2016-08-31 19:38:33 +03:00
|
|
|
# Change Log
|
|
|
|
|
|
|
|
|
|
## [Unreleased]
|
2017-04-29 02:11:54 +03:00
|
|
|
|
2019-12-14 00:32:26 +03:00
|
|
|
## [3.9.1] - 2019-12-13
|
|
|
|
|
### Added
|
|
|
|
|
- Support `--no-wait` on pool creation to allow the command to skip
|
|
|
|
|
waiting for the pool to become idle
|
|
|
|
|
- Allow ability to ignore GPU warnings, please see pool configuration
|
|
|
|
|
- Ubuntu 18.04 SR-IOV IB/RDMA Packer script
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- **Breaking Change:** improved per-job autoscratch setup. As part of this
|
|
|
|
|
change the `auto_scratch` property in the jobs configuration has changed.
|
|
|
|
|
- Provide ability to setup via dependency or blocking behavior
|
|
|
|
|
- Allow specifying the number of VMs to span
|
|
|
|
|
- Allow specifying the per-job autoscratch task id
|
|
|
|
|
- Allow multiple multi-instance tasks per job in non-`native` mode
|
|
|
|
|
- Update GlusterFS version to 7
|
|
|
|
|
|
2019-11-17 07:07:48 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Fix merge task regression with enhanced autogenerated task id support
|
2019-12-14 00:32:26 +03:00
|
|
|
- Fix job schedule submission regression
|
|
|
|
|
([#329](https://github.com/Azure/batch-shipyard/issues/329))
|
|
|
|
|
- Fix per-job autoscratch provisioning due to upstream dependency changes
|
2019-11-17 07:07:48 +03:00
|
|
|
|
2019-11-15 22:08:52 +03:00
|
|
|
## [3.9.0] - 2019-11-15 (SC19 Edition)
|
|
|
|
|
### Added
|
|
|
|
|
- Support for [Encrypted Singularity Containers](docs/50-batch-shipyard-encrypted-containers.md)
|
|
|
|
|
- Enhanced support for autogenerated task id schemas
|
|
|
|
|
([#324](https://github.com/Azure/batch-shipyard/issues/324))
|
|
|
|
|
- SR-IOV IB/RDMA support for Ubuntu-based images
|
|
|
|
|
- Support for CentOS/CentOS-HPC 7.7
|
|
|
|
|
- Support Hyper-V Gen2 boot
|
|
|
|
|
- Ubuntu 16.04 SR-IOV IB/RDMA Packer script
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- **Breaking Change:** the `singularity_images` property in the global
|
|
|
|
|
configuration has been modified to accommodate encrypted container images.
|
|
|
|
|
Please see the global configuration doc for more information.
|
|
|
|
|
- **Breaking Change:** non-`native` pools using `STANDARD_NC24rs_v3` will now
|
|
|
|
|
default to using SR-IOV IB/RDMA settings. `native` pools using this VM size
|
|
|
|
|
will rely on the Azure Batch runtime to bind the correct container settings.
|
|
|
|
|
Please see the
|
|
|
|
|
[Azure Batch Guidance issue](https://github.com/Azure/Batch/issues/73)
|
|
|
|
|
for more information.
|
|
|
|
|
- Improve Azure blob/fileshare mount logic and add retries
|
|
|
|
|
- Set proper `FI_PROVIDER` env var to `intel-ofi` MPI runtime selection
|
|
|
|
|
- Updated Docker CE to 19.03.5
|
|
|
|
|
- Updated Singularity to 3.5.0
|
|
|
|
|
- Updated NC/ND driver to 418.87.01, NV driver to 430.46
|
|
|
|
|
- Updated LIS to 4.3.4
|
|
|
|
|
- Updated blobxfer to 1.9.4
|
|
|
|
|
- Updated Python to 3.7.5 for pre-built binaries
|
|
|
|
|
- Updated dependencies to latest, where applicable
|
|
|
|
|
- Update OSUMicroBenchmarks recipe to MVAPICH-2.3.2
|
|
|
|
|
|
2019-09-13 21:39:57 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Fix task `output_data` to correctly honor virtual directories in remote
|
|
|
|
|
paths for native pools
|
|
|
|
|
([#313](https://github.com/Azure/batch-shipyard/issues/313))
|
2019-11-15 22:08:52 +03:00
|
|
|
- Fix blobfuse not properly remounting on reboot
|
|
|
|
|
([#320](https://github.com/Azure/batch-shipyard/issues/320))
|
|
|
|
|
- Fix services/images table overflow
|
|
|
|
|
([#327](https://github.com/Azure/batch-shipyard/issues/327))
|
|
|
|
|
- Fix `pre_execution_command` in non-native mode to not be wrapped in
|
|
|
|
|
a shell
|
|
|
|
|
- Fix potential race in `--tail` between task completion and streaming last
|
|
|
|
|
portion of file
|
|
|
|
|
- Fix ephemeral device detection
|
|
|
|
|
- Allow further retries on packager manager contention
|
|
|
|
|
- Fix RemoteFS issues
|
|
|
|
|
- Non-Samba enabled clusters
|
|
|
|
|
- Single disk servers
|
|
|
|
|
- Fix Slurm issues
|
|
|
|
|
- Slurm cluster provisioning without Public IP address ([#322](https://github.com/Azure/batch-shipyard/pull/322))
|
|
|
|
|
- Priority tier check for partitions ([#323](https://github.com/Azure/batch-shipyard/pull/323))
|
|
|
|
|
- Various documentation updates
|
|
|
|
|
([#314](https://github.com/Azure/batch-shipyard/issues/314),
|
|
|
|
|
[#315](https://github.com/Azure/batch-shipyard/issues/315),
|
|
|
|
|
[#319](https://github.com/Azure/batch-shipyard/issues/319))
|
|
|
|
|
|
|
|
|
|
### Removed
|
|
|
|
|
- Support for CentOS 7.5, CentOS-HPC 7.1/7.3, and WindowsServerSemiAnnual
|
|
|
|
|
Datacenter-Core-1709-with-Containers-smalldisk/Datacenter-Core-1803-with-Containers-smalldisk
|
2019-09-13 21:39:57 +03:00
|
|
|
|
2019-09-12 19:29:08 +03:00
|
|
|
## [3.8.2] - 2019-09-12
|
2019-09-11 20:51:13 +03:00
|
|
|
### Changed
|
2019-09-12 19:29:08 +03:00
|
|
|
- `blobxfer` program output for handling `input_data` and `output_data` on
|
|
|
|
|
non-native pools is now captured to a separate file named
|
|
|
|
|
`blobxfer-download.log` and `blobxfer-upload.log`, respectively. This
|
|
|
|
|
prevents pollution of the stdout/stderr streams by the data transfer
|
|
|
|
|
phases.
|
2019-09-11 20:51:13 +03:00
|
|
|
- Updated Docker CE to 19.03.2
|
|
|
|
|
- Updated NC/ND driver to 418.87.00
|
|
|
|
|
- Updated blobxfer to 1.9.2
|
|
|
|
|
- Updated dependencies
|
|
|
|
|
|
2019-09-12 19:29:08 +03:00
|
|
|
### Fixed
|
2019-09-13 21:39:57 +03:00
|
|
|
- Fix prefix filter not being applied on task factory `remote_path`
|
|
|
|
|
([#303](https://github.com/Azure/batch-shipyard/issues/303))
|
2019-09-12 19:29:08 +03:00
|
|
|
- Fix non-string pickling in recurring job definitions
|
|
|
|
|
([#306](https://github.com/Azure/batch-shipyard/issues/306))
|
|
|
|
|
- Fix potential null values on node error collections and node agent info
|
|
|
|
|
on preempted nodes
|
|
|
|
|
([#307](https://github.com/Azure/batch-shipyard/issues/307),
|
|
|
|
|
[#309](https://github.com/Azure/batch-shipyard/issues/309))
|
|
|
|
|
- Fix task termination for infinite retry tasks and in non-native mode over
|
|
|
|
|
SSH ([#308](https://github.com/Azure/batch-shipyard/issues/308))
|
|
|
|
|
- Fix non-native data transfer sequence coupling
|
|
|
|
|
([#310](https://github.com/Azure/batch-shipyard/issues/310))
|
|
|
|
|
- Prevent job submission on pools without task runner
|
|
|
|
|
([#312](https://github.com/Azure/batch-shipyard/issues/312))
|
|
|
|
|
- Fix task `output_data` with include filters for native pools
|
|
|
|
|
([#313](https://github.com/Azure/batch-shipyard/issues/313))
|
2019-09-13 21:39:57 +03:00
|
|
|
- Fix downloading of cascade logs on start task failure
|
2019-09-12 19:29:08 +03:00
|
|
|
- Update documentation regarding AAD and subscription id requirements
|
|
|
|
|
along with better error messages
|
|
|
|
|
([#305](https://github.com/Azure/batch-shipyard/issues/305))
|
|
|
|
|
- Update documentation regarding Windows vs Linux environment
|
|
|
|
|
variables
|
|
|
|
|
([#311](https://github.com/Azure/batch-shipyard/issues/311))
|
|
|
|
|
|
2019-08-19 19:24:33 +03:00
|
|
|
## [3.8.1] - 2019-08-19
|
|
|
|
|
### Changed
|
|
|
|
|
- Updated blobxfer to 1.9.1
|
|
|
|
|
|
2019-08-15 00:48:29 +03:00
|
|
|
### Fixed
|
2019-08-19 19:24:33 +03:00
|
|
|
- Task runner regressions for non-native mode pools including `input_data`,
|
|
|
|
|
`output_data` and `pre_execution_command` for native mode pools
|
|
|
|
|
([#301](https://github.com/Azure/batch-shipyard/issues/301))
|
2019-08-15 00:48:29 +03:00
|
|
|
- Provisioning Network Direct RDMA VM sizes (A8/A9/NC24rX/H16r/H16mr) resulted
|
|
|
|
|
in start task failures
|
2019-08-19 19:24:33 +03:00
|
|
|
([#299](https://github.com/Azure/batch-shipyard/issues/299))
|
2019-08-15 00:48:29 +03:00
|
|
|
|
2019-08-14 05:19:48 +03:00
|
|
|
## [3.8.0] - 2019-08-13
|
|
|
|
|
### Added
|
|
|
|
|
- Revamped Singularity support, including support for Singularity 3,
|
|
|
|
|
SIF images, and pull support from ACR registries for SIF images via ORAS.
|
|
|
|
|
Please see the global and jobs configuration docs for more information.
|
|
|
|
|
([#146](https://github.com/Azure/batch-shipyard/issues/146))
|
|
|
|
|
- New MPI interface in jobs configuration for seamless multi-instance task
|
|
|
|
|
executions with automatic configuration for SR-IOV RDMA VM sizes with support
|
|
|
|
|
for popular MPI runtimes including OpenMPI, MPICH, Intel MPI, and MVAPICH
|
|
|
|
|
([#287](https://github.com/Azure/batch-shipyard/issues/287))
|
|
|
|
|
- Support for Hb/Hc SR-IOV RDMA VM sizes
|
|
|
|
|
([#277](https://github.com/Azure/batch-shipyard/issues/277))
|
|
|
|
|
- Support for NC/NV/H Promo VM sizes
|
|
|
|
|
- Support for user-specified job preparation and release tasks on the host
|
|
|
|
|
([#202](https://github.com/Azure/batch-shipyard/issues/202))
|
|
|
|
|
- Support for conditional output data
|
|
|
|
|
([#230](https://github.com/Azure/batch-shipyard/issues/230))
|
|
|
|
|
- Support for bring your own public IP addresses on Batch pools.
|
|
|
|
|
Please see the pool configuration doc and the
|
|
|
|
|
[Virtual Networks and Public IPs guide](docs/64-batch-shipyard-byovnet.md)
|
|
|
|
|
for more information.
|
|
|
|
|
- Support for Shared Image Gallery for custom images
|
|
|
|
|
- Support for CentOS HPC 7.6 native conversion
|
|
|
|
|
- Additional Slurm configuration options
|
|
|
|
|
- New recipes: mpiBench across various configurations,
|
|
|
|
|
OpenFOAM-Infiniband-OpenMPI, OSUMicroBenchmarks-Infiniband-MVAPICH
|
|
|
|
|
|
|
|
|
|
### Changed
|
2019-09-11 20:51:13 +03:00
|
|
|
- **Breaking Change:** jobs cannot be submitted against pre-`3.8.0` pools.
|
|
|
|
|
Pools must be re-created with `3.8.0` or later.
|
2019-08-14 05:19:48 +03:00
|
|
|
- **Breaking Change:** the `singularity_images` property in the global
|
|
|
|
|
configuration has been modified to accomodate Singularity 3 support.
|
|
|
|
|
Please see the global configuration doc for more information.
|
|
|
|
|
([#146](https://github.com/Azure/batch-shipyard/issues/146))
|
|
|
|
|
- **Breaking Change:** the `gpu` property in the jobs configuration has
|
|
|
|
|
been changed to `gpus` to accommodate the new native GPU execution
|
|
|
|
|
support in Docker 19.03. Please see the jobs configuration doc for
|
|
|
|
|
more information.
|
|
|
|
|
([#293](https://github.com/Azure/batch-shipyard/issues/293))
|
|
|
|
|
- `pool images` commands now support Singularity
|
|
|
|
|
- Non-native task execution is now proxied via script
|
|
|
|
|
([#235](https://github.com/Azure/batch-shipyard/issues/235))
|
|
|
|
|
- Batch Shipyard images have been migrated to the Microsoft Container Registry
|
|
|
|
|
([#278](https://github.com/Azure/batch-shipyard/issues/278))
|
|
|
|
|
- Updated Docker CE to 19.03.1
|
|
|
|
|
- Updated blobxfer to 1.9.0
|
|
|
|
|
- Updated LIS to 4.3.3
|
|
|
|
|
- Updated NC/ND driver to 418.67, NV driver to 430.30
|
|
|
|
|
- Updated Batch Insights to 1.3.0
|
|
|
|
|
- Updated dependencies to latest, where applicable
|
|
|
|
|
- Updated Python to 3.7.4 for pre-built binaries
|
|
|
|
|
- Updated Docker images to use Alpine 3.10
|
|
|
|
|
- Various recipe updates to showcase the new MPI schema, HPLinpack and HPCG
|
|
|
|
|
updates to SR-IOV RDMA VM sizes
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Cargo Batch service client update missed
|
|
|
|
|
([#274](https://github.com/Azure/batch-shipyard/issues/274), [#296](https://github.com/Azure/batch-shipyard/issues/296))
|
|
|
|
|
- Premium File Shares were not enumerating correctly with AAD
|
|
|
|
|
([#294](https://github.com/Azure/batch-shipyard/issues/294))
|
|
|
|
|
- Per-job autoscratch setup failing for more than 2 nodes
|
|
|
|
|
|
2019-08-12 22:50:15 +03:00
|
|
|
### Removed
|
2019-08-14 08:22:32 +03:00
|
|
|
- Peer-to-peer image distribution support
|
2019-08-12 22:50:15 +03:00
|
|
|
- Python 3.4 support
|
|
|
|
|
|
2019-07-24 05:49:38 +03:00
|
|
|
## [3.7.1] - 2019-07-23
|
|
|
|
|
### Fixed
|
|
|
|
|
- Detection of graph root was broken with new version of Docker client (CLI)
|
|
|
|
|
on GPU pools ([#291](https://github.com/Azure/batch-shipyard/issues/291))
|
|
|
|
|
|
2019-03-01 00:10:14 +03:00
|
|
|
## [3.7.0] - 2019-02-28
|
2018-12-10 21:50:38 +03:00
|
|
|
### Added
|
2019-03-01 00:10:14 +03:00
|
|
|
- Slurm on Batch support: provision Slurm clusters with elastic cloud bursting
|
|
|
|
|
on Azure Batch pools. Please see the
|
|
|
|
|
[Slurm on Batch guide](https://github.com/Azure/batch-shipyard/blob/master/docs/69-batch-shipyard-slurm.md).
|
|
|
|
|
- Batch Insights integration ([#259](https://github.com/Azure/batch-shipyard/issues/259)),
|
|
|
|
|
please see the pool and credentials configuration docs.
|
2019-01-16 18:28:06 +03:00
|
|
|
- Support environment variables on additional node prep commands ([#253](https://github.com/Azure/batch-shipyard/pull/253))
|
2019-03-01 00:10:14 +03:00
|
|
|
- Support CentOS 7.6
|
2019-02-07 20:43:31 +03:00
|
|
|
- `pool exists` command
|
|
|
|
|
- `--recreate` flag for `pool add` to allow existing pools to be recreated
|
2019-03-01 00:10:14 +03:00
|
|
|
- `fs cluster orchestrate` command
|
|
|
|
|
- Sample Windows container recipes ([#246](https://github.com/Azure/batch-shipyard/issues/246))
|
2019-01-15 21:50:17 +03:00
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- **Breaking Change:** the `additional_node_prep_commands` property has
|
|
|
|
|
been migrated under the new `additional_node_prep` property as
|
2019-01-16 18:28:06 +03:00
|
|
|
`commands` ([#252](https://github.com/Azure/batch-shipyard/issues/252))
|
2019-01-16 23:45:17 +03:00
|
|
|
- Performance improvements to speed up job submission with large task
|
|
|
|
|
factories or large amount of tasks. Verbosity of task generation progress
|
|
|
|
|
has been increased which can be modified with `-v`.
|
2019-03-01 00:10:14 +03:00
|
|
|
- Updated blobxfer to 1.7.0 ([#255](https://github.com/Azure/batch-shipyard/issues/255))
|
|
|
|
|
- Updated LIS, NV driver to 410.92 and NC/ND driver to 410.104
|
2019-01-16 23:45:17 +03:00
|
|
|
- Updated other dependencies to latest
|
2018-12-10 21:50:38 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Some commands were incorrectly failing due to nodeid conflicts with
|
2019-01-16 18:28:06 +03:00
|
|
|
supplied parameters ([#249](https://github.com/Azure/batch-shipyard/issues/249))
|
2019-01-16 23:45:17 +03:00
|
|
|
- Azure Function extension installation failure ([#260](https://github.com/Azure/batch-shipyard/issues/260))
|
2019-03-01 00:10:14 +03:00
|
|
|
- Block job submission on non-active pools ([#251](https://github.com/Azure/batch-shipyard/issues/251))
|
|
|
|
|
- Missing files included in binary distributions ([#258](https://github.com/Azure/batch-shipyard/issues/258))
|
2019-03-23 00:47:44 +03:00
|
|
|
- Pools with accelerated networking would fail provisioning sometimes due to
|
2019-02-07 20:43:31 +03:00
|
|
|
infiniband devices being present for non-RDMA VM sizes
|
2018-12-10 21:50:38 +03:00
|
|
|
|
2019-03-23 00:47:44 +03:00
|
|
|
### Security
|
|
|
|
|
- Updated Docker CE to 18.09.2 to address the runc CVE-2019-5736
|
|
|
|
|
- Updated Singularity to 2.6.1 to address the shared mount propagation
|
|
|
|
|
vulnerability CVE-2018-19295
|
|
|
|
|
|
2018-12-03 20:03:24 +03:00
|
|
|
## [3.6.1] - 2018-12-03
|
|
|
|
|
### Added
|
2018-12-10 21:54:18 +03:00
|
|
|
- `force_enable_task_dependencies` property in jobs configuration to turn
|
|
|
|
|
on task dependencies on a job even when no task dependencies are present
|
|
|
|
|
initially. This is useful when tasks are added at a later time that may
|
|
|
|
|
have dependencies. Please consult the jobs documentation for more information.
|
2018-12-03 20:03:24 +03:00
|
|
|
- Windows Server 2019 support
|
|
|
|
|
- Genomics and Bioinformatics recipes: BLAST and RNASeq
|
|
|
|
|
- PyTorch recipes
|
|
|
|
|
|
2018-11-19 22:20:08 +03:00
|
|
|
### Changed
|
|
|
|
|
- Updated Docker CE to 18.09.0
|
|
|
|
|
- Updated blobxfer to 1.5.5
|
2018-12-03 20:03:24 +03:00
|
|
|
- Updated NC/ND driver to 410.79
|
|
|
|
|
- Updated NV driver to 410.71 with CUDA10 support
|
2018-11-19 22:20:08 +03:00
|
|
|
- Updated other dependencies to latest
|
|
|
|
|
|
|
|
|
|
### Fixed
|
2018-12-03 20:03:24 +03:00
|
|
|
- `--tail` console output occasionally repeating characters
|
|
|
|
|
- NV provisioning regressions
|
|
|
|
|
- Windows node prep issue
|
2018-11-19 22:20:08 +03:00
|
|
|
- fs cluster status issue
|
|
|
|
|
- Retry MSI provisioning for discrete VM resources
|
|
|
|
|
|
2018-11-07 01:21:50 +03:00
|
|
|
## [3.6.0] - 2018-11-06 (SC18 Edition)
|
|
|
|
|
### Added
|
|
|
|
|
- Kata containers support: run containers on Linux compute nodes with a higher
|
|
|
|
|
level of isolation through lightweight VMs. Please see the pool doc for more
|
|
|
|
|
information.
|
|
|
|
|
- Per-job distributed scratch space support: create on-demand scratch
|
|
|
|
|
space shared between tasks of a job which can be particularly useful for MPI
|
|
|
|
|
and multi-instance tasks without having to manage a GlusterFS-on-compute
|
|
|
|
|
shared data volume. Please see both the pool doc and jobs doc for more
|
|
|
|
|
information.
|
|
|
|
|
- Add `restrict_default_bind_mounts` option to jobs specifications. This
|
|
|
|
|
will restrict automatic host directory bindings to the container filesystem
|
|
|
|
|
only to `$AZ_BATCH_TASK_DIR`. This is particularly useful in combination with
|
|
|
|
|
container runtimes enforcing VM-level isolation such as Kata containers.
|
|
|
|
|
- Allow installation and selection of multiple container runtimes along with
|
|
|
|
|
a default container runtime for Docker invocations. Please see the pool doc
|
|
|
|
|
for more information under `container_runtimes`.
|
|
|
|
|
- Support for Standard SSD and Ultra SSD managed disks for RemoteFS clusters.
|
|
|
|
|
In conjunction with this change, Availability Zone support has been added
|
|
|
|
|
for manage disks and storage cluster VMs. Please see the relevant
|
|
|
|
|
documentation for more information.
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- **Breaking Change:** the `premium` property under `remote_fs`:`managed_disks`
|
|
|
|
|
has been replaced with `sku`. Please see the RemoteFS configuration doc for
|
|
|
|
|
more information.
|
|
|
|
|
- **Breaking Change:** the Singularity container runtime is no longer installed
|
|
|
|
|
by default, please see the pool doc to configure pools to install Singularity
|
|
|
|
|
as needed under `container_runtimes`:`install`.
|
|
|
|
|
- Renamed MADL recipe to HPMLA
|
|
|
|
|
- Updated NC/ND driver to 410.72 with CUDA 10 support
|
|
|
|
|
- Updated blobxfer to 1.5.4
|
|
|
|
|
- Updated LIS, Prometheus, and Grafana
|
|
|
|
|
- Updated other dependencies to latest
|
|
|
|
|
- Updated binary builds and Windows Docker images to Python 3.7.1
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- `input_data` utilizing Azure File shares ([#243](https://github.com/Azure/batch-shipyard/issues/243))
|
|
|
|
|
- New NV driver location ([#244](https://github.com/Azure/batch-shipyard/issues/244))
|
|
|
|
|
- Fixed non-public Azure region AAD login issues
|
|
|
|
|
- Fixed Singularity image download issues
|
|
|
|
|
- Fixed Grafana update regression with default Batch Shipayrd Dashboard
|
|
|
|
|
- Fixed SSH login to monitoring resource after federation feature merge
|
|
|
|
|
- Enable Singularity on Ubuntu 18.04
|
|
|
|
|
|
|
|
|
|
### Removed
|
|
|
|
|
- Debian 8 host support
|
|
|
|
|
|
2018-09-20 22:45:01 +03:00
|
|
|
## [3.6.0b1] - 2018-09-20
|
|
|
|
|
### Added
|
|
|
|
|
- Task and node count commands: `jobs tasks counts` and `pool nodes counts`
|
|
|
|
|
respectively ([#228](https://github.com/Azure/batch-shipyard/issues/228)).
|
|
|
|
|
Please see the usage doc for more information.
|
2018-10-04 23:51:27 +03:00
|
|
|
- Enhance blocked action tracking for federations. Please see the usage
|
|
|
|
|
doc for `fed jobs list` for more information.
|
2018-09-20 22:45:01 +03:00
|
|
|
- Support for Ubuntu 18.04
|
|
|
|
|
- Support for CentOS 7.5 in both non-native and native mode
|
|
|
|
|
- MacOS binary for the CLI
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Updated Docker to 18.06.1
|
|
|
|
|
- Updated Singularity to 2.6.0
|
|
|
|
|
- Updated blobxfer to 1.5.0
|
2018-10-04 23:51:27 +03:00
|
|
|
- Updated Nvidia driver for NC/ND-series to 396.44
|
2018-09-20 22:45:01 +03:00
|
|
|
- Update various other dependencies to latest
|
|
|
|
|
- Windows binary is now signed
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Batch Shipyard site extension on nuget.org has been restored ([#224](https://github.com/Azure/batch-shipyard/issues/224))
|
|
|
|
|
- Pool auto-scaling beyond low priority limit ([#239](https://github.com/Azure/batch-shipyard/issues/239))
|
2018-10-04 23:51:27 +03:00
|
|
|
- Fix `jobs tasks term` command without pool SSH info
|
2018-09-20 22:45:01 +03:00
|
|
|
- Fix task id generator for federations
|
|
|
|
|
|
2018-08-06 20:35:31 +03:00
|
|
|
## [3.6.0a1] - 2018-08-06
|
|
|
|
|
### Added
|
|
|
|
|
- Federation support. Please see the
|
|
|
|
|
[federation guide](https://github.com/Azure/batch-shipyard/blob/master/docs/68-batch-shipyard-federation.md)
|
|
|
|
|
for more information.
|
|
|
|
|
- `monitor status` command with `--raw` support
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Updated dependencies
|
|
|
|
|
|
2018-07-31 20:28:30 +03:00
|
|
|
## [3.5.3] - 2018-07-31
|
|
|
|
|
### Added
|
|
|
|
|
- Support Docker image preload delay for Linux native container pools.
|
|
|
|
|
Please see the global configuration docs for more information.
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Improve registry login robustness with retries
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Docker Hub private registry login failures
|
|
|
|
|
- Environment variable issues ([#234](https://github.com/Azure/batch-shipyard/issues/234))
|
|
|
|
|
|
2018-07-20 19:13:44 +03:00
|
|
|
## [3.5.2] - 2018-07-20
|
|
|
|
|
### Fixed
|
|
|
|
|
- Non-native pool allocation on N-series VMs failing due to unpinned
|
|
|
|
|
dependent package for nvidia-docker2 ([#231](https://github.com/Azure/batch-shipyard/issues/231))
|
|
|
|
|
|
2018-07-17 23:16:23 +03:00
|
|
|
## [3.5.1] - 2018-07-17
|
|
|
|
|
### Changed
|
|
|
|
|
- Update GlusterFS on Compute on CentOS to 4.1
|
|
|
|
|
- Updated NC/ND Nvidia driver to 396.37
|
|
|
|
|
- Updated NV Nvidia driver to 390.75
|
|
|
|
|
- Updated LIS, Prometheus and Grafana
|
|
|
|
|
- Updated dependencies
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Properly terminate image pull without fallback on failure
|
|
|
|
|
- Fix pool metadata dump check logic
|
|
|
|
|
- Fix storage cluster provisioning with Node Exporter options
|
|
|
|
|
|
2018-06-29 23:24:27 +03:00
|
|
|
## [3.5.0] - 2018-06-29
|
|
|
|
|
### Added
|
|
|
|
|
- CentOS 7.5 and Microsoft Windows Server semi-annual
|
|
|
|
|
`datacenter-core-1803-with-containers-smalldisk` host support. Please see
|
|
|
|
|
the platform image support doc for more information.
|
|
|
|
|
- `fallback_registry` to improve robustness during provisioning when Docker
|
|
|
|
|
Hub has an outage or is degraded
|
|
|
|
|
([#215](https://github.com/Azure/batch-shipyard/issues/215), [#217](https://github.com/Azure/batch-shipyard/issues/217))
|
|
|
|
|
- `misc mirror-images` command to help mirror Batch Shipyard system
|
2018-07-02 00:34:03 +03:00
|
|
|
images to the designated fallback registry
|
2018-06-29 23:24:27 +03:00
|
|
|
- Support for XFS filesystem in storage clusters ([#218](https://github.com/Azure/batch-shipyard/issues/219))
|
|
|
|
|
- Experimental support for disk array RAID expansion for mdadm-based devices
|
|
|
|
|
via `fs cluster expand`
|
|
|
|
|
- Option to auto-upload Batch compute node service logs on unusable ([#216](https://github.com/Azure/batch-shipyard/issues/216))
|
|
|
|
|
- Microsoft Azure Distributed Linear Learner recipe ([#195](https://github.com/Azure/batch-shipyard/pull/195))
|
|
|
|
|
|
2018-06-15 23:22:22 +03:00
|
|
|
### Changed
|
2018-06-29 23:24:27 +03:00
|
|
|
- `pool nodes list` can now filter nodes with start task failed and/or
|
|
|
|
|
unusable states
|
|
|
|
|
- `diag logs upload` command can generate a read only SAS for the target
|
|
|
|
|
container via `--generate-sas`
|
|
|
|
|
- `storage clear` and `storage del` now allow multiple `--poolid` arguments
|
|
|
|
|
along with `--diagnostics-logs` to clear/delete diagnostics logs containers
|
|
|
|
|
- `storage sas create` now allows container and file share level SAS creation
|
|
|
|
|
along with `--list` permission now available as an option
|
|
|
|
|
- Pools failing to allocate with unusable or start task failed nodes will now
|
|
|
|
|
dump a listing of problematic nodes detailing the error
|
|
|
|
|
- Updated RemoteFS storage clusters using GlusterFS and Ubuntu/Debian-based
|
|
|
|
|
GlusterFS-on-compute to 4.1
|
|
|
|
|
- Updated blobxfer to 1.3.1
|
|
|
|
|
- Updated Singularity to 2.5.1
|
2018-06-15 23:22:22 +03:00
|
|
|
- Updated dependencies
|
|
|
|
|
|
2018-06-29 23:24:27 +03:00
|
|
|
### Fixed
|
|
|
|
|
- GlusterFS on compute provisioning ([#220](https://github.com/Azure/batch-shipyard/issues/220))
|
|
|
|
|
- Regression in KeyVault credential conf loading ([#214](https://github.com/Azure/batch-shipyard/issues/214))
|
|
|
|
|
- Task file mover command arg left unpopulated ([#29](https://github.com/Azure/batch-shipyard/issues/29))
|
|
|
|
|
- Recurring job manager failing to unpickle tasks with dependencies ([#221](https://github.com/Azure/batch-shipyard/issues/221))
|
|
|
|
|
- Task add regression when collections are too large from individual slices
|
|
|
|
|
|
|
|
|
|
### Removed
|
|
|
|
|
- CentOS 7.3 host support
|
|
|
|
|
|
2018-06-13 23:25:15 +03:00
|
|
|
## [3.5.0b3] - 2018-06-13
|
|
|
|
|
### Changed
|
|
|
|
|
- All supported platform images support blobfuse, including native mode
|
|
|
|
|
|
|
|
|
|
### Fixed
|
2018-06-29 23:24:27 +03:00
|
|
|
- blobfuse check preventing valid pool provisioning ([#213](https://github.com/Azure/batch-shipyard/issues/213))
|
2018-06-13 23:25:15 +03:00
|
|
|
- Pool resize not adding SSH users if keys are specified
|
|
|
|
|
|
2018-06-12 02:43:21 +03:00
|
|
|
## [3.5.0b2] - 2018-06-12
|
2018-05-22 00:29:02 +03:00
|
|
|
### Added
|
2018-06-29 23:24:27 +03:00
|
|
|
- Support for Prometheus monitoring and Grafana visualization
|
|
|
|
|
([#205](https://github.com/Azure/batch-shipyard/issues/205)). Please see the
|
|
|
|
|
monitoring doc and
|
2018-06-12 02:43:21 +03:00
|
|
|
[guide](https://github.com/Azure/batch-shipyard/blob/master/docs/66-batch-shipyard-resource-monitoring.md)
|
|
|
|
|
for more information.
|
|
|
|
|
- Support for specifying a maximum increment per autoscale evaluation and
|
2018-06-29 23:24:27 +03:00
|
|
|
the ability to define weekdays and workhours ([#210](https://github.com/Azure/batch-shipyard/issues/210))
|
2018-06-12 02:43:21 +03:00
|
|
|
- Support for native container support Marketplace platform images. Please
|
2018-06-29 23:24:27 +03:00
|
|
|
see the platform image support doc for more information. ([#204](https://github.com/Azure/batch-shipyard/issues/204))
|
|
|
|
|
- Allow configuration to enable SSH users to access Docker daemon ([#206](https://github.com/Azure/batch-shipyard/issues/206))
|
|
|
|
|
- Support for GPUs on CentOS 7.4 ([#199](https://github.com/Azure/batch-shipyard/issues/199))
|
|
|
|
|
- Support for CentOS-HPC 7.4 ([#184](https://github.com/Azure/batch-shipyard/issues/184))
|
2018-05-22 00:29:02 +03:00
|
|
|
|
2018-05-17 17:30:31 +03:00
|
|
|
### Changed
|
|
|
|
|
- **Breaking Change:** You can no longer specify both an `account_key`
|
|
|
|
|
and `aad` with the `batch` section of the credentials config. The prior
|
|
|
|
|
behavior was that `account_key` would take precedence over `aad`. Now
|
|
|
|
|
these options are mutually exclusive. This will now break configurations
|
|
|
|
|
that specified `aad` at the global level while having a shared `account_key`
|
2018-06-29 23:24:27 +03:00
|
|
|
at the `batch` level. ([#197](https://github.com/Azure/batch-shipyard/issues/197))
|
2018-06-12 02:43:21 +03:00
|
|
|
- **Breaking Change:** `install.sh` now installs into a virtual env by
|
|
|
|
|
default. Use the `-u` switch to retain the old (non-recommended)
|
2018-06-29 23:24:27 +03:00
|
|
|
default behavior. ([#200](https://github.com/Azure/batch-shipyard/issues/200))
|
2018-06-12 02:43:21 +03:00
|
|
|
- GlusterFS for RemoteFS and gluster on compute updated to 4.0.
|
2018-05-22 00:29:02 +03:00
|
|
|
- Update NC driver to 396.26 supporting CUDA 9.2
|
2018-05-30 01:02:32 +03:00
|
|
|
- blobxfer updated to 1.2.1
|
2018-05-22 00:29:02 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
2018-06-12 02:43:21 +03:00
|
|
|
- Errant credentials check for configuration from commandline which affected
|
|
|
|
|
config load from KeyVault
|
|
|
|
|
- Blobxfer extra options regression
|
2018-06-29 23:24:27 +03:00
|
|
|
- Cache container/file share creations for data egress ([#211](https://github.com/Azure/batch-shipyard/issues/211))
|
2018-05-17 17:30:31 +03:00
|
|
|
|
2018-05-02 17:48:54 +03:00
|
|
|
## [3.5.0b1] - 2018-05-02
|
|
|
|
|
### Added
|
|
|
|
|
- Output to JSON for a subset of commands via `--raw` commmand line switch.
|
|
|
|
|
JSON output is directed to stdout. Please see the usage doc for which commands
|
2018-06-29 23:24:27 +03:00
|
|
|
are supported and important information regarding output stability. ([#177](https://github.com/Azure/batch-shipyard/issues/177))
|
2018-05-02 17:48:54 +03:00
|
|
|
- Allow AAD on the `storage` section in the credentials configuration.
|
2018-06-29 23:24:27 +03:00
|
|
|
Please see the credential doc for more information. ([#179](https://github.com/Azure/batch-shipyard/issues/179))
|
2018-05-02 17:48:54 +03:00
|
|
|
- Boot diagnostics are now enabled for all VMs provisioned for RemoteFS
|
2018-06-29 23:24:27 +03:00
|
|
|
clusters. This also enables serial console support in the portal. ([#193](https://github.com/Azure/batch-shipyard/issues/193))
|
2018-05-02 17:48:54 +03:00
|
|
|
- `product_iterables` task factory support. Please see the task factory
|
2018-06-29 23:24:27 +03:00
|
|
|
doc for more information. ([#187](https://github.com/Azure/batch-shipyard/issues/187))
|
2018-05-02 17:48:54 +03:00
|
|
|
- `default_working_dir` option as the job and task-level to set the
|
|
|
|
|
default working directory when a container executes as a task. Please
|
2018-06-29 23:24:27 +03:00
|
|
|
see the jobs doc for more information. ([#190](https://github.com/Azure/batch-shipyard/issues/190))
|
2018-05-02 17:48:54 +03:00
|
|
|
- `--no-generate-tunnel-script` option to `pool nodes grls`
|
|
|
|
|
|
2018-04-04 23:27:41 +03:00
|
|
|
### Changed
|
2018-05-02 17:48:54 +03:00
|
|
|
- Greatly improve throughput speed of many commands that internally iterated
|
2018-06-29 23:24:27 +03:00
|
|
|
sequences of actions ([#188](https://github.com/Azure/batch-shipyard/issues/188))
|
|
|
|
|
- RemoteFS clusters provisioned using Ubuntu 18.04-LTS
|
|
|
|
|
([#161](https://github.com/Azure/batch-shipyard/issues/161), [#185](https://github.com/Azure/batch-shipyard/issues/185))
|
2018-05-22 00:29:02 +03:00
|
|
|
- Update Nvidia NC driver to 390.46 supporting CUDA 9.1
|
2018-05-01 01:06:52 +03:00
|
|
|
- Update Nvidia NV driver to 390.42.
|
2018-05-02 17:48:54 +03:00
|
|
|
- Singularity updated to 2.5.0
|
|
|
|
|
- blobxfer updated to 1.2.0
|
2018-05-01 01:06:52 +03:00
|
|
|
- Docker CE updated to 18.03.1
|
2018-04-04 23:27:41 +03:00
|
|
|
- Update dependencies to latest
|
2018-06-29 23:24:27 +03:00
|
|
|
- Unify nodeprep scripts ([#176](https://github.com/Azure/batch-shipyard/issues/176))
|
|
|
|
|
- Integrate shellcheck ([#177](https://github.com/Azure/batch-shipyard/issues/177))
|
2018-05-02 17:48:54 +03:00
|
|
|
- Extend retry policy for all clients
|
|
|
|
|
- Add Windows file version info and icon to CLI binary
|
2018-04-04 23:27:41 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
2018-06-29 23:24:27 +03:00
|
|
|
- Kernel unattended upgrades causes GPU jobs to fail on reboot ([#174](https://github.com/Azure/batch-shipyard/issues/174))
|
2018-05-02 17:48:54 +03:00
|
|
|
- Task submission speed regression when using task factory or large task
|
2018-06-29 23:24:27 +03:00
|
|
|
arrays with unnamed tasks ([#183](https://github.com/Azure/batch-shipyard/issues/183))
|
2018-05-02 17:48:54 +03:00
|
|
|
- Fix determinism in cardinality of results from `pool nodes grls`
|
|
|
|
|
- Env var export for tasks without env vars
|
|
|
|
|
- Ensure Nvidia persistence mode on reboots
|
|
|
|
|
- Pin nvidia-docker2 installations
|
|
|
|
|
- Site extension broken install issues (and fallback manual recovery)
|
|
|
|
|
|
|
|
|
|
### Removed
|
2018-06-29 23:24:27 +03:00
|
|
|
- Ubuntu 14.04 host support ([#164](https://github.com/Azure/batch-shipyard/issues/164))
|
2018-04-04 23:27:41 +03:00
|
|
|
|
2018-03-26 23:15:44 +03:00
|
|
|
## [3.4.0] - 2018-03-26
|
2018-03-14 00:59:52 +03:00
|
|
|
### Added
|
2018-03-14 22:21:13 +03:00
|
|
|
- Support for adding network access rules to the remote access port (SSH or
|
|
|
|
|
RDP). Please see the pool configuration guide for more details.
|
2018-03-15 00:01:29 +03:00
|
|
|
- Support for adding certificate references to a pool. Please see the
|
2018-03-16 00:55:33 +03:00
|
|
|
pool configuration guide for more details. Also please see below for
|
|
|
|
|
improvements to the `cert` command.
|
2018-03-14 00:59:52 +03:00
|
|
|
- Support for
|
|
|
|
|
[NCv3 VM sizes](https://azure.microsoft.com/blog/ncv3-vms-generally-available-other-gpus-expanding-regions/).
|
|
|
|
|
Note that ND/NCv2/NCv3 all require separate quota approval; please raise a
|
|
|
|
|
ticket through the Azure Portal.
|
2018-03-26 21:58:45 +03:00
|
|
|
- Support for uploading Batch compute node service logs to the specified
|
|
|
|
|
Azure storage account used by Batch Shipyard. Please see the
|
|
|
|
|
`diag logs upload` command in the usage docs.
|
2018-03-14 00:59:52 +03:00
|
|
|
- Support for fine-tuning `/etc/exports` when creating NFS file servers via
|
|
|
|
|
`server_options` and `nfs`. Please see the remote FS configuration doc
|
|
|
|
|
for more information.
|
2018-03-16 19:25:35 +03:00
|
|
|
- Support for job-level default task exit condition options. These options
|
|
|
|
|
can be overriden on a per-task basis. Please see the job configuration doc
|
|
|
|
|
for more information.
|
2018-03-14 00:59:52 +03:00
|
|
|
|
2018-03-05 21:03:11 +03:00
|
|
|
### Changed
|
2018-03-16 00:55:33 +03:00
|
|
|
- Improve `cert` commands
|
|
|
|
|
- Support adding arbitrary cer, pem and pfx certificates to a Batch
|
|
|
|
|
account via command line options
|
|
|
|
|
- Support deleting arbitrary certificates by thumbprint, including
|
|
|
|
|
multiple at once; also ask for confirmation before deleting
|
|
|
|
|
- Support creating pem/pfx pairs with `cert create` without having
|
|
|
|
|
to define an `encryption` section in the global configuration for
|
|
|
|
|
use in scenarios outside of credential encryption
|
2018-03-14 00:59:52 +03:00
|
|
|
- `depends_on` and `depends_on_range` now apply to tasks generated by
|
|
|
|
|
`task_factory` (#173). Please see the job configuration doc for more
|
|
|
|
|
information.
|
2018-03-05 21:03:11 +03:00
|
|
|
- `pool nodes del` and `pool nodes reboot` now accept multiple `--nodeid`
|
|
|
|
|
arguments to specify deleting and rebooting multiple nodes at the same time,
|
|
|
|
|
respectively
|
|
|
|
|
- `pool nodes prune`, `pool nodes reboot`, `pool nodes zap` will now ask
|
|
|
|
|
for confirmation first. `-y` flag can be specified to suppress confirmation.
|
2018-03-14 00:59:52 +03:00
|
|
|
- Added Batch Shipyard version to user agent for all ARM clients
|
2018-03-26 19:26:38 +03:00
|
|
|
- Improved node prep scripts with more timestamp detail, Docker and
|
|
|
|
|
Nvidia details
|
2018-03-22 00:39:58 +03:00
|
|
|
- CUDA 9.1 support on ND/NCv2/NCv3 with Tesla Driver 390.30
|
2018-03-26 19:26:38 +03:00
|
|
|
- Docker CE updated to 18.03.0
|
2018-03-20 00:07:12 +03:00
|
|
|
- Singularity updated to 2.4.4
|
|
|
|
|
- Dependencies updated
|
2018-03-05 21:03:11 +03:00
|
|
|
|
2018-03-14 18:33:47 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Previous environment variable expansion fix applied to multi-instance tasks
|
2018-03-16 00:55:33 +03:00
|
|
|
- `jobs tasks list` command with undefined job action but with dependency
|
|
|
|
|
actions
|
2018-03-16 19:25:35 +03:00
|
|
|
- `job_action` for task default exit condition was being overwritten
|
|
|
|
|
incorrectly in certain scenarios
|
2018-03-14 18:33:47 +03:00
|
|
|
|
2018-03-01 19:25:49 +03:00
|
|
|
## [3.3.0] - 2018-03-01
|
2018-02-25 23:12:47 +03:00
|
|
|
### Added
|
2018-03-01 19:25:49 +03:00
|
|
|
- Support for specifying default task exit conditions (i.e., non-zero exit
|
|
|
|
|
codes). Please see the jobs configuration doc for more information.
|
|
|
|
|
- New commands (please see usage doc for more information):
|
|
|
|
|
- `pool nodes prune` will remove all unused Docker-related data on all
|
|
|
|
|
nodes in the pool (requires an SSH user)
|
|
|
|
|
- `pool nodes ps` performs a `docker ps -a` on all nodes in the pool
|
|
|
|
|
(requires an SSH user)
|
|
|
|
|
- `pool nodes zap` will kill (and optionally remove) **all** running
|
|
|
|
|
Docker containers on nodes in a pool (requires an SSH user). Note that
|
|
|
|
|
`jobs tasks term` is the preferred command to control individual
|
|
|
|
|
(or grouped) task termination or `jobs term --termtasks` to terminate
|
|
|
|
|
at the job level.
|
|
|
|
|
- `storage sas create` command added as a utility helper function to
|
|
|
|
|
create SAS tokens for given storage accounts in credentials
|
2018-03-01 00:10:14 +03:00
|
|
|
- Support for activating
|
|
|
|
|
[Azure Hybrid Use Benefit](https://azure.microsoft.com/pricing/hybrid-benefit/)
|
|
|
|
|
for Windows pools
|
2018-02-25 23:12:47 +03:00
|
|
|
|
2018-02-25 05:30:02 +03:00
|
|
|
### Changed
|
2018-03-01 19:25:49 +03:00
|
|
|
- Greatly expand pool, node, job, and task details for `list` sub-commands
|
|
|
|
|
- Expand error detail key/value pairs if present
|
2018-02-25 05:30:02 +03:00
|
|
|
- Name resources for RemoteFS with 3 digits (e.g., 000 instead of 0) for
|
|
|
|
|
improved alpha ordering
|
2018-03-01 19:25:49 +03:00
|
|
|
- Move site extension to nuget.org (#172)
|
2018-02-25 05:30:02 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
2018-03-01 19:25:49 +03:00
|
|
|
- Command lines in non-native mode now properly expand environment variables
|
|
|
|
|
with normal quoting
|
|
|
|
|
- RemoteFS cluster del with disks error
|
|
|
|
|
- RemoteFS cluster status decoding issues
|
2018-02-25 05:30:02 +03:00
|
|
|
- Unintended interaction between native mode and custom images
|
2018-03-01 19:25:49 +03:00
|
|
|
- Handle starting Docker service for all deployments
|
|
|
|
|
- Do not automatically mount storage cluster or custom linux mounts on boot
|
|
|
|
|
due to potential race conditions with the ephemeral disk mount
|
|
|
|
|
- Fix TLS issue with powershell (#171)
|
|
|
|
|
- Fix conflicts when using install.sh with Anaconda environments
|
2018-02-25 05:30:02 +03:00
|
|
|
- Update packer scripts and fix typos
|
|
|
|
|
- Minor doc updates
|
|
|
|
|
|
2018-02-21 19:26:13 +03:00
|
|
|
## [3.2.0] - 2018-02-21
|
2018-02-09 01:25:28 +03:00
|
|
|
### Added
|
2018-02-12 21:34:33 +03:00
|
|
|
- Custom Linux Mount support for `shared_data_volumes`. Please see the
|
|
|
|
|
global configuration doc for more information.
|
2018-03-01 19:25:49 +03:00
|
|
|
- New commands (please see usage doc for more information):
|
|
|
|
|
- `account` command added with the following sub-commands (requires
|
|
|
|
|
AAD auth):
|
|
|
|
|
- `info` provides information about a Batch account (including account
|
|
|
|
|
level quotas)
|
|
|
|
|
- `list` provides information about all (or a resource group subset)
|
|
|
|
|
of accounts within the subscription specified in credentials
|
|
|
|
|
- `quota` provides service level quota information for the
|
|
|
|
|
subscription for a given location
|
|
|
|
|
- `pool rdp` sub-command added, please see usage doc for more information.
|
|
|
|
|
Requires Batch Shipyard executing on Windows with target Windows
|
|
|
|
|
containers pools.
|
2018-02-18 00:24:52 +03:00
|
|
|
- `pool images update` command now supports updating Docker images
|
2018-02-12 21:34:33 +03:00
|
|
|
in native container support pools via SSH
|
2018-02-16 07:37:49 +03:00
|
|
|
- Ability to specify an AAD authority URL via the `aad`:`authority_url`
|
|
|
|
|
credential configuration, `--aad-authority-url` command line option or
|
|
|
|
|
`SHIPYARD_AAD_AUTHORITY_URL` environment variable. Please see relevant
|
|
|
|
|
documentation for credentials and usage.
|
2018-02-17 01:35:17 +03:00
|
|
|
- Support for CentOS 7.4 and Debian 9 compute node hosts. CentOS 7.4
|
2018-02-21 05:38:28 +03:00
|
|
|
on GPU nodes is currently unsupported; CentOS 7.3 will continue to work on
|
|
|
|
|
N-series.
|
2018-02-21 19:26:13 +03:00
|
|
|
- Support for publisher `MicrosoftWindowsServer`, offer
|
|
|
|
|
`WindowsServerSemiAnnual`, and sku
|
2018-02-17 01:35:17 +03:00
|
|
|
`Datacenter-Core-1709-with-Containers-smalldisk`
|
2018-02-21 05:38:28 +03:00
|
|
|
- `--delete-resource-group` option added to `fs disks del` command
|
|
|
|
|
- CentOS-HPC 7.1, CentOS 7.3 GPU, and CentOS 7.4 packer scripts added to
|
|
|
|
|
contrib area
|
2018-02-21 19:26:13 +03:00
|
|
|
- Add documentation for which `platform_image`s are supported
|
2018-02-09 01:25:28 +03:00
|
|
|
|
2018-02-09 20:33:25 +03:00
|
|
|
### Changed
|
2018-02-18 00:24:52 +03:00
|
|
|
- **Breaking Change:** `additional_node_prep_commands` is now a dictionary
|
|
|
|
|
of `pre` and `post` properties which are executed either before or after the
|
|
|
|
|
Batch Shipyard startup task. Please see the pool configuration doc for more
|
|
|
|
|
information.
|
2018-02-09 20:33:25 +03:00
|
|
|
- Allow provisioning of OpenLogic CentOS-HPC 7.1
|
2018-02-16 07:37:49 +03:00
|
|
|
- Default management endpoint for public Azure cloud updated
|
2018-02-17 00:56:46 +03:00
|
|
|
- Improve some error messages/handling
|
2018-02-15 01:20:27 +03:00
|
|
|
- Update dependencies to latest
|
2018-02-17 01:03:13 +03:00
|
|
|
- Linux pre-built binary is no longer gzipped
|
2018-02-21 05:38:28 +03:00
|
|
|
- Update packer scripts in contrib area
|
2018-02-09 20:33:25 +03:00
|
|
|
|
2018-02-07 19:55:10 +03:00
|
|
|
### Fixed
|
2018-02-16 07:37:49 +03:00
|
|
|
- AAD auth for ARM endpoints in non-public Azure cloud regions
|
2018-02-09 01:25:28 +03:00
|
|
|
- Custom image + native mode deployment for Linux pools
|
|
|
|
|
- Potential command launch problems in native mode
|
2018-02-07 19:55:10 +03:00
|
|
|
- Minor schema validation updates
|
2018-02-09 01:25:28 +03:00
|
|
|
- AAD check logic for different points in pool allocation
|
|
|
|
|
- `--ssh` parameter for `pool images update` was not correctly set as a flag
|
2018-02-15 01:20:27 +03:00
|
|
|
- `--jobs` was not properly being merged with `--configdir` (#163)
|
2018-02-18 00:24:52 +03:00
|
|
|
- Fix regression in `pool images update` that would not login to
|
2018-02-15 01:20:27 +03:00
|
|
|
registries in multi-instance mode
|
2018-02-18 00:24:52 +03:00
|
|
|
- Fix `pool images` commands to more reliably work with SSH
|
2018-02-18 04:54:20 +03:00
|
|
|
- Fix `output_data` with windows containers pools (#165)
|
2018-02-07 19:55:10 +03:00
|
|
|
|
2018-01-30 22:21:22 +03:00
|
|
|
## [3.1.0] - 2018-01-30
|
2018-01-22 19:03:27 +03:00
|
|
|
### Added
|
|
|
|
|
- Configuration validation. Validator supports both YAML and JSON
|
|
|
|
|
configuration, please see special note in the Removed section below (#145)
|
2018-01-24 01:23:39 +03:00
|
|
|
- Support for Azure Blob storage container mounting via blobfuse (#159)
|
2018-01-24 22:30:34 +03:00
|
|
|
- Support for merge tasks which depend on all tasks specified in the
|
|
|
|
|
`tasks` array. Please see the jobs configuration guide for more
|
|
|
|
|
information (#149).
|
2018-01-26 00:37:29 +03:00
|
|
|
- Support for accelerated networking in RemoteFS storage clusters (#158)
|
2018-01-22 19:03:27 +03:00
|
|
|
|
|
|
|
|
### Changed
|
2018-01-30 22:21:22 +03:00
|
|
|
- Update Docker CE to 17.12.0 for Ubuntu/CentOS
|
|
|
|
|
- Update nvidia-docker 1.0.1 to nvidia-docker2
|
|
|
|
|
- Update blobxfer to 1.1.1
|
2018-01-22 19:03:27 +03:00
|
|
|
- Updated dependencies to latest
|
|
|
|
|
|
2018-01-26 00:37:29 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Disabling `remove_container_after_exit`, `gpu`, `infiniband` at the
|
|
|
|
|
task-level was not being honored properly
|
|
|
|
|
|
2018-01-22 19:03:27 +03:00
|
|
|
### Removed
|
|
|
|
|
- Integration of the schema validator has now removed or enforced strict
|
|
|
|
|
behavior for the following previously deprecated configuration properties:
|
|
|
|
|
- `credentials`:`batch`:`account` has been removed
|
2018-01-24 22:30:34 +03:00
|
|
|
- `pool_specification`:`vm_count` must be a map of `dedicated` and
|
|
|
|
|
`low_priority` VM counts
|
2018-01-22 19:03:27 +03:00
|
|
|
- `pool_specification`:`vm_configuration` must be specified instead of
|
|
|
|
|
directly specifying `publisher`, `offer`, `sku` on `pool_specification`
|
2018-01-30 22:21:22 +03:00
|
|
|
- `global_resources`:`docker_volumes` is no longer valid and must be
|
|
|
|
|
replaced with `global_resources`:`volumes`
|
|
|
|
|
- `job_specifications`:`tasks`:`image` is no longer valid and must be
|
|
|
|
|
replaced with `job_specifications`:`tasks`:`docker_image`
|
2018-01-22 19:03:27 +03:00
|
|
|
|
2018-01-22 19:39:12 +03:00
|
|
|
## [3.0.3] - 2018-01-22
|
|
|
|
|
### Security
|
|
|
|
|
- Update NV driver to 384.111 to work with updated Linux kernels with
|
|
|
|
|
speculative execution side channel vulnerability patches (#154)
|
|
|
|
|
|
2018-01-12 20:06:57 +03:00
|
|
|
## [3.0.2] - 2018-01-12
|
2017-12-09 03:28:25 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Errant bind option being propagated to volume name
|
2018-01-12 20:06:57 +03:00
|
|
|
- Clarify error path on attempting `infiniband` on non-supported images
|
|
|
|
|
- Fix quickstart recipe links from Read the Docs (#150)
|
|
|
|
|
|
|
|
|
|
### Security
|
|
|
|
|
- Update NC driver to 384.111 to work with updated Linux kernels with
|
2018-01-22 19:39:12 +03:00
|
|
|
speculative execution side channel vulnerability patches
|
2017-12-09 03:28:25 +03:00
|
|
|
|
2017-11-22 19:30:13 +03:00
|
|
|
## [3.0.1] - 2017-11-22
|
|
|
|
|
### Fixed
|
|
|
|
|
- Fix on-disk file naming for Docker images pulled with Singularity
|
|
|
|
|
- Data movement regressions
|
|
|
|
|
- Public IP configs in RemoteFS recipes
|
|
|
|
|
- Support more than 16 data disks per VM for RemoteFS servers
|
|
|
|
|
- Documentation and other typos
|
|
|
|
|
|
2017-11-13 09:45:19 +03:00
|
|
|
## [3.0.0] - 2017-11-13 (SC17 Edition)
|
2017-11-09 23:25:34 +03:00
|
|
|
### Added
|
|
|
|
|
- CLI Singularity image (#135)
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Start LUN numbering for remote fs disks at 0
|
2017-11-10 20:22:18 +03:00
|
|
|
- Allow path to `python.exe` to be specified in `install.cmd`
|
2017-11-13 09:45:19 +03:00
|
|
|
- Ensure persistence daemon/mode is enabled for GPUs
|
2017-11-10 20:22:18 +03:00
|
|
|
- Update dependencies to latest
|
2017-11-09 23:25:34 +03:00
|
|
|
|
2017-11-10 00:07:10 +03:00
|
|
|
### Fixed
|
2017-11-13 10:22:15 +03:00
|
|
|
- Non-Ubuntu/CentOS cascade failures from non-existent Singularity
|
2017-11-10 00:07:10 +03:00
|
|
|
- Default Singularity tagged image names on disk
|
2017-11-10 20:22:18 +03:00
|
|
|
- Circular dependency in `task_factory` and `settings`
|
2017-11-13 02:11:26 +03:00
|
|
|
- `misc tensorboard` command broken from latest TF image
|
2017-11-13 09:45:19 +03:00
|
|
|
- Update NV driver
|
2017-11-10 00:07:10 +03:00
|
|
|
|
2017-11-09 00:33:51 +03:00
|
|
|
## [3.0.0rc1] - 2017-11-08
|
2017-11-06 23:55:17 +03:00
|
|
|
### Changed
|
2017-11-08 23:23:23 +03:00
|
|
|
- Update VM size support
|
|
|
|
|
- SSH private key filemode check no longer results in an exception if it
|
|
|
|
|
fails. Instead a warning is issued - this is to allow SSH invocations on WSL.
|
|
|
|
|
- Install scripts now uninstall `azure-storage` first due to conflicts with
|
|
|
|
|
the Azure Storage Python split library.
|
2017-11-06 23:55:17 +03:00
|
|
|
- Updated to blobxfer 1.0.0
|
|
|
|
|
|
2017-11-06 19:12:49 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Job submission on custom image pools with 4.0 SDK changes
|
2017-11-08 23:23:23 +03:00
|
|
|
- Empty coordination command issue for Docker tasks
|
2017-11-08 05:16:30 +03:00
|
|
|
- Singularity registries with passwords in keyvault
|
2017-11-06 19:12:49 +03:00
|
|
|
|
2017-11-06 02:58:29 +03:00
|
|
|
## [3.0.0b1] - 2017-11-05
|
2017-10-29 19:38:46 +03:00
|
|
|
### Added
|
2017-11-06 02:58:29 +03:00
|
|
|
- Singularity support (#135)
|
|
|
|
|
- Preliminary Windows server support (#7)
|
|
|
|
|
- Pre-built binaries for CLI for some Linux distributions and Windows (#131)
|
2017-11-06 01:06:52 +03:00
|
|
|
- Windows Docker image for CLI
|
2017-11-06 02:58:29 +03:00
|
|
|
- Singularity HPCG and TensorFlow-GPU recipes
|
2017-10-29 19:38:46 +03:00
|
|
|
|
2017-11-01 23:27:14 +03:00
|
|
|
### Changed
|
2017-11-03 23:43:18 +03:00
|
|
|
- **Breaking Change:** Many commands have been placed under more
|
|
|
|
|
appropriate hierachies. Please see the major version migration guide for
|
|
|
|
|
more information.
|
2017-11-01 23:27:14 +03:00
|
|
|
|
2017-10-29 19:38:46 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Mount `/opt/intel` into Singularity containers
|
2017-10-29 21:13:32 +03:00
|
|
|
- Retry image configuration error pulls from Docker registries
|
2017-11-01 23:27:14 +03:00
|
|
|
- AAD MFA token cache on Python2
|
2017-11-05 22:48:08 +03:00
|
|
|
- Non-native coordination command fix, if not specified
|
2017-11-06 02:58:29 +03:00
|
|
|
- Include min node counts in autoscale scenarios (#139)
|
|
|
|
|
- `jobs tasks list` when there is a failed task (#142)
|
2017-10-29 19:38:46 +03:00
|
|
|
|
2017-10-27 21:35:48 +03:00
|
|
|
## [3.0.0a2] - 2017-10-27
|
2017-10-05 03:59:30 +03:00
|
|
|
### Added
|
2017-10-06 20:55:46 +03:00
|
|
|
- Major version migration guide (#134)
|
2017-10-05 03:59:30 +03:00
|
|
|
- Support for mounting multiple Azure File shares as `shared_data_volumes`
|
|
|
|
|
to a pool (#123)
|
2017-10-21 06:15:46 +03:00
|
|
|
- `bind_options` support for `data_volumes` and `shared_data_volumes`
|
2017-10-24 18:50:30 +03:00
|
|
|
- More packer samples for custom images
|
2017-10-27 21:35:48 +03:00
|
|
|
- Singularity HPLinpack recipe
|
2017-10-05 03:59:30 +03:00
|
|
|
|
|
|
|
|
### Changed
|
2017-10-17 22:57:02 +03:00
|
|
|
- **Breaking Change:** `global_resources`:`docker_volumes` is now named
|
|
|
|
|
`global_resources`:`volumes`. Although backward compatibility is maintained
|
|
|
|
|
for this property, it is recommended to migrate as volumes are now shared
|
|
|
|
|
between Docker and Singularity containers.
|
2017-10-05 03:59:30 +03:00
|
|
|
- Azure Files (with `volume_driver` of `azurefile`) specified under
|
|
|
|
|
`shared_data_volumes` are now mounted directly to the host (#123)
|
|
|
|
|
- The internal root mount point for all `shared_data_volumes` is now under
|
|
|
|
|
`$AZ_BATCH_NODE_ROOT_DIR/mounts` to reduce clutter/confusion under the
|
|
|
|
|
old root mount point of `$AZ_BATCH_NODE_SHARED_DIR`. The container mount
|
|
|
|
|
points (i.e., `container_path`) are unaffected.
|
2017-10-16 23:03:54 +03:00
|
|
|
- Canonical UbuntuServer 16.04-LTS is no longer pinned to a specific
|
|
|
|
|
release. Please avoid using the version `16.04.201709190`.
|
2017-10-27 21:35:48 +03:00
|
|
|
- Update to blobxfer 1.0.0rc3
|
2017-10-21 06:15:46 +03:00
|
|
|
- Updated custom image guide
|
2017-10-05 03:59:30 +03:00
|
|
|
|
2017-10-27 21:35:48 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Multi-instance Docker-based application command was not being launched
|
|
|
|
|
under a user identity if specified
|
|
|
|
|
- Allow min node allocation with `bias_last_sample` without required
|
|
|
|
|
sample percentage (#138)
|
|
|
|
|
|
2017-10-03 19:12:24 +03:00
|
|
|
## [3.0.0a1] - 2017-10-04
|
2017-09-26 21:08:37 +03:00
|
|
|
### Added
|
2017-09-29 08:13:14 +03:00
|
|
|
- Support for deploying compute nodes to an ARM Virtual Network with Batch
|
2017-10-03 03:29:55 +03:00
|
|
|
Service Batch accounts (#126)
|
|
|
|
|
- Support for deploying custom image compute nodes from an ARM Image resource
|
|
|
|
|
(#126)
|
|
|
|
|
- Support for multiple public and private container registries (#127)
|
2017-09-26 21:08:37 +03:00
|
|
|
- YAML configuration support. JSON formatted configuration files will continue
|
|
|
|
|
to be supported, however, note the breaking change with the corresponding
|
|
|
|
|
environment variable names for specifying individual config files from the
|
2017-10-03 19:12:24 +03:00
|
|
|
commandline. (#122)
|
2017-10-04 18:59:03 +03:00
|
|
|
- Option to automatically attempt recovery of unusable nodes during
|
2017-10-03 19:12:24 +03:00
|
|
|
pool allocation or resize. See the `attempt_recovery_on_unusable` option in
|
|
|
|
|
the pool configuration doc.
|
2017-09-29 04:39:15 +03:00
|
|
|
- Virtual Network guide
|
2017-09-26 21:08:37 +03:00
|
|
|
|
|
|
|
|
### Changed
|
2017-10-03 23:40:47 +03:00
|
|
|
- **Breaking Change:** Docker image tag for the CLI has been renamed to
|
2017-10-04 23:31:21 +03:00
|
|
|
`alfpark/batch-shipyard:<version>-cli` where `<version>` is the release
|
|
|
|
|
version or `latest` for whatever is in `master`. (#130)
|
2017-09-30 00:59:15 +03:00
|
|
|
- **Breaking Change:** Fully qualified Docker image names are now required
|
|
|
|
|
under both the global config `global_resources`.`docker_images` and jobs
|
2017-10-04 23:31:21 +03:00
|
|
|
`task` array `docker_image` (or `image`). The `docker_registry` property
|
|
|
|
|
in the global config file is no longer valid. (#106)
|
2017-09-30 00:59:15 +03:00
|
|
|
- **Breaking Change:** Docker private registries backed to Azure Storage blobs
|
2017-10-03 19:12:24 +03:00
|
|
|
are no longer supported. This is not to be confused with the Classic Azure
|
2017-10-03 03:29:55 +03:00
|
|
|
Container Registries which are still supported. (#44)
|
2017-09-30 00:59:15 +03:00
|
|
|
- **Breaking Change:** `docker_registry` property in the global config is
|
|
|
|
|
no longer required. An `additional_registries` option is available for any
|
|
|
|
|
additional registries that are not present from the `docker_images`
|
2017-10-06 20:55:46 +03:00
|
|
|
array in `global_resources` but require a valid login. (#106)
|
2017-10-03 03:29:55 +03:00
|
|
|
- **Breaking Change:** Data ingress/egress from/to Azure Storage along with
|
|
|
|
|
`task_factory`:`file` has changed to accommodate `blobxfer 1.0.0` commandline
|
|
|
|
|
and options. There are new expanded options available, including multiple
|
|
|
|
|
`include` and `exclude` along with `remote_path` explicit specifications
|
|
|
|
|
(instead of general `container` or `file_share`). Please see the appropriate
|
|
|
|
|
global config, pool or job configuration docs for more information. (#47)
|
|
|
|
|
- **Breaking Change:** `image_uris` in the `vm_configuration`:`custom_image`
|
|
|
|
|
property of the pool configuration has been replaced with `arm_image_id`
|
|
|
|
|
which is a reference to an ARM Image resource. Please see the custom image
|
2017-10-03 19:12:24 +03:00
|
|
|
guide for more information. (#126)
|
2017-09-26 21:08:37 +03:00
|
|
|
- **Breaking Change:** environment variables `SHIPYARD_CREDENTIALS_JSON`,
|
|
|
|
|
`SHIPYARD_CONFIG_JSON`, `SHIPYARD_POOL_JSON`, `SHIPYARD_JOBS_JSON`, and
|
|
|
|
|
`SHIPYARD_FS_JSON` have been renamed to `SHIPYARD_CREDENTIALS_CONF`,
|
|
|
|
|
`SHIPYARD_CONFIG_CONF`, `SHIPYARD_POOL_CONF`, `SHIPYARD_JOBS_CONF`, and
|
2017-10-03 19:12:24 +03:00
|
|
|
`SHIPYARD_FS_CONF` respectively. (#122)
|
2017-10-04 00:31:33 +03:00
|
|
|
- `--configdir` or `SHIPYARD_CONFIGDIR` now defaults to the current working
|
|
|
|
|
directory (i.e., `.`) if no other conf file options are specified.
|
2017-09-29 04:39:15 +03:00
|
|
|
- `aad` can be specified at a "global" level in the credentials configuration
|
|
|
|
|
file, which is then applied to `batch`, `keyvault` and/or `management`
|
|
|
|
|
section. Please see the credentials configuration guide for more information.
|
2017-09-29 08:13:14 +03:00
|
|
|
- `docker_image` is now preferred over the deprecated `image` property in
|
|
|
|
|
the `task` array in the jobs configuration file
|
2017-10-03 19:12:24 +03:00
|
|
|
- `gpu` and `infiniband` under the jobs configuration are now optional. GPU
|
|
|
|
|
and/or RDMA capable compute nodes will be autodetected and the proper
|
|
|
|
|
devices and other settings will be automatically be applied to tasks running
|
|
|
|
|
on these compute nodes. You can force disable GPU and/or RDMA by setting
|
|
|
|
|
`gpu` and `infiniband` properties to `false`. (#124)
|
2017-10-03 03:29:55 +03:00
|
|
|
- Update Docker CE to 17.09.0
|
2017-10-03 19:12:24 +03:00
|
|
|
- Update NC driver to 384.81 (CUDA 9.0 support)
|
2017-09-26 21:08:37 +03:00
|
|
|
|
2017-10-03 18:03:47 +03:00
|
|
|
## [2.9.6] - 2017-10-03
|
|
|
|
|
### Added
|
2019-02-27 01:30:09 +03:00
|
|
|
- Migrate to Read the Docs for [documentation](https://batch-shipyard.readthedocs.io/en/latest/)
|
2017-10-03 18:03:47 +03:00
|
|
|
|
2017-09-26 21:09:47 +03:00
|
|
|
### Fixed
|
|
|
|
|
- RemoteFS disk attach fixes
|
|
|
|
|
- Nvidia docker volume mount check
|
|
|
|
|
|
2017-09-25 05:50:14 +03:00
|
|
|
## [2.9.5] - 2017-09-24
|
2017-09-23 04:23:24 +03:00
|
|
|
### Added
|
|
|
|
|
- Optional `version` support for `platform_image`. This property can be
|
|
|
|
|
used to set a host OS version to prevent possible issues that occur with
|
|
|
|
|
`latest` image versions.
|
2017-09-25 05:50:14 +03:00
|
|
|
- `--all-starting` option for `pool delnode` which will delete all nodes
|
|
|
|
|
in starting state
|
2017-09-23 04:23:24 +03:00
|
|
|
|
2017-09-22 18:21:16 +03:00
|
|
|
### Changed
|
|
|
|
|
- Prevent invalid configuration of HPC offers with non-RDMA VM sizes
|
|
|
|
|
- Expanded network tuning exemptions for new Dv3 and Ev3 sizes
|
2017-09-23 04:23:24 +03:00
|
|
|
- Temporarily override Canonical UbuntuServer 16.04-LTS latest version to
|
2017-09-25 05:50:14 +03:00
|
|
|
a prior version due to recent linux-azure kernel issues
|
2017-09-22 18:21:16 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- NV driver updates
|
|
|
|
|
- Various OS updates and Docker issues
|
|
|
|
|
- CentOS 7.3 to 7.4 Nvidia driver breakage
|
|
|
|
|
- Regression in `pool ssh` on Windows
|
|
|
|
|
- Exception in unusable nodes with pool stats on allocation
|
2017-09-22 22:32:31 +03:00
|
|
|
- Handle package manager db locks during conflicts for local package installs
|
2017-09-22 18:21:16 +03:00
|
|
|
|
2017-09-12 18:56:03 +03:00
|
|
|
## [2.9.4] - 2017-09-12
|
|
|
|
|
### Changed
|
|
|
|
|
- Update dependencies to latest available
|
|
|
|
|
- Improve Docker builds
|
|
|
|
|
|
2017-08-31 20:28:22 +03:00
|
|
|
### Fixed
|
2017-09-12 18:56:03 +03:00
|
|
|
- Missing `join_by` function in blobxfer helper script (#115)
|
|
|
|
|
- Fix `clear()` for `pool udi` with Python 2.7 (#118)
|
2017-08-31 20:28:22 +03:00
|
|
|
|
2017-08-29 18:05:34 +03:00
|
|
|
## [2.9.3] - 2017-08-29
|
|
|
|
|
### Fixed
|
|
|
|
|
- Ignore `resize_timeout` for autoscale-enabled pools
|
|
|
|
|
- Present a warning for `jobs migrate` indicating Docker image requirements
|
|
|
|
|
- Various doc typos and updates
|
|
|
|
|
|
2017-08-16 18:59:53 +03:00
|
|
|
## [2.9.2] - 2017-08-16
|
2017-08-10 18:48:24 +03:00
|
|
|
### Added
|
2017-08-15 23:28:55 +03:00
|
|
|
- Deep learning Jupyter notebooks (thanks to @msalvaris and @thdeltei)
|
2017-08-10 18:48:24 +03:00
|
|
|
- Automatic site-extensions NuGet package updates with tagged releases via
|
|
|
|
|
AppVeyor builds
|
2017-08-15 23:28:55 +03:00
|
|
|
- Caffe2-CPU and Caffe2-GPU Recipes
|
2017-08-10 18:48:24 +03:00
|
|
|
|
2017-08-10 18:11:46 +03:00
|
|
|
### Changed
|
|
|
|
|
- Python 3.3 is no longer supported (due to `cryptography` dropping support
|
|
|
|
|
for 3.3).
|
2017-08-16 18:59:53 +03:00
|
|
|
- Use multi-stage build for Cascade to improve build times and reduce
|
|
|
|
|
Docker image size
|
2017-08-10 18:11:46 +03:00
|
|
|
|
2017-08-15 23:28:55 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Provide more helpful feedback for invalid clients
|
|
|
|
|
- Fix provisioning clusters with disks larger than 2TB
|
2017-08-16 18:59:53 +03:00
|
|
|
- RemoteFS issues in resize and expand
|
|
|
|
|
- Various site extension issues, will now proper install/upgrade to the
|
|
|
|
|
associated tagged version
|
2017-08-15 23:28:55 +03:00
|
|
|
|
2017-08-09 18:57:33 +03:00
|
|
|
## [2.9.0rc1] - 2017-08-09
|
2017-08-07 20:06:19 +03:00
|
|
|
### Added
|
2017-08-08 23:57:57 +03:00
|
|
|
- Recurring job support (job schedules). Please see jobs configuration doc
|
|
|
|
|
for more information.
|
2017-08-07 20:06:19 +03:00
|
|
|
- `custom` task factory. See the task factory guide for more information.
|
2017-08-08 23:57:57 +03:00
|
|
|
- `--all-jobschedules` option for `jobs term` and `jobs del`
|
|
|
|
|
- `--jobscheduleid` option for `jobs disable`, `jobs enable` and
|
|
|
|
|
`jobs migrate`
|
|
|
|
|
- `--all` option for `jobs listtasks`
|
2017-08-07 20:06:19 +03:00
|
|
|
|
2017-08-08 18:24:15 +03:00
|
|
|
### Changed
|
|
|
|
|
- `autogenerated_task_id_prefix` configuration setting is now named
|
|
|
|
|
`autogenerated_task_id` and is a complex property. It has member properties
|
|
|
|
|
named `prefix` and `zfill_width` to control how autogenerated task ids are
|
|
|
|
|
named.
|
2017-08-08 23:57:57 +03:00
|
|
|
- `jobs list` will now output job schedules in addition to jobs
|
|
|
|
|
- `--all` parameter for `jobs term` and `jobs del` renamed to `--all-jobs`
|
|
|
|
|
- list subcommands now output in a more human readable format
|
2017-08-08 18:24:15 +03:00
|
|
|
|
2017-08-05 00:51:27 +03:00
|
|
|
## [2.9.0b2] - 2017-08-04
|
2017-08-03 20:41:06 +03:00
|
|
|
### Added
|
2017-08-07 20:06:19 +03:00
|
|
|
- `random` and `file` task factories. See the task factory guide for more
|
2017-08-04 20:19:04 +03:00
|
|
|
information.
