Release Notes: IBM Aspera Enterprise Server, Connect Server, Point-to-Point Client, and Desktop Client 3.6.2

 

Product Release: July 5, 2016
Release Notes Updated: September 11, 2017

WHAT'S NEW (Windows Version)

For Linux features, see WHAT'S NEW (Linux Version) below.

Release 3.6.2 is an update of IBM Aspera Enterprise Server, Connect Server, Point-to-Point Client, and Desktop Client. It provides the new features, fixes, and other changes listed below that have been made since release 3.6.0. Release 3.6.2 includes IBM Aspera Sync 3.6.2.

New Features

  • The command to resume ascp with the command-line argument k2 or k3 now works when an FIPS-140 compatible cipher is set.
  • Added support for tokens for parent paths in single path transfers. For example, a token built for /general will also allow a transfer of a single file or directory named/general/other/subsidiary.
  • ascp now supports multiple private SSH key files that are called on the command line, specified with -i arguments. The keys are tried in order and the process ends when a key passes authentication or when all keys have been tried and have failed authentication.
  • Added support for encryption-at-rest (EAR) in the watchfolder configuration.
  • Console now reports the activity of the server node on the Activity Overview Page, in the Sync Jobs list, when async_activity_logging is set in aspera.conf.
  • aswatchadmin now allows specification of the escape character when using --format.
  • Improved performance (by increasing the speed) when the system is deleting a large number of files in Azure.
  • Node API /files/browse now uses a configurable aspera.conf option to retrieve the preserved modification times for ascp uploads.
  • In Node API, the parameters iteration_token and after_time can now both be used with the APIs /ops/transfers and /events; the parameters can be used together or separately. The parameters are set in aspera.conf according to the following policy:
    • If no value is available for count, use the value set in max_response_entries.
    • If count is present, return the lower number of responses of count and max_response_entries.
  • Support for cache-control metadata for S3.
  • The default read/write block size is now set to 256 K (262144 KB) in ascp.
  • ascp forward proxy needs to support DNS names pointing to multiple proxies.
  • asperawatchfolderd support for aspera proxy (dnat/dnats).
  • A new option in aspera.conf for client-side encryption setting frees users from having to specify a matching value at the command line (-T).
  • async now accepts external cookies, tags, and reports through management.
  • Added a client-side configuration option to define a per-user or global default of the SSH key (to help clients used to running ssh without specifying a key if $HOME/.ssh/id_rsa exists).
  • Added support for the hmac-sha2-256 and hmac-sha2-512 algorithms.
  • Added support for the diffie-hellman-group-exchange-sha256 key-exchange algorithm.
  • Support for full URIs in file lists and file-pair lists.
  • Amazon's new Infrequent Access storage option is now available as a storage class, through docroot configuration.
  • Node API /transfers and /opt/transfers take a multi_session field to split a given transfer job between multiple ascp sessions, possibly over multiple nodes when using a cluster as an initiator. Functionality: this adds a default multipart aspera.conf configuration directive to use when one is not specified by the API call.
  • Support for the multi-session threshold feature in node (file.json) transfers.
  • Support for Amazon KMS when uploading to S3 (pushes data directly to S3 and supports KMS encryption).
  • ascp now reads session_timeout_sec from aspera.conf.
  • Azure support now includes storage to blob pages.
  • Transfers to Amazon S3 storage now have the capacity to set cache-control request headers. For more information, see the API Reference in the AWS documentation at the following URL (subject to change):
    http://docs.aws.amazon.com/AmazonS3/latest/API/RESTObjectPUT.html
  • The option ascp --remove-empty-source-directory for the --remove-after-transfer feature can now be enabled in Aspera Central in the SOAP API; it removes the source directory itself.
  • Support for an inclusion filter with Console smart transfers.
    Note: This feature is not available in Console 3.0.6 (the current release).
  • In async, there is support for periodic scan on remote.
  • A new procedure removes the --master option from asperawatchd and introduces a procedure to elect a leader among the running daemons. This helps to overcome the situation where the master instance crashes or loses the connection to the other instances.
  • asperawatchd now supports domains (namespaces) in redis that distinguish the different asperawatchd instances running on the same redis instance, which enables asperawatchd to start multiple asperawatchd instances using different users.

Breaking Changes

  • When you export and import connections in the GUI, they do not include passwords. Because the password database encryption uses a per-user/per-machine cryptographic key, the encrypted passwords cannot be used on other machines. Since the passwords are not transferred, they must be re-entered when the connection is imported.
    Note: This is an intentional change in behavior that was made for security reasons.
  • Parallel transfers started with ascp ascp.exe on any version of Enterprise Server (ES), and transferring to ES (version 3.6.0 and above) with the -C option, fail with the error Session Stop (Error: Session initiation failed).
  • [Watchfolder] Upgrading to Enterprise Server 3.6.2 deletes all watchfolders. Workaround: Before upgrading, locate the watchfolder identifier generated in ES 3.6.0. Do this by opening the watchfolderd logs and locating a line similar to the following:
    2016-06-23 20:05:18 [b6b-3caaa700] LOG [Redis_communicator::Worker] Run(): Start listening for incoming messages on address [t=CHANNEL,n='wfd:b7520a90-86f4-4485-9178-f746c9ef98c5:input']
    In the example above, b7520a90-86f4-4485-9178-f746c9ef98c5 is the generated identifier. Save the identifier somewhere to access it later.
    After upgrading to 3.6.2, run the following to reload your preexisting watchfolder configuration:
    asperawatchfolderd --identifier watchfolder_id
    For example:
    asperawatchfolderd --identifier b7520a90-86f4-4485-9178-f746c9ef98c5

Other Changes

  • [Windows] After upgrading from the currently installed 3.6.2 version of Enterprise Server, Connect Server, or Point-to-Point to a more recent build version of 3.6.2, the admin user may be prompted to restart the system.
  • This release offers a new set of aswatchadmin options reorganized by function, which can be used instead of the --tool option. The first command below is replaced by the second:
    Old
    aswatchadmin --db-spec redis:localhost:31415 --tool=daemon --create /mnt
    New
    aswatchadmin create-watch /mnt --db-spec redis:locahost:31415
  • The aspera.conf options for asperawatchd snapshot creation have changed.

Issues Fixed in this Release

(This release contains tickets created from two tracking systems; thus there are two different formats for ticket identifiers.)

#36187 - [Windows, ES and P2P] An upgrade from 3.6.0 to 3.6.2 results in a problem with hotfolders; a failed to authenticate message appears.

#WAT-230 - [watchfolderd] Post-processing is not performed for failed drops to a watchfolder: when a drop fails because it has exceeded its configured retry period (nothing transfers for xseconds), the post-processing for the drop is not executed and the corresponding files do not reach a final state. The logs incorrectly show the file state TRANSFER_FINISHED.

#WAT-223 - When creating a watchfolder with post-processing transfer_delete, the source files are not removed after they are successfully transferred

#WAT-190 - Watchfolder does not retransfer changed files because it does not detect a change that happens after the cool-off of the file but before the completion of the drop.

#WAT-170 - async stops mid-transfer when using asperawatchd.

#WAT-168 - Local post-processing "transfer_delete" is executed for failed files.

#WAT-136 - On a watchfolder, timestamps are not preserved, even when they are specified in the configuration.

#WAT-119 - async noncontinuous mode does not sync files after a cancelled session.

#WAT-92 - In asperawatchfolderd, there is now an ability to reenable remove post-processing.

#WAT-86 - aswatchadmin now has a feature to test and set system values.

#WAT-64 - asperawatchd now supports namespaces in redis to distinguish the different asperawatchd instances running on the same redis instance; this enables starting multiple asperawatchd instances using different users.

#WAT-51 - Watchfolder status is stuck in Processing when files are skipped by ascp.

#36088, #36087, #36085, #36084, #36083, #36082, #36073, #36069 - [Windows] Users in a translated UI (Spanish, French, Chinese, Japanese) see error messages when they change Server Configuration settings.

#35893 - When a docroot is configured on a remote server and a new hot folder is configured with a source folder equal to (with any existing local folder as the destination folder) the underlying ascp transfer fails because the asperasync services removes trailing slashes from directory paths, resulting in an empty source path.

#35883 - [Windows] For an Enterprise Server that has inline validation configured and is running on Windows, with domain users (DOMAIN\USER) for authentication, the payload that the Aspera Server sends to the validator does not have valid JSON syntax.

#35378 - asperahttpd crashes when resuming an existing ascp transfer.

#35295 - A destination top-level directory does not have write access and the user is using the -t option to preserve times in their async.

#35253 - Alee logs show that alee reports 400 bad request when the redis database is empty or all the keys have 0.

#35002 - When the user attempts to transfer a file containing the character '&' in the name while using the AWS signature V4 (AWS4-HMAC-SHA256), the transfer fails. Users who use aws:kmsserver-side encryption will experience the issue since using KMS implies the use of signature V4. To get trapd to self -configure to use the signature V4, do one of the following:
  • Use the KMS server side encryption (sse:kms).
  • Use a server in a region where this is mandatory - like Frankfurt (eu-central-1).

#34419 - The user is unable to download encrypted files with an incorrect decryption passphrase. (This is stilll an issue when using HTTP fallback; see ticket #34811 in the Known Issues section).

#34228 - The error ERR cannot read content-protect parameters should display as a warning instead.

#34166 - Memory leaks in asperanoded result in system crashes.

#34139 - Delete-before-transfer is not working for a SMB\CIFS path.

#34124 - When transfers are initiated from Watchfolder Transfers, the Name field does not populate with data.

#33978 - Ability to check for sshkey path in $HOME/.ssh not functioning properly.

#33975 - (Point-to-Point) A Console transfer with SSH keys is not working after an upgrade to Point-to-Point 3.6.0. The issue occurs when Console is set up to have smart transfers where the following is true:
  • The non-root user has a docroot set on aspera.conf.
  • The remote endpoint is associated with an SSH private key.
  • The remote host is a unmanaged node.

#33874 - [Windows] During installation of Enterprise Server 3.6.0, the UI does not start and displays the error message Exception occured during initialization. This in an issue on Windows Server 2008 with a non-unicode locale set to several non-US countries.

#33567 - When server-side encryption-at-rest (EAR) is configured for local storage usingfile:///<path> docroot, the aspera.conf chunk_size parameter must be set.

#33566 - Installation of the Connect Server web UI on Windows 10 fails.

#33536 - When the database is configured on a remote machine, and db_host is set, ascpattempts to connect to the local database.

#33313 - When using CLI to transfer and preserve XATTRs, XATTRs are not preserved with a file reporting as 0 KB.

#33233 - Basic tokens are stored and can be reused to make arbitrary transfers.

#33204 - A file name "\" causes a crash.

#33011 - asdelete fails to connect to the server.

WHAT'S NEW (Linux Version)

Release 3.6.2 is an update of IBM Aspera Enterprise Server, Connect Server, Point-to-Point Client, and Desktop Client. It provides the new features, fixes, and other changes listed below for Linux since release 3.6.1. Release 3.6.2 includes IBM Aspera Sync 3.6.2.

New Features

  • The command to resume ascp with the command-line argument k2 or k3 now works when an FIPS-140 compatible cipher is set.
  • Added support for tokens for parent paths in single path transfers. For example, a token built for /general will also allow a transfer of a single file or directory named /general/other/subsidiary.
  • ascp now supports multiple private SSH key files that are called on the command line, specified with -i arguments. The keys are tried in order and the process ends when a key passes authentication or when all keys have been tried and have failed authentication.
  • Added support for encryption-at-rest (EAR) in the watchfolder configuration.

Breaking Changes

  • When you export and import connections in the GUI, they do not include passwords. Because the password database encryption uses a per-user/per-machine cryptographic key, the encrypted passwords cannot be used on other machines. Since the passwords are not transferred, they must be re-entered when the connection is imported.
    Note: This is an intentional change in behavior that was made for security reasons.
  • Parallel transfers from the desktop client to Enterprise Server (ES 3.6.0 and above) with the -C option fail with the error Session Stop (Error: Session initiation failed).
  • [Watchfolder] Upgrading to Enterprise Server 3.6.2 deletes all watchfolders. Workaround: Before upgrading, locate the watchfolder identifier generated in ES 3.6.0. Do this by opening the watchfolderd logs and locating a line similar to the following:
    2016-06-23 20:05:18 [b6b-3caaa700] LOG [Redis_communicator::Worker] Run(): Start listening for incoming messages on address [t=CHANNEL,n='wfd:b7520a90-86f4-4485-9178-f746c9ef98c5:input']
    In the example above, b7520a90-86f4-4485-9178-f746c9ef98c5 is the generated identifier. Save the identifier somewhere to access it later.
    After upgrading to 3.6.2, run the following to reload your preexisting watchfolder configuration:
    asperawatchfolderd --identifier watchfolder_id
    For example:
    asperawatchfolderd --identifier b7520a90-86f4-4485-9178-f746c9ef98c5

Issues Fixed in this Release

(This release contains tickets created from two tracking systems; thus there are two different formats for ticket identifiers.)

#WAT-230 - [watchfolderd] Post-processing is not performed for failed drops to a watchfolder: when a drop fails because it has exceeded its configured retry period (nothing transfers for xseconds), the post-processing for the drop is not executed and the corresponding files do not reach a final state. The logs incorrectly show the file state TRANSFER_FINISHED.

#WAT-223 - When creating a watchfolder with post-processing transfer_delete, the source files are not removed after they are successfully transferred

#36052 - [Linux] For Point-to-Point and Client, the passwords set in Global Preferences (under the Mail and Proxy tabs) are not saved properly. After the software is shut down and then reopened, the admin user may find that the password fields in Global Preferences are empty and the features requiring those passwords no longer work. Thus, the passwords need to be recreated in the UI or in aspera.conf each time.
Note: Because of this Known Issue in 3.6.1, admin users upgrading from 3.6.1 to 3.6.2 may need to recreate these passwords after the upgrade is complete.

#35893 - When a docroot is configured on a remote server and a new hot folder is configured with a source folder equal to (with any existing local folder as the destination folder) the underlying ascp transfer fails because the asperasync services removes trailing slashes from directory paths, resulting in an empty source path.

#33872 - [Linux] After installing the license, an incorrect error message is displayed: The server has a client license.

SYSTEM REQUIREMENTS

Windows:  2008r2, 2012r2, 7, 8, 10
Browsers: Microsoft Edge, Firefox 27-48, Chrome 32-50

Linux 64-bit and 32-bit: Ubuntu 12-15, RHEL 6-7, CentOS 6-7, SLES 11-12 Kernel 2.4 or higher and libc version GLIB 2.3.4+, Debian 6-7, Fedora 16-20 
Browsers (for Connect Server clients): Firefox 27-48, Google Chrome 32-50

PACKAGE INFORMATION

Enterprise/Connect Server

Linux 32-bit (deb): aspera-entsrv-3.6.2.127538-linux-32.deb
md5: 3bab47412e6484f3a9e434a22c3ac3b6
sha1: f68afa6b25d938b1f09571443fdb70806c6e64d6
Linux 32-bit (rpm): aspera-entsrv-3.6.2.127538-linux-32.rpm
md5: 6d00d5f4b31ca42adc6e3b2e3929c8d6
sha1: 09c376fc709f6c1b87e2ea5cf57aa18567d6cfd3
Linux 64-bit (deb): aspera-entsrv-3.6.2.127538-linux-64.deb
md5: 3410a375766c62a3592fb96bfb65547e
sha1: aab22e33993b619ff6d25b1ef7325b61bd79b5fe
Linux 64-bit (rpm): aspera-entsrv-3.6.2.127538-linux-64.rpm
md5: 02d908a0f0cc22bfa9cf38eb7ed9e892
sha1: 2cd36462fed1ac504623b6012ecc9e1d9c5a454d
Windows: AsperaEnterpriseServer-ML-3.6.2.130227-win-v100-32.exe
md5: e3812ecc1f3350de033f14f958f20f18
sha1: ca821cc4f49db3511f191e9cec14f09dadb30910

Point-to-Point

Linux 32-bit (deb): aspera-scp-p2p-3.6.2.127538-linux-32.deb
md5: d008ec0889d31810de5ce68e0bbeb849
sha1: cdc53c71cc42fda0b45a0002f80086ee0a762773
Linux 32-bit (rpm): aspera-scp-p2p-3.6.2.127538-linux-32.rpm
md5: 494d8f53c0132367def57fcb0e1d625e
sha1: 8d5373d05040dc434a03cd1944d1d94cf072d9fb
Linux 64-bit (deb): aspera-scp-p2p-3.6.2.127538-linux-64.deb
md5: 04dec663f3e349ef9a25881efd41b725
sha1: e7803d915cd363799383423f5d7fdee6f37506e2
Linux 64-bit (rpm): aspera-scp-p2p-3.6.2.127538-linux-64.rpm
md5: 871e43a91ee7da38687f3ea83ccd0535
sha1: 50cbf1a93ecd90984833ee6e41791cbb32187fe5
Windows: AsperaP2P-ML-3.6.2.130227-win-v100-32.exe
md5: 90f8a2a1190b449e21c2e7ba42c9cef0
sha1: 3867ef8d977ff51153f559d576438c3720b3cd07

Client

Linux 32-bit (deb): aspera-scp-client-3.6.2.127538-linux-32.deb
md5: cb9653866bd435db6df2e190b5ca1539
sha1: 0bf265925078831847412e42273b59c9c411516e
Linux 32-bit (rpm): aspera-scp-client-3.6.2.127538-linux-32.rpm
md5: d213afd106a7228e1e956d5847048a80
sha1: 36df0199ad067c25e62de4c638323dfc00731d56
Linux 64-bit (deb): aspera-scp-client-3.6.2.127538-linux-64.deb
md5: 53b0c259b00a12b950ebdf7e506af9de
sha1: 6eca4c347c670034273c355a61d73bfbba04a1f0
Linux 64-bit (rpm): aspera-scp-client-3.6.2.127538-linux-64.rpm
md5: b7e581441570bec6cdb774626eb99ebc
sha1: 1dfc4e961be8cd6027fa29b827ac0fe8ab346536
Windows: AsperaClient-ML-3.6.2.130227-win-v100-32.exe
md5: 115d2370dd1a1564e8d27482490ab48b
sha1: f0df9fddbac55d8f21bcf79a737e15fe5e273a23

KNOWN ISSUES (Windows and Linux)

(This release contains tickets created from two tracking systems; thus there are two different formats for ticket identifiers.)

ATT-174 - ascp fails to download and decrypt files that were uploaded and encrypted using HTTP fallback.

#36373 - In cases of a failed transfer, the files counter in session statistics is not updated with the data for skipped files.

#36276 - When the <ignore_skipped_files> setting is enabled (true) on the ES node, Console continues to show all skipped files for failed transfers on the Activity page.

#WAT-174 - Watchfolder uses excessive memory when watching 10 million files.

#WAT-169 - [Watchfolder] When using top_level_dirs drop detection with x top level directories, 7(x)+ drops are created. The drop count continually increases.

#WAT-158 - [Watchfolder] Upgrading to Enterprise Server 3.6.1 deletes all watchfolders. Workaround: Before upgrading, locate the watchfolder identifier generated in ES 3.6.0. Do this by opening the watchfolderd logs and locating a line similar to the following:
2016-06-23 20:05:18 [b6b-3caaa700] LOG [Redis_communicator::Worker] Run(): Start listening for incoming messages on address [t=CHANNEL,n='wfd:b7520a90-86f4-4485-9178-f746c9ef98c5:input']
In the example above, b7520a90-86f4-4485-9178-f746c9ef98c5 is the generated identifier. Save the identifier somewhere to access it later.
After upgrading to ES 3.6.1, run the following to reload your preexisting watchfolder configuration:
asperawatchfolderd --identifier watchfolder_id
For example:
asperawatchfolderd --identifier b7520a90-86f4-4485-9178-f746c9ef98c5

#35952 - asunprotect cannot decrypt a re-protected file.

#35857 - On CentOS7, the UI doesn't start unless there is data in the clipboard.

#35453 - Parallel transfers from the desktop client to Enterprise Server (ES 3.6.0 and above) with the -C option fail with the error Session Stop (Error: Session initiation failed).

#35421 - Shares console becomes unresponsive during a deletion command that takes a long time to run.

#35255 - When Japanese characters that include the voiced sound mark are used to name a directory in Connect Server, the following issues may occur:
  • The upload fails without an error message.
  • The download fails with the error code 43: The specified path cannot be found.

#34811 - The user is unable to download encrypted files with an incorrect decryption passphrase when using HTTP fallback.

#34674 - When Japanese language is set on the ES UI, it doesn't respect aspera.conf settings; all docroot settings are set to false, and the other settings fail with attached Japanese errors.

#34322 - ascp fails to authenticate SHH with a large banner file size (approximately 2000 bytes) on CentOS 7.2.

#33877 - Aspera Enterprise licenses that are watchfolder-enabled don't display watchfolders as an enabled setting when running the license switch ascp -A.

#33673 - Transferring symbolic links on Windows results in the creation of a zero-byte file at the destination instead of the expected copying of the target of the symlink.

#33374 - The unimplemented function error does not appear when the user attempts to create a symbolic link to a file on cloud storage (S3) from the Node API. (Symbolic link capability is only available on local storage.)

#33298 - The ES Windows installer fails to install with the /quiet option.

#33229 - The client is no longer able to browse a file on cloud storage using a /files/browseAPI request.

#33214 - Transfers to and from cloud storage (S3)—using authorization tokens with URIs that do not have a docroot specified—are unsupported.

#33206 - /ops/transfers erroneously shows some queued transfers (which are farther down in the queue) as failed before they have completed.

#33202 - Node API allows the user to move a file in cloud storage into a directory with a/files/rename request.

#33094 - ascp delete-before-transfer does not support URI storage.

#32941 - Because pvcl does not support the --no-read --no-write option, ascp should fail transfers that use these options with pvcl. However, ascp ignores the options and continues the transfer.

#32934 - Some HTTP fallback transfers may appear to complete but then lose connection with the server and subsequently attempt to retransfer. This behavior has been observed only when the client system has an installation of the Internet accountability software Covenant Eyes, which captures the entire HTTP transmission before forwarding it to the server. If the file is large enough that this process takes longer than about 20 seconds, the server generates an inactivity timeout and cancels the session. Workaround: Reduce the probability of timeout by increasing the timeout length. This can be done by changing the Session Activity Timeout value in the HTTP Fallback tab in the server's configuration settings, or by using the asconfigurator settings, for example:
asconfigurator -x "http_server;session_activity_timeout,1200"

#32890 - In a successful transfer from the client to the server, ascp fails to copy the metafile and thus it does not transfer with the data file.

#32680 - The option to create a directory (ascp -d) may create a directory at a destination before an expected session failure.

#32669 - When a directory is linked from a subdirectory, it does not appear in the search result for a /files/search request in the Node API.

#32627 - When a filename is just a dot and an extension, (for example, .pdf), then it is reported as a file with "content_type"=>"application/pdf" or a hidden file named PDF; for example:
{"id"=>"27", "name"=>".pdf", "size"=>12, "content_type"=>"application/pdf", "type"=>"file", "modified_time"=>"2015-09-10T15:24:01Z", "access_level"=>"edit", "permission_count"=>0}

#32553 - When the FASP Session log source file list exceeds 500 bytes and contains multibyte UTF-8 characters, the output is truncated in a manner that creates an invalid UTF-8 sequence.

#32517 - Retransfer requests are unencrypted when transfers are encrypted. This can cause transfer failures in some scenarios, such as a network device dropping the retransfer request because it detects a bit sequence it considers malicious.

#32080 - An error message spams the log file used by asperanoded each time the console attempts to check the node for the status. A second error can occur when the end user fills in a docroot but doesn't have a valid system user; this results in an error that also spams the log file used by asperanoded.

#31791 - The user is unable to transfer files with the file extension .aspx. Workaround: Edit the resume_suffix setting in the aspera.conf file on the server, and then copy aspera.conf to each source user.

#31712 - For both AWS and Softlayer Swift storage, /files returns modified_time for files but not for folders.

#31423 - It is possible for an ascp transfer of a file on a full disk to be reported as successful by both the sender and the receiver.

