Recent Posts

Pages: 1 [2] 3 4 ... 10
11
Known Bugs, Fixes, and Patches / Qube 7.0 Perl-based jobtypes fail to start on linux
« Last post by jburk on August 03, 2018, 08:57:36 PM »
We have reports of the Perl-based jobtypes on linux (Maya "LoadOnce"jobtype is the only current one that is still Perl-based) failing to start on linux workers with the qube-core-7.0 package installed, and exiting with the error:

Can't load '/usr/local/pfx/qube//api/perl/qb.so' for module qb: libQt5Core.so.5: cannot open shared object file: No such file or directory at /usr/lib64/perl5/DynaLoader.pm line 190.

We will be fixing this in our next build of the qube-core, but for now the work-around is to run the following command as the root user on each worker running Qube 7:

echo /usr/local/pfx/qube/lib/Qt > /etc/ld.so.conf.d/pfx.conf; /sbin/ldconfig
12
Announcements / CentOS-7.4 Now Supported in Qube! Maintenance Release 7.0-1c
« Last post by jburk on August 02, 2018, 06:31:16 PM »
Full support for CentOS-7.4 has been added to the existing 7.0-1c maintenance release of Qube.  This includes all core functionality and the graphical user interfaces.

See What's New in Qube and Qube Release Notes
13
Announcements / Maintenance/Patch Release 7.0-1c of Qube! Now Available
« Last post by pipelinescott on July 26, 2018, 06:04:43 PM »
This morning we released the 7.0-1c maintenance or patch version of 7.0-1 for Qube!. It contains several fixes that we believe may be of interest to several client installations of Qube!. The highlights from those fixes are as follows:

@FIX: Added time filters to job info API call (qb.jobinfo())
@FIX: Agenda items failing due to timeout won’t auto-retry
@FIX: 7.0-1 supe installer won’t work on platforms with python 2.6 (issue with init_supe_db.py script)
@FIX: Inaccurate cumulative cpu time for agenda items

You can download this latest version of Qube! here:
https://www.pipelinefx.com/downloadversions/

See the latest Qube! compatibility and tech specs:
https://www.pipelinefx.com/technical-specifications/

The Release Notes can be viewed here:
http://docs.pipelinefx.com/display/RELNOTES/PipelineFX+Release+Notes
14
Announcements / Maintenance Release 6.10-1 of Qube! Now Available
« Last post by jburk on July 24, 2018, 04:55:02 AM »
This is a maintenance release of the Qube! Core/Supervisor/Worker/ArtistView/WranglerView products.

This is a recommended release for all customers running Qube v6.10-0; customers already running v6.10-0a need only upgrade if they are impacted by any issues addressed by this release.

Notable changes and fixes are:
  • Workers are always "auto-remove"d when shut down, even if "auto_remove" is not set in worker_flags.
  • Absurdly inaccurate cumulative cpu time for agenda items noted in rare cases (4000+ days)
  • Data warehouse slot counts (farm size) can be inaccurate at times; a worker whose configured slot count changes worker is added to the worker_dim dimension table as many times as there are expired entries for that same worker.   This can cause the pfx_dw.slotcount_fact table which tracks farm size to grow very large.
Please see http://docs.pipelinefx.com/display/RELNOTES/PipelineFX+Release+Notes for the complete release notes.
15
Announcements / CentOS-7.5 Now Supported in Qube! Maintenance Release 7.0-1
« Last post by jburk on July 20, 2018, 12:11:10 AM »
Full support for CentOS-7.5 has been added to the existing 7.0-1 maintenance release of Qube.  This includes all core functionality and the graphical user interfaces.

See What's New in Qube and Qube Release Notes
16
Announcements / Maintenance Release 7.0-1 of Qube! Now Available
« Last post by pipelinescott on July 10, 2018, 12:02:57 AM »
Today, we have released the 7.0-1 version of Qube!. It is a maintenance release with fixes for the recently released version of Qube! 7.0-0 that came with Postgres.

See the associated changes in the release notes (if these notes are not live on the evening of 7/9/2018 Pacific Time, please check back on 7/10):
http://docs.pipelinefx.com/display/RELNOTES/PipelineFX+Release+Notes

Download the Qube Installer:
https://www.pipelinefx.com/downloadversions/

Check the Technical Specs:
https://www.pipelinefx.com/technical-specifications/
17
We've had a report from one customer that installing the Qube 7.0-0a supervisor on a Windows Server 2008 R2 SP1 host fails within the first few seconds.  We were able to reproduce this on the same version of Windows Server.

The problem can be identified by:
  • supervisor installation fails very quickly (~2s)
  • Running the "C:\Program Files\pfx\qube\utils\qbservice.exe" program fails with a pop-up dialog saying some DLLs are missing, namely: api-ms-win-crt-runtime-l1-1-0.dll
The solution is to download and install the Visual C++ Redistributable for Visual Studio 2015 and then re-attempt the supervisor installation.
18
  • Versions affected: 7.0-0, 7.0-0a
  • Platforms affected: all
  • Severity: Newly-installed Windows and linux workers are unable to run python-based jobs.  Upgraded workers on Windows and linux, and all workers on OS X may exhibit more subtle behavior mimicking a version mismatch between API and jobtype versions

We've noted that the qube-core installer has been incorrectly packaged, and will place the files that should be in the QBDIR/api/python/qb/.. directory tree all into the api/python directory.  When upgrading from a previous version of Qube, the compiled .pyc files that were generated from the original.py modules are left behind, so the issue may not be immediately obvious as the .pyc can be used in place of the .py files. 

On Windows and linux, the files are only in the api/python directory, while on OS X the files are duplicated into api/python.

A subtle side effect of this bug is that if older .pyc files are loaded, it can appear that the API is an older version than the jobtypes despite all Qube components being of the same version, and some python-based jobtypes will fail with the following error:

AttributeError: 'CmdRangeBackEnd' object has no attribute 'outputPaths_required'


19
This can occur either during the upgrade procedure from Qube 6 to Qube 7 or if the supervisor configuration is modified with WranglerView 7.0-0 or 7.0-0a at any time after a successful installation or upgrade.

The root cause of both of these issues is due to a

database_port=3300 (or 3306)

line present in the supervisor's qb.conf file.  If your supervisor's supelog file contains entries like:

[Jun 13, 2018 17:37:29] jburk-15-mbPro[74627]: ERROR: QbDatabasePostgreSQL::_connect() Unable To Init Database: could not connect to server: No such file or directory
   Is the server running locally and accepting
   connections on Unix domain socket "/tmp/.s.PGSQL.3306"?


please open the qb.conf in a text editor, comment out the database_port line, and restart the supervisor service.

We will be patching the installation and upgrade procedures, as well as WranglerView, to address this in the next few days.
20
Announcements / Maintenance/Patch Release 7.0-0a of Qube! Now Available
« Last post by jburk on May 31, 2018, 07:56:04 PM »
This is a supervisor-only patch release of Qube! 7.0.0 that includes the following key fix:
  • @FIX: jobs submitted that reserve a global resource never run
This bug was introduced at 7.0-0, where jobs specifying any global resource reservation would be stuck indefinitely in "pending" state. 

It was previously noted in our 'Known Issues' board as topic: Qube 7.0-0 supervisor - jobs reserving a global resource never start

Complete release notes are available on our documentation site.
Pages: 1 [2] 3 4 ... 10