|
2017-08-05 00:47:16 +03:00
|
|
|
- Summary statistics: `pool stats` and `jobs stats`. See the usage doc for
|
|
|
|
|
more information.
|
2017-08-03 20:41:06 +03:00
|
|
|
- Delete unusable nodes from pool with `--all-unusable` option for
|
|
|
|
|
`pool delnode`
|
2017-08-07 20:06:19 +03:00
|
|
|
- CentOS-HPC 7.3 support
|
2017-08-03 21:46:28 +03:00
|
|
|
- CNTK-GPU-Infiniband-IntelMPI recipe
|
2017-08-03 20:41:06 +03:00
|
|
|
|
2017-08-03 01:11:20 +03:00
|
|
|
### Changed
|
|
|
|
|
- `remove_container_after_exit` now defaults to `true`
|
2017-08-04 20:19:04 +03:00
|
|
|
- `input_data`:`azure_storage` files with an include filter that does not
|
|
|
|
|
include wildcards (i.e., targets a single file) will now be placed at
|
|
|
|
|
the `destination` directly as specified.
|
2017-08-03 01:11:20 +03:00
|
|
|
- Nvidia Tesla driver updated to 384.59
|
|
|
|
|
- TensorFlow recipes updated for 1.2.1. TensorFlow-Distributed `launcher.sh`
|
|
|
|
|
script is now generalized to take a script as the first parameter and
|
|
|
|
|
relocated to `/shipyard/launcher.sh`.
|
2017-08-04 00:40:28 +03:00
|
|
|
- CNTK recipes updated for 2.1. `run_cntk.sh` script now takes in CNTK
|
|
|
|
|
Python scripts for execution.
|
2017-08-03 01:11:20 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Task termination with force failing due to new task generators
|
2017-08-03 18:51:29 +03:00
|
|
|
- pool udi over SSH terminal mangling
|
2017-08-03 01:11:20 +03:00
|
|
|
|
2017-08-01 01:05:58 +03:00
|
|
|
## [2.9.0b1] - 2017-07-31
|
2017-07-18 00:15:48 +03:00
|
|
|
### Added
|
2017-08-01 01:05:58 +03:00
|
|
|
- Autoscale support. Please see the Autoscale guide for more information.
|
|
|
|
|
- Autopool support
|
|
|
|
|
- Task factory and parametric sweep support. Please see the task factory
|
|
|
|
|
guide for more information.
|
|
|
|
|
- Job priority support
|
2017-08-08 23:57:57 +03:00
|
|
|
- Job migration support with new command `jobs migrate`
|
2017-08-01 01:05:58 +03:00
|
|
|
- Compute node fill type support
|
2017-08-05 00:47:16 +03:00
|
|
|
- New commands: `jobs enable` and `jobs disable`. Please see the usage doc
|
|
|
|
|
for more information.
|
2017-07-18 00:15:48 +03:00
|
|
|
- From Scratch: Step-by-Step guide
|
2017-08-01 01:05:58 +03:00
|
|
|
- Azure Cloud Shell information
|
2017-07-18 00:15:48 +03:00
|
|
|
|
2017-07-31 22:46:41 +03:00
|
|
|
### Changed
|
|
|
|
|
- Auto-generated task prefix is now `task-`. This can now be overridden with
|
|
|
|
|
the `autogenerated_task_id_prefix` property in the global configuration.
|
2017-08-01 01:05:58 +03:00
|
|
|
- Update dependencies to latest except for azure-mgmt-compute due to
|
|
|
|
|
broken changes
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- RemoteFS regressions
|
|
|
|
|
- Pool deletion with poolid argument cleanup
|
2017-07-31 22:46:41 +03:00
|
|
|
|
2017-07-06 21:12:24 +03:00
|
|
|
## [2.8.0] - 2017-07-06
|
2017-07-01 09:50:21 +03:00
|
|
|
### Added
|
2017-07-06 21:12:24 +03:00
|
|
|
- Support for CentOS 7.3 NC/NV gpu pools
|
2017-07-01 09:50:21 +03:00
|
|
|
- `--all-start-task-failed` parameter for `pool delnode`
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Improve robustness of docker image pulls within node prep scripts
|
|
|
|
|
- Restrict node list queries until pool allocation state emerges from resizing
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Remove nvidia gpu driver property from FFmpeg recipe
|
|
|
|
|
- Further improve retry logic for docker image pulls in cascade
|
|
|
|
|
|
2017-06-30 20:20:19 +03:00
|
|
|
## [2.8.0rc2] - 2017-06-30
|
2017-06-29 18:48:05 +03:00
|
|
|
### Added
|
|
|
|
|
- Support Mac OS X and Windows Subsystem for Linux installations via
|
2017-06-30 20:20:19 +03:00
|
|
|
`install.sh` (#101)
|
2017-06-29 18:48:05 +03:00
|
|
|
- Guide for Windows Subsystem for Linux installations
|
2017-06-30 20:20:19 +03:00
|
|
|
- Automated Nvidia driver install for NV-series
|
2017-06-29 18:48:05 +03:00
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Drop unsupported designations for Mac OS X and Windows
|
2017-06-30 20:20:19 +03:00
|
|
|
- Update Docker engine to 17.06 for Ubuntu, Debian, CentOS and 17.04 for
|
|
|
|
|
OpenSUSE
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Regression in private registry image pulls
|
2017-06-29 18:48:05 +03:00
|
|
|
|
2017-06-27 21:38:36 +03:00
|
|
|
## [2.8.0rc1] - 2017-06-27
|
2017-06-27 19:27:17 +03:00
|
|
|
### Added
|
|
|
|
|
- Version metadata added to pools and jobs with warnings generated for
|
|
|
|
|
mismatches (#89)
|
2017-06-27 20:12:56 +03:00
|
|
|
- Cloud shell installation support
|
2017-06-27 19:27:17 +03:00
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Update Docker images to Alpine 3.6 (#65)
|
|
|
|
|
- Improve robustness of package downloads
|
|
|
|
|
- Add retries for docker pull within cascade context
|
|
|
|
|
- Download cascade.log on start up failures
|
|
|
|
|
|
2017-06-09 21:34:26 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Patch job for auto completion (#97)
|
2017-06-27 19:27:17 +03:00
|
|
|
- Tensorboard command with custom images
|
2017-06-29 18:48:05 +03:00
|
|
|
- conda-forge detection in installation scripts (#100)
|
2017-06-09 21:34:26 +03:00
|
|
|
|
2017-06-07 18:30:04 +03:00
|
|
|
## [2.8.0b1] - 2017-06-07
|
2017-06-06 18:29:44 +03:00
|
|
|
### Added
|
|
|
|
|
- Custom image support, please see the pool configuration doc and custom
|
2017-06-27 19:27:17 +03:00
|
|
|
image guide for more information. (#94)
|
2017-06-06 18:29:44 +03:00
|
|
|
- `contrib` area with `packer` scripts
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- **Breaking Change:** `publisher`, `offer`, `sku` is now part of a complex
|
|
|
|
|
property named `vm_configuration`:`platform_image`. This change is to
|
|
|
|
|
accommodate custom images. The old configuration schema is now deprecated and
|
|
|
|
|
will be removed in a future release.
|
|
|
|
|
- Updated NVIDIA Tesla driver to 375.66
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Improved pool resize/allocation logic to fail early with low priority core
|
|
|
|
|
quota reached with no dedicated nodes
|
|
|
|
|
|
2017-05-31 17:43:14 +03:00
|
|
|
## [2.7.0] - 2017-05-31
|
2017-05-24 19:54:09 +03:00
|
|
|
### Added
|
|
|
|
|
- `--poolid` parameter for `pool del` to specify a specific pool to delete
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Prompt for confirmation for `jobs cmi`
|
|
|
|
|
- Updated to latest dependencies
|
2017-05-29 22:58:53 +03:00
|
|
|
- Split low-priority considerations into separate doc
|
2017-05-24 19:54:09 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Remote FS allocation issue with `vm_count` deprecation check
|
|
|
|
|
- Better handling of package index refresh errors
|
2017-05-31 17:43:14 +03:00
|
|
|
- `pool udi` over SSH issues (#92)
|
2017-05-29 00:57:52 +03:00
|
|
|
- Duplicate volume checks between job and task definitions
|
2017-05-24 19:54:09 +03:00
|
|
|
|
2017-05-24 17:43:19 +03:00
|
|
|
## [2.7.0rc1] - 2017-05-24
|
2017-05-23 05:45:25 +03:00
|
|
|
### Added
|
2017-05-24 00:14:29 +03:00
|
|
|
- `pool listimages` command which will list all common Docker images on
|
|
|
|
|
all nodes and provide warning for mismatched images amongst compute nodes.
|
|
|
|
|
This functionality requires a provisioned SSH user and private key.
|
2017-05-23 19:29:00 +03:00
|
|
|
- `max_wall_time` option for both jobs and tasks. Please consult the
|
|
|
|
|
documentation for the difference when specifying this option at either the
|
|
|
|
|
job or task level.
|
2017-05-23 05:45:25 +03:00
|
|
|
- `--poll-until-tasks-complete` option for `jobs listtasks` to block the CLI
|
2017-05-24 17:43:19 +03:00
|
|
|
from exiting until all tasks under jobs for which the command is run have
|
|
|
|
|
completed
|
2017-05-23 06:03:35 +03:00
|
|
|
- `--tty` option for `pool ssh` and `fs cluster ssh` to enable allocation
|
|
|
|
|
of a pseudo-tty for the SSH session
|
2017-05-23 05:45:25 +03:00
|
|
|
|
2017-05-23 19:29:00 +03:00
|
|
|
### Changed
|
|
|
|
|
- `remove_container_after_exit`, `retention_time`, `shm_size`, `infiniband`,
|
2017-05-24 17:43:19 +03:00
|
|
|
`gpu` can now be specified at the job-level and overriden at the task-level
|
2017-05-23 19:29:00 +03:00
|
|
|
in the jobs configuration
|
2017-05-24 17:43:19 +03:00
|
|
|
- `data_volumes` and `shared_data_volumes` can now be specified at the
|
|
|
|
|
job-level and any volumes specified at the task level will be *merged* with
|
|
|
|
|
the job-level volumes to be exposed for the container
|
2017-05-23 19:29:00 +03:00
|
|
|
|
2017-05-23 05:45:25 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Add missing deprecation path for `pool_specification_vm_count` for
|
|
|
|
|
multi-instance tasks. Please upgrade your jobs configuration to explicitly
|
|
|
|
|
use either `pool_specification_vm_count_dedicated` or
|
|
|
|
|
`pool_specification_vm_count_low_priority`.
|
|
|
|
|
- Speed up task collection additions by caching last task id
|
2017-05-24 00:14:29 +03:00
|
|
|
- Issues with pool resize and wait logic with low priority
|
2017-05-23 05:45:25 +03:00
|
|
|
|
2017-05-18 19:10:16 +03:00
|
|
|
## [2.7.0b2] - 2017-05-18
|
2017-05-13 05:23:29 +03:00
|
|
|
### Changed
|
2017-05-16 19:49:01 +03:00
|
|
|
- Allow the prior `vm_count` behavior, but provide a deprecation warning. The
|
|
|
|
|
old `vm_count` behavior will be removed in a future release. (#84)
|
2017-05-18 19:10:16 +03:00
|
|
|
- Add tasks via collection (#86)
|
2017-05-13 05:23:29 +03:00
|
|
|
- Log if node is dedicated in `pool listnodes`
|
|
|
|
|
- Updated all recipes with new `vm_count` changes
|
|
|
|
|
|
2017-05-16 19:49:01 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Improve pool resize wait logic for pools with mixed node types
|
2017-05-18 19:10:16 +03:00
|
|
|
- Do not override workdir if specified (#87)
|
|
|
|
|
- Prevent container scanning for data ingress from Azure Storage if include
|
|
|
|
|
filter contains no wildcards (#88)
|
2017-05-16 19:49:01 +03:00
|
|
|
|
2017-05-13 00:40:06 +03:00
|
|
|
## [2.7.0b1] - 2017-05-12
|
2017-05-08 19:05:22 +03:00
|
|
|
### Added
|
2018-03-01 00:10:14 +03:00
|
|
|
- Support for [Low Priority Batch Compute Nodes](https://docs.microsoft.com/azure/batch/batch-low-pri-vms)
|
2017-05-13 00:40:06 +03:00
|
|
|
- `resize_timeout` can now be specified on the pool specification
|
2017-05-08 19:05:22 +03:00
|
|
|
- `--clear-tables` option to `storage del` command which will delete
|
|
|
|
|
blob containers and queues but clear table entries
|
2017-05-13 00:40:06 +03:00
|
|
|
- `--ssh` option to `pool udi` command which will force the update Docker
|
|
|
|
|
images command to update over SSH instead of through a Batch job. This is
|
2017-05-13 05:23:29 +03:00
|
|
|
useful if you want to perform an out-of-band update of Docker image(s), e.g.,
|
2017-05-13 00:40:06 +03:00
|
|
|
your pool is currently busy processing tasks and would not be able to
|
|
|
|
|
accommodate another task.
|
2017-05-08 19:05:22 +03:00
|
|
|
|
2017-05-11 19:21:20 +03:00
|
|
|
### Changed
|
2017-05-13 00:40:06 +03:00
|
|
|
- **Breaking Change:** `vm_count` in the pool specification is now a
|
|
|
|
|
complex property consisting of the properties `dedicated` and `low_priority`
|
2017-05-11 19:21:20 +03:00
|
|
|
- Updated all dependencies to latest
|
|
|
|
|
|
2017-05-08 22:29:37 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Improve node startup time for GPU NC-series by removing extraneous
|
|
|
|
|
dependencies
|
2017-05-11 19:21:20 +03:00
|
|
|
- `fs cluster ssh` storage cluster id and command argument ordering was
|
|
|
|
|
inverted. This has been corrected to be as intended where the command
|
|
|
|
|
is the last argument, e.g., `fs cluster ssh mynfs -- df -h`
|
2017-05-08 22:29:37 +03:00
|
|
|
|
2017-05-05 18:42:53 +03:00
|
|
|
## [2.6.2] - 2017-05-05
|
|
|
|
|
### Added
|
|
|
|
|
- Docker image build for `develop` branch
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Allow NVIDIA license agreement to be auto-confirmed via `-y` option
|
|
|
|
|
- Use requests for file downloading since it is already being installed
|
|
|
|
|
as a dependency
|
|
|
|
|
- Update dependencies to latest versions
|
|
|
|
|
|
2017-05-02 23:16:57 +03:00
|
|
|
### Fixed
|
|
|
|
|
- TensorFlow image not being set if no suitable image is found for
|
|
|
|
|
`misc tensorboard` command
|
2017-05-03 06:11:24 +03:00
|
|
|
- Authentication for running images not present in global config sourced
|
|
|
|
|
from a private registry
|
2017-05-02 23:16:57 +03:00
|
|
|
|
2017-05-02 08:16:48 +03:00
|
|
|
## [2.6.1] - 2017-05-01
|
2017-04-29 02:11:54 +03:00
|
|
|
### Added
|
|
|
|
|
- `misc tensorboard` command added which automatically instantiates a
|
|
|
|
|
Tensorboard instance on the compute node which is running or has ran a
|
2017-04-30 08:16:35 +03:00
|
|
|
task that has generated TensorFlow summary operation compatible logs. An
|
2017-04-29 02:11:54 +03:00
|
|
|
SSH tunnel is then created so you can view Tensorboard locally on the
|
2017-05-02 08:16:48 +03:00
|
|
|
machine running Batch Shipyard. This requires a valid SSH user that has been
|
2017-04-30 08:16:35 +03:00
|
|
|
provisioned via Batch Shipyard with private keys available. This command
|
|
|
|
|
will work on Windows if `ssh.exe` is available in `%PATH%` or the current
|
|
|
|
|
working directory. Please see the usage guide for more information about
|
|
|
|
|
this command.