#30690 - ascp fails with an inaccurate message—Error: failed to authenticate—when a server is configured to only accept ciphers that are no longer supported.

#30542 -/files PUT (file rename) should be fixed to involve only one PVCL operation but still return the proper 409 code when there is a destination conflict, and PVCL needs to return proper error codes stating that the move operation failed because of a destination conflict.

#30365 - In the Sync Job Detail page for pull operations, the size for pending files is reported as 0 Bytes. (Transfer sizes are reported correctly after the transfer completes.)

#30324 - During an ascp upload to cloud storage, if a mid-file read failure occurs on the sending computer (which is very rare) it can cause the server-side ascp to crash and possibly fail to report transfer completion. This read failure can be caused when a source file is truncated during transfer, a drive or file system fails, or a transfer is cancelled with Ctrl+C or other means.

#30231 - [Windows] If a pre-post script includes a path that contains ".." and the client is running on Windows, ascp fails with the following message:
Cannot open -e pre-post command: Unknown error

#30082 - [Windows] When the upload of a source file to the remove server fails, no error message appears in the UI.

#29848 - When <write_allowed><read_allowed>, and <dir_allowed> are all set to false inaspera.conf, Node API calls to URLs such as /files/browse are returning response code 500 Internal Server Error: instead of another code that better indicates that access to the resource is denied.

#29787 - When the docroot is not configured, the HTTP error code 500 ("Internal Server Error") is returned.

#29613 - On Windows, the folder timestamp is not preserved when top-level directories are transferred with the -p option. (The issue does not occur when folders are created on the day of transfer.)

#29255 - Download from SoftLayer of a file larger than 62GB is unsuccessful. Workaround: Do not use time-stamp preservation with SoftLayer.

#29187 - For content in cloud storage, the Node API does not display all the files in the docroot directory. Workaround: Use the /files/info request to browse the docroot directory when content is in cloud-based storage.

#29172 - When using --file-list or --file-pair-list with SoftLayer, some files may not be transferred. Note: This issue is due to a defect in Swift 1.12. Correcting it depends on its correction in Swift and the Swift upgrade on SoftLayer.

#29138 - For files in S3, the node daemon does not return the correct file modification time.

#29078 - When an access key is created using the standard node user authorization, the access key inherits that node user and its associated system user. Afterwards, asnodeadmin can be used to associate a new system user to the node user, but the new system user is not updated for the access key.

#29043 - Under serious WAN impairment, hotfolder push transfers with a prepost script intermittently report an assh error approximately 10-15 seconds after Transfer Statistics are reported to the log. This delay in process shutdown can hold the handle to the filelist such that subsequent ascp processes fail.

#29038 - The default checksum behavior with S3 (as with any cloud storage) is "none". An existing file on S3 is considered identical to the local file when their sizes are equal. Therefore the file on S3 is not overwritten, even when the content of S3 differs from the content of the local file. Using overwrite=always does not overwrite the file.

#28939 - If command line ascp neglects to specify a destination host, then the failed transfer (error: "no remote host specified") gets recorded in SQLite with client_node_id NULL, instead of being populated with the uuid of the node. This causes an issue with Console.

#28887 - When a file or directory is moved using the --exclude command, the old file or directory is not removed from the remote side.

#28820 - The following operations attempted in a Limelight directory fail or are unacceptably slow: copy/paste file, rename file, delete folder.

#28747 - Connect Server security: Encryption over HTTP is not enforced, and must be set on the server by an administrator.

#28686 - Multi-session uploads to S3 with file splitting must have the ascp chunk size and the S3 part size set to equal values. (File-splitting only works with S3.) File-splitting needs to respect a minimum split size, which for cloud storage is a part. That is, each ascp call needs to deliver full parts. Minimum split size is specified with the ascp option --multi-session-threshold. However, note that if the file size is greater than the multi-session threshold and smaller than the chunk size, then there is no split. Set chunk size and part size as follows:
  1. In aspera.conf set the chunk size to some value greater than 5 MB; for example:
    <chunk_size>67108864</chunk_size>   <!-- 64 MB -->
  2. In /opt/aspera/etc/trapd/s3.properties:
    • Set the upload part size (default 64 MB) to the same value as the chunk size.
    • Use a ONE_TO_ONE gathering policy:
      aspera.transfer.upload.part-size=64MB
      aspera.transfer.gathering-policy=ONE_TO_ONE

#28683 - asp-check.sh docroot warning message: If global docroot is enabled, asp-check.shreports the root user as insecure.

#28679 - In some cases, the fallback server does not respond.

#28288 - When the FIPS mode in aspera.conf is set to true, ascp fails with the error message, "ascp.exe: failed to authenticate, exiting.", without prompting the user for a password.

#28219 - asperanoded fails to delete a directory name containing backslashes. Workaround:
  1. Stop node.
  2. At the command prompt, issue the command ulimit -S -n 2048.
  3. Restart node in the same command-line shell (the limit is shell-specific).
  4. Send in the same /files/delete request as you did before; the head and everything in it will be removed.
Note: Prior to the command in Step 2 above, you can find out the current limit by running the following command: ulimit -a | grep "open files".

#27924 - Installing Connect Server on CentOS 7 on Google Cloud requires that Perl and the Perl module Data::Dumper be installed first.

#27908 - When asperacentral has database errors, it continues to allow new transfers.

#27879 - always_set_home does not work if the user's home directory does not exist.

#27823 - The user's Faspex Server goes down with an "Transfer Server errors detected" error message. An error, "trying to read events from a parser that already reported an error before" appears. Workaround: move the central-store.db to allow re-creation.

#27311 - Sync: An apply-local-docroot pull copies the local docroot path into the same path. For example, /home/user1/sync is copied into /home/user1/sync.

#27056 - There is no ascmd support for server-side symlink configuration.

#27007 - Aspera clients on UNIX systems use the HTTP_PROXY and HTTPS_PROXY environment variables when configured to use a FASP proxy; it does not report this in the logs. These variables are not used in HTTP Fallback.

#26618 - The ascmd process does not exit properly on newer versions of SLES (SUSE Linux Enterprise Server).

#26281 - At approximately 100 (or a similarly high number) of concurrent uploads to s3, intermittent transfer session failures occur.

#26185 - During an upload to S3 storage, an error may result if ascp reports a successful file transfer before the transfer to S3 completes.

#25915 - When the sync function is set to continuous, overwriting an existing file on the initiating node may result in the deletion of the previous version of the file on the responding node. Workaround: Perform the sync manually to recreate the deleted file.

#25913 - Aspera Drive attempts to sync a file that is still uploading to Shares, resulting in a file error.

#25865 - Allowing symbolic links to be copied also allows access to locations outside the docroot.

#25832 - Hot Folders: When creating or modifying hot folders, some controls in the File Handling tab do not work correctly: The option for resuming incomplete files does not work, but is still necessary to enable options for checking whether files at the source and files at the destination are the same. Selecting Compare File Attributes only checks whether the files are the same size. Selecting Compare Sparse File Checksums actually compares full checksums.

#25791 - Azure: Concurrent upload fails with the error Data transfer stalled, timed out.

#25636 - To use a larger chunk size to transfer big files to AWS S3 storage, some users modify the memory settings in the trapd initialization script, asperatrapd_init.sh. If you do so, be sure to preserve the script manually during upgrades to prevent it from being overwritten.

#25631 - When transferring from Windows to Mac using preserve-acls=native and remote-preserve-acls=native, ACL data is saved as xattr. Workaround: Do not use the native setting when transferring or syncing across platforms.

#25467 - async can now be used with a cloud storage.

#25127 - HTTP fallback temporary files (*.haspx) are not filtered out by the Node API.

#25083 - When hot folder transfers fail due to lack of permissions, the hot folders tab gives no indication that the failures are due to insufficient permissions.

#25042 - In the server-side aspera.conf, the none option for file checksum reporting is no longer supported; only md5 and sha1 are supported. The any option means allow the checksum format to be whichever format the client requests. On the client side, the none option is still available, as a command-line option. A setting of any on the client side results in an error with the message "ascp: unknown file checksum type any".

#24918 - When shares are added to or removed from a user while the user is connected to Shares in the Enterprise Server desktop app, the remote file browser is not updated. The list of shares displayed in the remote file browser bar is not refreshed until you disconnect from Shares and reconnect to it. In the case where a share has been removed, the user will be unable to perform certain actions and it will not be clear that this occurring because the user is no longer authorized for the share.

#24805 - For a synced directory, while all of the files and subdirectories under the source folder will have ACL preserved at destination, the ACL for the source folder itself will not be replicated to the remote destination folder.

#24780 - Both --preserve-acls and --remote-preserve-acls need to be specified in order for the target side of the pull to apply the acls.

#24549 - Drag-and-drop with the Linux version of the desktop client is not supported. Attempting a drag-and-drop upload does nothing. Attempting a drag-and-drop download causes the mouse cursor to display a plus sign, suggesting incorrectly that the feature is available.

#24390 - HTTP Fallback upload operations fails with the message "Failed to create HTTP session on server!".

#23954 - Sync: When the --preserve-acls or --preserve-xattrs option is used, async will not preserve the acl or xattr when either file acl or xattr is modified and (a) file content is unmodified or (b) file content is unmodified and the file is renamed.

#23876 - On S3, when always overwrite is set, files are missing at the destination.

#23583 - Certain option values for asperacentral job submission are case-sensitive.

#23503 - Akamai: Uploads of zero-byte files appear to be successful, but no file is present at the destination.

#23434 - Node API: Files that start with "._" are not returned by the Node API browse action.

#23400 - async 1.5 permits the root directory to be synced on Linux.

#23246 - Warnings are not generated about files skipped due to source base setting.

#23070 - If a transfer of several files is interrupted, the retries generate a no such file error for any already-transferred files.

