Thread getting stuck for more than 10 minutes during login #578
Comments
|
Can you show the other threads as well? Without looking too deep into it, my experience in this area are broken DFS scenarios with dangling Server references. Did you try to target a server directly and turn DFS of? |
|
Hi ecki, Thanks and Regards, |
|
@SubhasishDey Any update on this? |
|
Hi @hierynomus / @ecki , I toggled the dfs setting for different server and dfs shares and then figured that the actual issue occurs when we try to connect to a non DFS server with DFS enabled in the smbj client.
Thanks and Regards, |
|
@SubhasishDey Current master contains a fix for the hanging of the |
|
Thanks @hierynomus …I will take the update. Anyways.. thanks again...I will test wit your inputs. |
|
I'm adding one additional fix, should be ready end of day today. |
|
Additional fix is also merged. Good luck testing. |
|
Hi @hierynomus The fix is working fine for us , will it be part of the official release of 0.11.2? If so when can we expect it to be released? |
|
Yes, it will be part of the next release. |
* Ignore VSCode * Add NtStatus.STATUS_UNSUCCESSFUL * Add NtStatus.STATUS_INSUFF_SERVER_RESOURCES (#611) * Add NtStatus.STATUS_IO_REPARSE_TAG_NOT_HANDLED (#514) * Update gradle build * Update release plugin * Release version: 0.11.0 * Fix signing task dependency * Release version: 0.11.1 * Use BCSecurityProvider by default for SMB3 compatibility (Fixes #638) * Ensure DFS Path Referral times out after transactTimeout (Fixes #578) * Only add DFSPathResolver if both client and server support DFS (#640) * Only add DFSPathResolver if both client and server support DFS * Fix indentation problems * Fix incorrectly reformatted javadoc * Format using java formatter * One more indentation fix * Upgrade Bouncy Castle to 1.68 to fix vulnerability report (#641) * address issue #604 - stop closing the dfs share connection immediately. (#609) * stop closing the dfs share connection immediately. * Add explanatory comment Co-authored-by: Jeroen van Erp <[email protected]> * Add support for unregistering server from serverlist (Fixes #644) (#647) * Add support for unregistering server from serverlist (Fixes #644) * Fix indentation * Reducing logging for smb3 (#650) For each smb3 packet there's an info log message which produces a tremendous amount of output. I would suggest to reduce log level to debug (or trace; similarly as in one of the other packet reciever classes). * Consolidate SMBv1 error messages * Upgrade BouncyCastle to 1.69 * Release version: 0.11.2 * Ensure artifact is signed * Release version: 0.11.3 * Fix #665: Allow JCE KDF to work (#666) * Fix #665: Allow JCE KDF to work * Add header * Add KDF unit test * Use correct maxPayloadSize for encrypted packets (Fixes #668) (#683) * Read fileId as long (#693) * Read fileId of FileIdBothDirectoryInformation into a long * Read fileId of FileIdFullDirectoryInformation into a long * File the issue that nested folder creation throw NAME EXIST error. (#685) * File the issue that nested folder creation throw NAME EXIST error. * formatting * Updated build status badges (#684) - Added GitHub badge for Build SMBJ - Corrected Codacy badge link - Removed bintray badge link - Removed Travis CI badge and configuration - Removed Java profiler link - Removed CircleCI configuration - Removed unused github-ci configuration * Use AceSize field when reading ACEs (#696) Fixes issue seen in the wild where unnecessary padding at the end of an ACE confused Smbj (but not Windows). * Ensure that enough bytes are cached from InputStream to get a correct bytesLeft count for SMB2Write (fixes #669) * GzipOutputStream integration test * Ensure that enough bytes are cached from InputStream to get a correct bytesLeft count for SMB2Write Co-authored-by: Stanislav Kardashov <[email protected]> Co-authored-by: Jeroen van Erp <[email protected]> * Add GH workflow for publishing * Update dependencies and build file * Rename test class to *Spec * Release version: 0.11.5 * Use the hostname part of the TargetHint for DFS step 9 (fixes \#419) (#722) * Slightly reduce the locking in Connection.send and DirectTcpTransport (fixes \#732) * Fixed indentation * Converting bytes written to long (Fixes #740) Signed-off-by: Jeroen van Erp <[email protected]> * Upgrading gradle to 8.0.2 Signed-off-by: Jeroen van Erp <[email protected]> * Add Implementation manifest attributes (Fixes #743) * Revert accidental comment of integration docker tasks * Do not send SMB2EncryptionCapabilities NegotiationContext is !isEncry… (#752) * Do not send SMB2EncryptionCapabilities NegotiationContext is !isEncryptionSupported (Fixes #747) * Add test for SMB2EncryptionCapabilities * Add preliminary changelog for new release * Ensure we call flip() on Buffer to avoid Java8 problems (Fixes #705) Signed-off-by: Jeroen van Erp <[email protected]> * Ensure path is set for rmdir to prevent accidents (Fixes #756) Signed-off-by: Jeroen van Erp <[email protected]> * Add support for reading / writing NIO ByteBuffers (#759) * Add support for reading / writing NIO ByteBuffers Currently one can transfer data using streams or array, but it would be great to have the opportunity to use NIO buffers. This adds two new method to the File class that accept a NIO ByteBuffer. * Implemented ByteBuffer write using ByteChunkProvider Signed-off-by: Jeroen van Erp <[email protected]> --------- Signed-off-by: Jeroen van Erp <[email protected]> Co-authored-by: Christoph Läubrich <[email protected]> Co-authored-by: Jeroen van Erp <[email protected]> * Fix some sonatype warnings * resolve conflict with master * Ignore non-semver tags for release workflow * Small warning cleanup Signed-off-by: Jeroen van Erp <[email protected]> * Setup ConnectionContext and AuthenticationContext for NTLM improvements Signed-off-by: Jeroen van Erp <[email protected]> * Refactor TargetInfo/AvPairs * Added null check and rename field * Refactor NtlmFunctions * Change hierarchy of Ntlm messages Signed-off-by: Jeroen van Erp <[email protected]> * Next step of NTLM refactor * NtlmNegotiate sends Domain/Workstation/Version fields * Filter negotiateflags and use clientTargetInfo * Rework keys in NtlmAuthenticator Signed-off-by: Jeroen van Erp <[email protected]> * Change to structure of NtlmAuthenticate Signed-off-by: Jeroen van Erp <[email protected]> * Added last changes Required to put withIntegrity = false still, due to missing mechListMIC Signed-off-by: Jeroen van Erp <[email protected]> --------- Signed-off-by: Jeroen van Erp <[email protected]> Co-authored-by: Nicholas DiPiazza <[email protected]> Co-authored-by: ndimitry <[email protected]> Co-authored-by: Patrick Boyd <[email protected]> Co-authored-by: Hannes <[email protected]> Co-authored-by: pyzhou <[email protected]> Co-authored-by: exceptionfactory <[email protected]> Co-authored-by: Chris Pacejo <[email protected]> Co-authored-by: Stanislav Kardashov <[email protected]> Co-authored-by: Stanislav Kardashov <[email protected]> Co-authored-by: Christoph Läubrich <[email protected]> Co-authored-by: Christoph Läubrich <[email protected]>
Hi,
We are finding that the SMBJ client stucks during connect for a period of 10 mins or more even if the socket timeouts are set in the SMBConfig.
We see the below thread dump, any hint from anyone will be deeply appreciated.
Id=488 in TIMED_WAITING
at sun.misc.Unsafe.park(Native Method)
- waiting on <0x08e6d37d> (a java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject)
at java.util.concurrent.locks.LockSupport.parkNanos(LockSupport.java:215)
at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await(AbstractQueuedSynchronizer.java:2163)
at com.hierynomus.protocol.commons.concurrent.Promise.tryRetrieve(Promise.java:170)
at com.hierynomus.protocol.commons.concurrent.Promise.retrieve(Promise.java:134)
at com.hierynomus.protocol.commons.concurrent.PromiseBackedFuture.get(PromiseBackedFuture.java:58)
at com.hierynomus.protocol.commons.concurrent.CancellableFuture.get(CancellableFuture.java:84)
at com.hierynomus.protocol.commons.concurrent.Futures.get(Futures.java:38)
at com.hierynomus.smbj.connection.Connection.multiProtocolNegotiate(Connection.java:340)
at com.hierynomus.smbj.connection.Connection.negotiateDialect(Connection.java:310)
at com.hierynomus.smbj.connection.Connection.connect(Connection.java:120)
at com.hierynomus.smbj.SMBClient.getEstablishedOrConnect(SMBClient.java:103)
- locked <0x04f57371> (a com.hierynomus.smbj.SMBClient)
The text was updated successfully, but these errors were encountered: