login/security context, so
the worker's inherited Subject is normally the context/login UGI. This is why a
frequent YARN failure would be surprising and why YARN is not listed as a
confirmed affected deployment type, but it is not a platform-level exemption:
the decisive condition remains the Subject inherited by the first worker.
*Root cause*
HadoopDelegationTokenReceiver targets getCurrentUser() of the refresh thread.
That target is not guaranteed to be the UGI held by HadoopSecurityContext or
the UGI captured by a long-lived checkpoint storage or job-result-store
DFSClient.
*Proposed change*
This issue addresses the case where a token is obtained successfully but the
refreshed credentials do not reach long-lived Hadoop clients, primarily
checkpoint storage.
* Add an accessor for the UGI held by HadoopSecurityContext.
* In HadoopDelegationTokenReceiver, add refreshed credentials to the
security-context UGI and login UGI, with the current UGI as a fallback,
deduplicated by object identity.
* Add a test where the refresh thread's current user differs from the
security-context user, and verify that the security-context/login UGI receives
the credentials.
Out of scope for this issue:
* The separate case where HadoopFSDelegationTokenProvider evaluates
delegation-token requirements against the wrong current UGI and does not obtain
a token at all.
* Closing FileSystem instances created under the short-lived fresh UGI; this is
separate lifecycle cleanup.
* SecurityUtils.install idempotency.
*Testing*
* A JVM probe confirmed that a fixed thread pool created under Subject A but
first submitted under Subject B runs subsequent tasks under Subject B.
* Receiver test as described above.
* A deployed application-mode job should survive the 24h token deadline and
continue checkpointing without SecretManager$InvalidToken. Verify that the
checkpoint storage DFSClient receives the new token; the job-result-store
client should receive the same credentials because it uses the same target UGI.
*Acceptance criteria*
* A token refresh updates the UGI used by checkpoint storage and other
long-lived Hadoop clients.
* Refresh does not depend on the Subject inherited by the shared cluster-io
worker.
* A long-running application-mode job survives the 24h HDFS token deadline.
> Hadoop delegation token refresh can target a different UGI than Hadoop clients
> ------------------------------------------------------------------------------
>
> Key: FLINK-41435
> URL: https://issues.apache.org/jira/browse/FLINK-41435
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Configuration
> Affects Versions: 2.0.0, 1.19.0, 1.20.0, 2.1.0, 2.2.0, 2.3.0, 2.4.0
> Environment: Flink 1.19.0 and later (1.19.x, 1.20.x, 2.0 through
> 2.4), plus master (2.5-SNAPSHOT); Kubernetes and standalone application mode
> with Kerberos and HDFS delegation tokens. The mismatch is more likely when
> FileSystemJobResultStore and the token manager share the cluster-io executor
> (FLINK-27204).
> Reporter: LuChengxiang
> Priority: Major
>
> *Background*
> Flink periodically obtains HDFS delegation tokens and distributes them
> through HadoopDelegationTokenReceiver. The receiver currently adds the new
> credentials to UserGroupInformation.getCurrentUser() on the refresh thread.
> Long-lived Hadoop clients, such as checkpoint storage and the job-result
> store, capture a UGI when their DFSClient is created and use tokens from that
> UGI for later RPC connections.
> If the refresh thread's current UGI is not the UGI captured by the client,
> the client continues using the startup token even after a new token was
> created successfully. The job later fails with SecretManager$InvalidToken at
> the 24h deadline.
> *Observed failure*
> This was observed in a Kubernetes application-mode job. The primary failure
> was in checkpoint storage:
> CheckpointCoordinator.initializeCheckpointLocation
> -> FsCheckpointStorageAccess.initializeLocationForCheckpoint
> -> HadoopFileSystem.mkdirs
> -> DFSClient.mkdirs
> The checkpoint storage client reported InvalidToken for the startup token
> after the refresh had created a new HDFS token and the receiver reported
> success. The same JVM also showed a
> FileSystemJobResultStore.hasCleanJobResultEntryInternal stack using the
> expired token during cleanup. The job-result store is therefore a secondary
> long-lived client with the same failure mode, not the primary trigger.
> An earlier test build that skipped the second security installation
> (`Security configuration is already installed`) still hit the 24h deadline;
> install idempotency alone is therefore not sufficient.