|
2017-05-01 19:43:41 +03:00
|
|
|
- Pool-level `resource_files` support
|
2017-04-29 02:11:54 +03:00
|
|
|
|
|
|
|
|
### Changed
|
2017-05-01 00:22:13 +03:00
|
|
|
- Added optional `COMMAND` argument to `pool ssh` and `fs cluster ssh`
|
|
|
|
|
commands. If `COMMAND` is specified, the command is run non-interactively
|
|
|
|
|
with SSH on the target node.
|
2017-04-29 02:11:54 +03:00
|
|
|
- Added some additional sanity checks in the node prep script
|
|
|
|
|
- Updated TensorFlow-CPU and TensorFlow-GPU recipes to 1.1.0. Removed
|
|
|
|
|
specialized Docker build for TensorFlow-GPU. Added `jobs-tb.json` files
|
|
|
|
|
to TensorFlow-CPU and TensorFlow-GPU recipes as Tensorboard samples.
|
2017-05-02 08:16:48 +03:00
|
|
|
- Optimize some Batch calls
|
2017-04-29 02:11:54 +03:00
|
|
|
|
2017-04-22 05:22:03 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Site extension issues
|
2017-04-29 02:11:54 +03:00
|
|
|
- SSH user add exception on Windows
|
2017-04-30 08:16:35 +03:00
|
|
|
- `jobs del --termtasks` will now disable the job prior to running task
|
|
|
|
|
termination to prevent active tasks in job from running while tasks are
|
|
|
|
|
being terminated
|
|
|
|
|
- `jobs listtasks` and `data listfiles` will now accept a `--jobid` that
|
|
|
|
|
does not have to be in `jobs.json`
|
2017-05-01 18:38:45 +03:00
|
|
|
- Data ingress on pool create issue with single node
|
2017-04-20 19:45:37 +03:00
|
|
|
|
|
|
|
|
## [2.6.0] - 2017-04-20
|
|
|
|
|
### Changed
|
|
|
|
|
- Update to latest dependencies
|
|
|
|
|
|
2017-04-18 23:51:05 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Checks that prevented ssh/scp/openssl interaction on Windows
|
2017-04-20 19:45:37 +03:00
|
|
|
- SSH private key regression in data ingress direct to compute node
|
2017-04-13 23:13:06 +03:00
|
|
|
|
2017-04-14 22:54:43 +03:00
|
|
|
## [2.6.0rc1] - 2017-04-14
|
2017-04-13 19:31:35 +03:00
|
|
|
### Added
|
|
|
|
|
- Richer SSH options with new `ssh_public_key_data` and `ssh_private_key`
|
|
|
|
|
properties in `ssh` configuration blocks (for both `pool.json` and
|
|
|
|
|
`fs.json`).
|
|
|
|
|
- `ssh_public_key_data` allows direct embedding of SSH public keys in
|
|
|
|
|
OpenSSH format into the config files.
|
|
|
|
|
- `ssh_private_key` specifies where the private key is located with
|
|
|
|
|
respect to pre-created public keys (either `ssh_public_key` or
|
|
|
|
|
`ssh_public_key_data`). This allows transparent `pool ssh` or
|
|
|
|
|
`fs cluster ssh` commands with pre-created keys.
|
2017-04-13 23:13:06 +03:00
|
|
|
- RemoteFS-GlusterFS+BatchPool recipe
|
|
|
|
|
|
|
|
|
|
### Changed
|
2017-04-14 22:54:43 +03:00
|
|
|
- Docker installations are now pinned to a specific Docker version which
|
|
|
|
|
should reduce sudden breaking changes introduced upstream by Docker and/or
|
|
|
|
|
the distribution
|
2017-04-13 23:13:06 +03:00
|
|
|
- Fault domains for multi-vm storage clusters are now set to 2 by default but
|
|
|
|
|
can be configured using the `fault_domains` property. This was lowered from
|
|
|
|
|
the prior default of 3 due to managed disks and availability set restrictions
|
|
|
|
|
as some regions do not support 3 fault domains with this combination.
|
2017-04-14 22:54:43 +03:00
|
|
|
- Updated NC-series Tesla driver to 375.51
|
2017-04-13 19:31:35 +03:00
|
|
|
|
2017-04-05 21:30:00 +03:00
|
|
|
### Fixed
|
2017-04-14 22:54:43 +03:00
|
|
|
- Broken Docker installations due to gpgkey changes
|
2017-04-05 21:30:00 +03:00
|
|
|
- Possible race condition between disk setup and glusterfs volume create
|
2017-04-13 19:31:35 +03:00
|
|
|
- Forbid SSH username to be the same as the samba username
|
|
|
|
|
- Allow smbd.service to auto-restart with delay
|
2017-04-13 23:13:06 +03:00
|
|
|
- Data ingress to glusterfs on compute with no remotefs settings
|
2017-04-03 19:29:23 +03:00
|
|
|
|
2017-04-18 23:51:05 +03:00
|
|
|
### Removed
|
|
|
|
|
- Host support for OpenSUSE 13.2 and SLES 12
|
|
|
|
|
|
2017-04-03 19:29:23 +03:00
|
|
|
## [2.6.0b3] - 2017-04-03
|
2017-03-24 01:01:20 +03:00
|
|
|
### Added
|
2017-03-28 23:51:55 +03:00
|
|
|
- Created [Azure App Service Site Extension](https://www.siteextensions.net/packages/batch-shipyard).
|
|
|
|
|
You can now one-click install Batch Shipyard as a site extension (after you
|
2017-04-04 17:26:32 +03:00
|
|
|
have Python installed) and use Batch Shipyard from an Azure Function trigger.
|
2017-03-30 23:18:47 +03:00
|
|
|
- Samba support on storage cluster servers
|
2017-03-24 01:01:20 +03:00
|
|
|
- Add sample RemoteFS recipes for NFS and GlusterFS
|
2017-04-01 07:57:41 +03:00
|
|
|
- `install.cmd` installer for Windows. `install_conda_windows.cmd` has been
|
|
|
|
|
replaced by `install.cmd`, please see the install doc for more information.
|
2017-03-24 01:01:20 +03:00
|
|
|
|
2017-03-29 05:58:55 +03:00
|
|
|
### Changed
|
2017-04-03 20:47:52 +03:00
|
|
|
- **Breaking Change:** `multi_instance_auto_complete` under
|
|
|
|
|
`job_specifications` is now named `auto_complete`. This property will apply
|
|
|
|
|
to all types of jobs and not just multi-instance tasks. The default is now
|
|
|
|
|
`false` (instead of `true` for the old `multi_instance_auto_complete`).
|
2017-03-29 05:58:55 +03:00
|
|
|
- **Breaking Change:** `static_public_ip` has been replaced with a `public_ip`
|
2017-04-04 17:26:32 +03:00
|
|
|
complex property. This is to accommodate for situations where public IP for
|
2017-03-29 05:58:55 +03:00
|
|
|
RemoteFS is disabled. Please see the Remote FS configuration doc for more
|
|
|
|
|
info.
|
2017-04-01 07:57:41 +03:00
|
|
|
- `install.sh` now handles Anaconda Python environments
|
2017-03-31 23:21:22 +03:00
|
|
|
- `--cardinal 0` is now implicit if no `--hostname` or `--nodeid` is specified
|
|
|
|
|
for `fs cluster ssh` or `pool ssh` commands, respectively
|
2017-04-04 00:16:21 +03:00
|
|
|
- Allow `docker_images` in `global_resources` to be empty. Note that it is
|
|
|
|
|
always recommended to pre-load images on to pools for consistent scheduling
|
|
|
|
|
latencies from pool idle.
|
2017-03-29 05:58:55 +03:00
|
|
|
|
2017-03-24 01:01:20 +03:00
|
|
|
### Fixed
|
2017-03-24 03:37:53 +03:00
|
|
|
- Removed requirement of a `batch` credential section for pure `fs` operations
|
2017-03-28 23:51:55 +03:00
|
|
|
- Multi-instance auto complete setting not being properly read
|
2017-03-29 18:56:19 +03:00
|
|
|
- `install.sh` virtual environment issues
|
2017-03-30 23:18:47 +03:00
|
|
|
- Fix pool ingress data calls with remotefs (#62)
|
2017-03-31 01:15:19 +03:00
|
|
|
- Move additional node prep commands to last set of commands to execute in
|
|
|
|
|
start task (#63)
|
2017-03-31 18:07:23 +03:00
|
|
|
- `glusterfs_on_compute` shared data volume issues
|
2017-03-31 19:55:34 +03:00
|
|
|
- future and pathlib compat issues
|
2017-03-31 23:21:22 +03:00
|
|
|
- Python2 unicode/str issues with management libraries
|
2017-03-22 19:53:30 +03:00
|
|
|
|
|
|
|
|
## [2.6.0b2] - 2017-03-22
|
2017-03-18 01:48:20 +03:00
|
|
|
### Added
|
|
|
|
|
- Added virtual environment install option for `install.sh` which is now
|
|
|
|
|
the recommended way to install Batch Shipyard. Please see the install
|
|
|
|
|
guide for more information. (#55)
|
|
|
|
|
|
2017-03-22 19:53:30 +03:00
|
|
|
### Changed
|
|
|
|
|
- Force SSD optimizations for btrfs with premium storage
|
|
|
|
|
|
2017-03-18 01:48:20 +03:00
|
|
|
### Fixed
|
2017-03-22 19:53:30 +03:00
|
|
|
- Incorrect FS server options parsing at script time
|
2017-03-18 01:48:20 +03:00
|
|
|
- KeyVault client not initialized in `fs` contexts (#57)
|
2017-03-22 19:53:30 +03:00
|
|
|
- Check pool current node count prior to executing `pool udi` task (#58)
|
|
|
|
|
- Initialization with KeyVault uri on commandline (#59)
|
2017-03-17 03:25:40 +03:00
|
|
|
|
|
|
|
|
## [2.6.0b1] - 2017-03-16
|
2017-03-09 20:40:16 +03:00
|
|
|
### Added
|
2017-03-10 01:38:16 +03:00
|
|
|
- Support for provisioning storage clusters via the `fs cluster` command
|
2017-03-11 09:54:16 +03:00
|
|
|
- Support for NFS (single VM, scale up)
|
2017-03-16 18:58:53 +03:00
|
|
|
- Support for GlusterFS (multi VM, scale up and out)
|
2017-03-10 01:38:16 +03:00
|
|
|
- Support for provisioning managed disks via the `fs disks` command
|
2017-03-12 21:14:11 +03:00
|
|
|
- Support for data ingress to provisioned storage clusters
|
2017-03-22 19:53:30 +03:00
|
|
|
- Support for
|
2018-03-01 00:10:14 +03:00
|
|
|
[UserSubscription Batch accounts](https://docs.microsoft.com/azure/batch/batch-account-create-portal#user-subscription-mode)
|
2017-03-09 20:40:16 +03:00
|
|
|
- Azure Active Directory authentication support for Batch accounts
|
2017-03-12 21:14:11 +03:00
|
|
|
- Support for specifying a virtual network to use with a compute pool
|
2017-04-04 17:26:32 +03:00
|
|
|
- `allow_run_on_missing_image` option to jobs that allows tasks to execute
|
|
|
|
|
under jobs with Docker images that have not been pre-loaded via the
|
2017-03-10 01:38:16 +03:00
|
|
|
`global_resources`:`docker_images` setting in config.json. Note that, if
|
|
|
|
|
possible, you should attempt to specify all Docker images that you intend
|
|
|
|
|
to run in the `global_resources`:`docker_images` property in the global
|
|
|
|
|
configuration to minimize scheduling to task execution latency.
|
2017-03-17 20:00:58 +03:00
|
|
|
- Support for running containers as a different user identity (uid/gid)
|
2017-03-11 09:54:16 +03:00
|
|
|
- Support for Canonical/UbuntuServer/16.04-LTS. 16.04-LTS should be used over
|
2017-03-10 01:38:16 +03:00
|
|
|
the old 16.04.0-LTS sku due to
|
2017-03-11 09:54:16 +03:00
|
|
|
[issue #31](https://github.com/Azure/batch-shipyard/issues/31) and is no
|
|
|
|
|
longer receiving updates.
|
2017-03-09 20:40:16 +03:00
|
|
|
|
2017-03-01 23:45:32 +03:00
|
|
|
### Changed
|
2017-03-09 07:18:58 +03:00
|
|
|
- **Breaking Change:** `glusterfs` `volume_driver` for `shared_data_volumes`
|
2017-03-11 09:54:16 +03:00
|
|
|
should now be named as `glusterfs_on_compute`. This is to distinguish between
|
2017-03-16 18:58:53 +03:00
|
|
|
co-located GlusterFS on compute nodes with standalone GlusterFS
|
2017-03-11 09:54:16 +03:00
|
|
|
`storage_cluster` remote mounted distributed file system.
|
|
|
|
|
- Logging now has less verbose details (call origin) by default. Prior
|
|
|
|
|
behavior can be restored with the `-v` option.
|
2017-03-10 01:38:16 +03:00
|
|
|
- Pool existance is now checked prior to job submission and can now proceed
|
|
|
|
|
to add without an active pool.
|
|
|
|
|
- Batch `account` (name) is now an optional property in the credentials config
|
|
|
|
|
- Configuration doc broken up into multiple pages
|
|
|
|
|
- Update all recipes using Canonical/UbuntuServer/16.04.0-LTS to use
|
|
|
|
|
Canonical/UbuntuServer/16.04-LTS instead
|
2017-03-17 03:25:40 +03:00
|
|
|
- Configuration is no longer shown with `-v`. Use `--show-config` to dump
|
|
|
|
|
the complete configuration being used for the command.
|
2017-03-11 09:54:16 +03:00
|
|
|
- Precompile Python files during build for Docker images
|
2017-03-01 23:45:32 +03:00
|
|
|
- All dependencies updated to latest versions
|
|
|
|
|
- Update Batch API call compatibility for `azure-batch 2.0.0`
|
2017-02-23 19:06:27 +03:00
|
|
|
|
2017-03-11 09:54:16 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Logging time format and incorrect Zulu time designation.
|
2017-03-11 20:21:33 +03:00
|
|
|
- `scp` and `multinode_scp` data movement capability is now supported in
|
|
|
|
|
Windows given `ssh.exe` and `scp.exe` can be found in `%PATH%` or the current
|
|
|
|
|
working directory. `rsync` methods are not supported on Windows.
|
|
|
|
|
- Credential encryption is now supported in Windows given `openssl.exe` can
|
|
|
|
|
be found in `%PATH%` or the current working directory.
|
2017-03-11 09:54:16 +03:00
|
|
|
|
2017-03-08 19:39:51 +03:00
|
|
|
## [2.5.4] - 2017-03-08
|
|
|
|
|
### Changed
|
|
|
|
|
- Downloaded files are now verified via SHA256 instead of MD5
|
|
|
|
|
- Updated NC-series Tesla driver to 375.39
|
|
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- `nvidia-docker` updated to 1.0.1 for compatibility with Docker CE
|
|
|
|
|
|
2017-03-01 18:39:00 +03:00
|
|
|
## [2.5.3] - 2017-03-01
|
2017-02-28 20:41:05 +03:00
|
|
|
### Added
|
2017-03-01 06:34:22 +03:00
|
|
|
- `pool rebootnode` command added which allows single node reboot control.
|
2017-03-01 23:45:32 +03:00
|
|
|
Additionally, the option `--all-start-task-failed` will reboot all nodes in
|
2017-03-01 06:34:22 +03:00
|
|
|
the specified pool with the start task failed state.
|
2017-02-28 20:41:05 +03:00
|
|
|
- `jobs del` and `jobs term` now provide a `--termtasks` option to
|
|
|
|
|
allow the logic of `jobs termtasks` to precede the delete or terminate
|
|
|
|
|
action to the job. This option requires a valid SSH user to the remote nodes
|
|
|
|
|
as specified in the `ssh` configuration property in `pool.json`. This new
|
|
|
|
|
option is normally not needed if all tasks within the jobs have completed.