#23004 - Console does not draw a line between the involved nodes for an async transfer. This issue occurs when using two managed nodes with two interfaces on two different networks.

#22998 - If the overwrite setting in the server's aspera.conf is deny, a destination file with the same name as the transfer file is still overwritten.

#22905 - On S3, when copying a file with ascp, if a slash is appended to the destination -- for example, /path/ -- the file is renamed path/. Because of the trailing slash, it appears to be a directory, but is actually a file.

#22848 - On S3, uploading a file to a non-existent destination path, and without the -d option, creates the destination path anyway.

#22741 - When the --file-checksum feature is enabled, transferring ES 3.3.3 to ES 3.3.0, the file is corrupted at the destination with additional characters appended to the end.

#22633 - Sync does not support large xattr/ResourceForks.

#22619 - In the Node API, file searches now follow symbolic links.

#22618 - The behavior of the Node API paths filter changed with Enterprise Server 3.4. It now returns only the directory matching the filter, not the directory's content.

#22044 - Sync: asyncadmin reports all sessions locked after the last actual running session.

#21978 - ES is unable to view logs as superuser.

#21629 - Connect Server aspera-dirlist.pl does not accurately reflect file permissions for user actions.

#21604 - Extremely large socket buffer value causes overflow, resulting in an error when creating a UDP socket.

#21014 - When creating a file with vi during a sync, the swap file is in conflict.

#20906 - async can't create a watch on an unreadable directory; therefore it does not get notified when permissions change. In addition, async treats an unreadable directory as "skip" rather than reporting an error or conflict.

#20767 - Sync: -R log dir creation from Linux to Windows with spaces in the path will only get as far as the first space in the path.

#20576 - Target Rate must be in the range 0 – 100 if TargetRateAsPercentage is yes; however, setting a target rate greater than 100 does not return an error. (Note that rate percentage is a deprecated feature.)

#20002 - Inconsistent behavior regarding symbolic link following: /files/browse doesn't follow the links and reports links and their target (final type and next name), while /files/info reports symbolic links as files or directories.

#19945 - asyncadmin creates SHM and WAL files for read-only operations. Once asyncadmin is run as the root, async run by the user does not have permission to access the existing SHM and WAL files and thus async fails. This is due to a bug in SQLite.

#18913 - Sync database logging: cancelled or stopped jobs are reported as completed in fasp_sessions.status.

#18780 - An async push conflict occurs when a new file is renamed quickly.

#18659 - Searching with very long pathnames (over 520 characters) results in an "insufficient buffer space" error.

#18530 - FASP and Sync now accept and verify SSH remote-host keys.

#18368 - Files with a backslash in the filename are not displayed in the list when the user browses the remote source on the new package page.

#17243 - An S3 download fails when a filename contains square brackets or a caret.

#17231 - On Linux, an async pull does not restore access time on the source side.

#16911 - Characters in the async session option that are not preceded by a "-" or "--" are ignored, therefore failing to trigger an error message. Any session options specified (such as -lor -a) after the string of characters not preceded by a "-" or "--" are also ignored. The session will still run, using the default values, without notifying the user that the new settings have not been applied.

#16390 - Unicode filenames appear incorrectly in pre-processing and post-processing email notifications.

#13761 - If filenames contain "\" or new line, async will try to transfer but fail, causing the internal transfer queue to become full and the synchronization to stall (or hang after all the other files have completed).

#13645 - When a directory is renamed during transfer, Sync continues running and never completes.

#11550 - If preserve_attributes is set to none on the server side, when downloading a single file with the -p option, the file has an invalid atime value.

#9126 - Sync logs events at a fast rate. The log destination must have room for large log files.

#8620 - Enterprise Server for Linux fails to install if the dependency perl-CGI is not installed. Workaround: Install perL(CGI), then rerun the Enterprise Server installation. For example, to install perl(CGI) on Fedora, run yum install perl-CGI.

#8538 - If a file fails to be renamed from .partial after it transfers, the transfer is still considered successful and the session will continue (with additional files being renamed successfully). The receiver side logs an error message, but no error is logged on the sender side. Successful transfers that were not properly renamed from .partial should be manually renamed.

#8534 - IPv6 requires ascp to have the -6 option, which is currently disabled on the server side if the user is configured for aspshell.

#8412 - Backslashes in filenames is not supported.

#6448 - When your regional and language options are set to language standards and formats that do not use "." to represent the decimal point, values are displayed inconsistently using both "." and the native character to represent the decimal point.

#6443 - If admin A creates a new hot folder and exits the desktop client, and then admin B exits, the new hot folder created by admin A created is lost.

PRODUCT SUPPORT

For on-line support resources for Aspera products, including raising new support tickets, please visit the Aspera Support Portal. Note that you may have an existing account if you contacted the Aspera support team in the past. Before creating a new account, first try setting a password for the email that you use to interact with us. You may also call one of our regional support centers.

 
Have more questions? Submit a request

0 Comments

Please sign in to leave a comment.
Powered by Zendesk