> FLINK-27204 (1.19+) moved FileSystemJobResultStore I/O onto the shared
> cluster-io executor. DefaultDelegationTokenManager uses the same executor for
> periodic token refresh. The provider and receiver are invoked sequentially on
> the same refresh thread. The provider's freshUGI.doAs() scope ends before the
> receiver runs, so the receiver's getCurrentUser() reflects the Subject
> inherited by the cluster-io worker, not the provider's short-lived UGI.
> HadoopSecurityContext is created by HadoopSecurityContextFactory from
> UserGroupInformation.getLoginUser() during security installation. It stores
> that particular UGI/Subject, and clients created inside
> SecurityContext.runSecured() capture it. The token refresh is different:
> DefaultDelegationTokenManager dispatches the refresh with
> scheduledExecutor.schedule(() -> ioExecutor.execute(this::startTokensUpdate),
> ...). A worker in that fixed thread pool captures its AccessControlContext
> when the worker is created. If the first cluster-io task is submitted from a
> different Subject lifecycle, later refresh executions on that worker see that
> inherited Subject through UserGroupInformation.getCurrentUser(). The
> provider's freshUGI.doAs() only wraps the provider call and ends before the
> receiver runs; the receiver therefore uses the worker's inherited Subject,
> not the context/login UGI. The shared cluster-io executor is what makes that
> worker/Subject mismatch reachable in this path; checkpoint storage and the
> job-result store are both long-lived clients whose DFSClient can remain
> pinned to a UGI that does not receive the refreshed credentials.
> The 1.18 path had synchronous FileSystemJobResultStore I/O, so this
> worker/Subject interaction was not triggered in the deployments observed.
> 1.15 did not have the unified delegation-token manager path. These version
> differences explain the observed frequency; they do not prove that those
> versions are immune.
> YARN with ZooKeeper HA uses ZooKeeperLeaderElectionHaServices, which
> constructs FileSystemJobResultStore.fromConfiguration(configuration,
> executor); ClusterEntrypoint passes the same ioExecutor to the token manager.
> The executor-sharing code path therefore exists on YARN as well. The startup
> order differs: YarnApplicationClusterEntrypoint does not install the security
> context before getPackagedProgram(); the only installation is in
> ClusterEntrypoint.startCluster(), before runCluster() starts the cluster
> services. In the normal YARN application-mode path, the first cluster-io
> tasks are submitted from threads already running under the login/security
> context, so the worker's inherited Subject is normally the context/login UGI.
> This is why a frequent YARN failure would be surprising and why YARN is not
> listed as a confirmed affected deployment type, but it is not a
> platform-level exemption: the decisive condition remains the Subject
> inherited by the first worker.
> *Root cause*
> HadoopDelegationTokenReceiver targets getCurrentUser() of the refresh thread.
> That target is not guaranteed to be the UGI held by HadoopSecurityContext or
> the UGI captured by a long-lived checkpoint storage or job-result-store
> DFSClient.
> *Proposed change*
> This issue addresses the case where a token is obtained successfully but the
> refreshed credentials do not reach long-lived Hadoop clients, primarily
> checkpoint storage.
> * Add an accessor for the UGI held by HadoopSecurityContext.
> * In HadoopDelegationTokenReceiver, add refreshed credentials to the
> security-context UGI and login UGI, with the current UGI as a fallback,
> deduplicated by object identity.
> * Add a test where the refresh thread's current user differs from the
> security-context user, and verify that the security-context/login UGI
> receives the credentials.
> Out of scope for this issue:
> * The separate case where HadoopFSDelegationTokenProvider evaluates
> delegation-token requirements against the wrong current UGI and does not
> obtain a token at all.
> * Closing FileSystem instances created under the short-lived fresh UGI; this
> is separate lifecycle cleanup.
> * SecurityUtils.install idempotency.
> *Testing*
> * A JVM probe confirmed that a fixed thread pool created under Subject A but
> first submitted under Subject B runs subsequent tasks under Subject B.
> * Receiver test as described above.
> * A deployed application-mode job should survive the 24h token deadline and
> continue checkpointing without SecretManager$InvalidToken. Verify that the
> checkpoint storage DFSClient receives the new token; the job-result-store
> client should receive the same credentials because it uses the same target
> UGI.
> *Acceptance criteria*
> * A token refresh updates the UGI used by checkpoint storage and other
> long-lived Hadoop clients.
> * Refresh does not depend on the Subject inherited by the shared cluster-io
> worker.
> * A long-running application-mode job survives the 24h HDFS token deadline.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