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- The Docker image used for blobxfer is now tied to the specific Batch
|
|
|
|
|
Shipyard release
|
|
|
|
|
- Default SSH user expiry time if not specified is now 30 days
|
|
|
|
|
- All recipes now have the default config.json storage account set to the
|
2017-03-01 23:45:32 +03:00
|
|
|
link as named in the provided credentials.json file. Now, only the credentials
|
|
|
|
|
file needs to be modified to run a recipe.
|
2017-02-28 20:41:05 +03:00
|
|
|
|
2017-02-23 19:06:27 +03:00
|
|
|
## [2.5.2] - 2017-02-23
|
2017-02-23 06:23:51 +03:00
|
|
|
### Added
|
|
|
|
|
- Chainer-CPU and Chainer-GPU recipes
|
2017-02-23 07:25:17 +03:00
|
|
|
- [Troubleshooting guide](docs/96-troubleshooting-guide.md)
|
2017-02-23 06:23:51 +03:00
|
|
|
|
2017-02-18 00:14:00 +03:00
|
|
|
### Changed
|
2017-02-23 09:11:27 +03:00
|
|
|
- Perform automatic container path substitution with host path for
|
|
|
|
|
GlusterFS data ingress/egress from/to Azure Storage (#37)
|
2017-02-18 00:14:00 +03:00
|
|
|
- Allow NAMD-TCP recipe to be run on a single node
|
|
|
|
|
|
|
|
|
|
### Fixed
|
2017-02-23 19:06:27 +03:00
|
|
|
- CNTK-GPU-OpenMPI run script fixed to allow multinode+singlegpu executions
|
|
|
|
|
- TensorFlow recipes updated for 1.0.0 release
|
2017-02-23 06:23:51 +03:00
|
|
|
- blobxfer data ingress on Windows (#39)
|
2017-02-23 07:25:17 +03:00
|
|
|
- Minor delete job and terminate tasks fixes
|
2017-02-01 22:06:26 +03:00
|
|
|
|
|
|
|
|
## [2.5.1] - 2017-02-01
|
2017-01-25 00:54:26 +03:00
|
|
|
### Added
|
|
|
|
|
- Support for max task retries (#23). See configuration doc for more
|
|
|
|
|
information.
|
2017-01-31 20:31:21 +03:00
|
|
|
- Support for task data retention time (#30). See configuration doc for
|
|
|
|
|
more information.
|
2017-01-25 00:54:26 +03:00
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- **Breaking Change:** `environment_variables_secret_id` was erroneously
|
|
|
|
|
named and has been renamed to `environment_variables_keyvault_secret_id` to
|
|
|
|
|
follow the other properties with similar behavior.
|
|
|
|
|
- Include Python 3.6 Travis CI target
|
|
|
|
|
|
|
|
|
|
### Fixed
|
2017-01-26 01:49:23 +03:00
|
|
|
- Automatically assigned task ids are now in the format `dockertask-NNNNN`
|
|
|
|
|
and will increment properly past 99999 but will not be padded after that (#27)
|
2017-02-01 22:06:26 +03:00
|
|
|
- Defect in list tasks for tasks that have not run (#28)
|
2017-01-25 00:54:26 +03:00
|
|
|
- Docker temporary directory not being set properly
|
2017-01-26 21:44:43 +03:00
|
|
|
- SLES-HPC will now install all Intel MPI related rpms
|
2017-01-31 02:05:03 +03:00
|
|
|
- Defect in task file mover for unencrypted credentials (#29)
|
2017-01-19 21:15:32 +03:00
|
|
|
|
|
|
|
|
## [2.5.0] - 2017-01-19
|
2017-01-12 20:23:25 +03:00
|
|
|
### Added
|
|
|
|
|
- Support for
|
2018-03-01 00:10:14 +03:00
|
|
|
[Task Dependency Id Ranges](https://docs.microsoft.com/azure/batch/batch-task-dependencies#task-id-range)
|
2017-01-12 20:23:25 +03:00
|
|
|
with the `depends_on_range` property under each task json property in `tasks`
|
|
|
|
|
in the jobs configuration file. Please see the configuration doc for more
|
|
|
|
|
information.
|
2017-01-19 21:15:32 +03:00
|
|
|
- Support for `environment_variables_secret_id` in job and task definitions.
|
|
|
|
|
Specifying these properties will fetch manually added secrets (in the form of
|
|
|
|
|
a string representation of a json key-value dictionary) from the specified
|
|
|
|
|
KeyVault using AAD credentials. Please see the configuration doc for more
|
|
|
|
|
information.
|
2017-01-03 19:51:19 +03:00
|
|
|
|
2017-01-19 09:03:26 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Remove extraneous import (#12)
|
|
|
|
|
- Defect in handling per key secret ids (#13)
|
|
|
|
|
- Defect in environment variable dict merge (#17)
|
2017-01-19 21:15:32 +03:00
|
|
|
- Update Nvidia Docker to 1.0.0 (#21)
|
2017-01-19 09:03:26 +03:00
|
|
|
|
2017-01-11 20:14:24 +03:00
|
|
|
## [2.4.0] - 2017-01-11
|
|
|
|
|
### Added
|
|
|
|
|
- Support for credentials stored in Azure KeyVault
|
|
|
|
|
- `keyvault` command added. Please see the usage doc for more information.
|
|
|
|
|
- `*_keyvault_secret_id` properties added for keys and passwords in
|
|
|
|
|
credentials json. Please see the configuration doc for more information.
|
|
|
|
|
- Using Azure KeyVault with Batch Shipyard guide
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Updated NC-series Tesla driver to 375.20
|
|
|
|
|
|
2017-01-03 19:51:19 +03:00
|
|
|
## [2.3.1] - 2017-01-03
|
2016-12-15 22:04:33 +03:00
|
|
|
### Added
|
|
|
|
|
- Add support for nvidia-docker with ssh docker tunnel
|
2016-12-09 22:32:05 +03:00
|
|
|
|
2017-01-03 19:51:19 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Fix multi-job bug with jpcmd
|
|
|
|
|
|
2016-12-15 06:01:50 +03:00
|
|
|
## [2.3.0] - 2016-12-15
|
|
|
|
|
### Added
|
|
|
|
|
- `pool ssh` command. Please see the usage doc for more information.
|
|
|
|
|
- `shm_size` json property added to the json object within the `tasks` array
|
|
|
|
|
of a job. Please see the configuration doc for more information.
|
2017-01-03 19:51:19 +03:00
|
|
|
- SSH, Interactive Sessions and Docker SSH Tunnel guide
|
2016-12-15 06:01:50 +03:00
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- Improve usability of the generated SSH docker tunnel script
|
|
|
|
|
|
2016-12-09 22:32:05 +03:00
|
|
|
## [2.2.0] - 2016-12-09
|
|
|
|
|
### Added
|
|
|
|
|
- CNTK-CPU-Infiniband-IntelMPI recipe
|
|
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- `/opt/intel` is now automatically mounted once again for infiniband-enabled
|
|
|
|
|
containers on SUSE SLES-HPC hosts.
|
|
|
|
|
|
2016-12-01 01:35:44 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Fix masked KeyErrors on `input_data` and `output_data`
|
|
|
|
|
- Fix SAS key generation for data movement
|
|
|
|
|
- Typo in ssh public key check on Windows prevented pool add actions
|
|
|
|
|
- Pin version of tfm docker image on data transfers
|
2016-11-30 19:12:07 +03:00
|
|
|
|
|
|
|
|
## [2.1.0] - 2016-11-30
|
2016-11-28 20:47:02 +03:00
|
|
|
### Added
|
|
|
|
|
- Allow `--configdir`, `--credentials`, `--config`, `--jobs`, `--pool` config
|
|
|
|
|
options to be specified as environment variables. Please see the usage doc
|
|
|
|
|
for more information.
|
2016-11-30 01:58:52 +03:00
|
|
|
- Added subcommand `listskus` to the `pool` command to list available
|
|
|
|
|
VM configurations (publisher, offer, sku) for the Batch account
|
2016-11-23 20:06:37 +03:00
|
|
|
|
2016-11-30 19:12:07 +03:00
|
|
|
### Changed
|
|
|
|
|
- Nodeprep now references cascade and tfm docker images by version instead
|
|
|
|
|
of latest to prevent breaking changes affecting older versions. Docker builds
|
|
|
|
|
of cascade and tfm based on latest commits are now disabled.
|
|
|
|
|
|
2016-11-29 01:14:44 +03:00
|
|
|
### Fixed
|
2016-11-30 19:12:07 +03:00
|
|
|
- Cascade docker image run not propagating exit code
|
2016-11-29 01:14:44 +03:00
|
|
|
|
2016-11-23 20:06:37 +03:00
|
|
|
## [2.0.0] - 2016-11-23
|
2016-11-19 19:54:52 +03:00
|
|
|
### Added
|
|
|
|
|
- Support for any Internet accessible container registry, including
|
2018-03-01 00:10:14 +03:00
|
|
|
[Azure Container Registry](https://azure.microsoft.com/services/container-registry/).
|
2016-11-19 21:39:58 +03:00
|
|
|
Please see the configuration doc for information on how to integrate with
|
|
|
|
|
a private container registry.
|
|
|
|
|
|
2016-11-20 12:55:35 +03:00
|
|
|
### Changed
|
|
|
|
|
- GPU driver for `STANDARD_NC` instances defined in the
|
|
|
|
|
`gpu`:`nvidia_driver`:`source` property is no longer required. If omitted,
|
|
|
|
|
an NVIDIA driver will be downloaded automatically with an NVIDIA License
|
|
|
|
|
agreement prompt. For `STANDARD_NV` instances, a driver URL is still required.
|
2016-11-21 01:55:36 +03:00
|
|
|
- Docker container name auto-tagging now prepends the job id in order to
|
|
|
|
|
prevent conflicts in case of un-named simultaneous tasks from multiple jobs
|
2016-11-22 10:56:07 +03:00
|
|
|
- Update CNTK docker images to 2.0beta4 and optimize GPU images for use
|
|
|
|
|
with NVIDIA K80/M60
|
2016-11-21 20:03:35 +03:00
|
|
|
- Update Caffe docker image, default to using OpenBLAS over ATLAS, and
|
|
|
|
|
optimize GPU images for use with NVIDIA K80/M60
|
2016-11-23 01:27:33 +03:00
|
|
|
- Update MXNet GPU docker image optimized for use with NVIDIA K80/M60
|
2016-11-21 20:03:35 +03:00
|
|
|
- Update TensorFlow docker images to 0.11.0 and optimize GPU images for use
|
|
|
|
|
with NVIDIA K80/M60
|
2016-11-20 12:55:35 +03:00
|
|
|
|
2016-11-19 21:39:58 +03:00
|
|
|
### Fixed
|
|
|
|
|
- Cascade thread exceptions will terminate with non-zero exit code
|
|
|
|
|
- Some improvements with node prep and reboots
|
2016-11-21 01:55:36 +03:00
|
|
|
- Task termination will only issue `docker rm` if the container exists
|
2016-11-14 21:54:23 +03:00
|
|
|
|
|
|
|
|
## [2.0.0rc3] - 2016-11-14 (SC16 Edition)
|
2016-11-04 17:25:55 +03:00
|
|
|
### Added
|
|
|
|
|
- `install_conda_windows.cmd` helper script for installing Batch Shipyard
|
|
|
|
|
under Anaconda for Windows
|
2016-11-07 21:23:43 +03:00
|
|
|
- Added `relative_destination_path` json property for `files` ingress into
|
2016-11-10 02:37:55 +03:00
|
|
|
node destinations. This allows arbitrary specification of where ingressed
|
|
|
|
|
files should be placed relative to the destination path.
|
|
|
|
|
- Added ability to ingress directly into the host without the requirement
|
|
|
|
|
of GlusterFS for pools with one compute node. A GlusterFS shared volume is
|
|
|
|
|
required for pools with more than one compute node for direct to pool data
|
|
|
|
|
ingress.
|
2016-11-12 23:35:56 +03:00
|
|
|
- New commands and options:
|
2016-11-09 22:59:01 +03:00
|
|
|
- `pool udi`: Update docker images on all compute nodes in a pool. `--image`
|
|
|
|
|
and `--digest` options can restrict the scope of the update.
|
2016-11-04 17:25:55 +03:00
|
|
|
- `data stream`: `--disk` will stream the file as binary to disk instead
|
|
|
|
|
of as text to the local console
|
2016-11-09 02:06:40 +03:00
|
|
|
- `data listfiles`: `--jobid` and `--taskid` allows scoping of the list
|
|
|
|
|
files action
|
2016-11-09 22:59:01 +03:00
|
|
|
- `jobs listtasks`: `--jobid` allows scoping of list tasks to a specific job
|
2016-11-13 09:31:15 +03:00
|
|
|
- `jobs add`: `--tail` allows tailing the specified file for the last job
|
|
|
|
|
and task added
|
2016-11-13 12:43:01 +03:00
|
|
|
- Keras+Theano-CPU and Keras+Theano-GPU recipes
|
|
|
|
|
- Keras+Theano-CPU added as an option in the quickstart guide
|
2016-11-02 18:26:39 +03:00
|
|
|
|
2016-11-07 21:23:43 +03:00
|
|
|
### Changed
|
2016-11-10 20:48:00 +03:00
|
|
|
- **Breaking Change:** Properties of `docker_registry` have changed
|
2016-11-14 21:54:23 +03:00
|
|
|
significantly to support eventual integration with the Azure Container
|
|
|
|
|
Registry service. Credentials for docker logins have moved to the credentials
|
|
|
|
|
json file. Please see the configuration doc for more information.
|
2016-11-10 02:37:55 +03:00
|
|
|
- `files` data ingress no longer creates a directory where files to
|
|
|
|
|
be uploaded exist. For example if uploading from a path `/a/b/c`, the
|
|
|
|
|
directory `c` is no longer created at the destination. Instead all files
|
|
|
|
|
found in `/a/b/c` will be immediately placed directly at the destination
|
|
|
|
|
path with sub-directories preserved. This behavior can be modified with
|
|
|
|
|
the `relative_destination_path` property.
|
2016-11-13 22:35:09 +03:00
|
|
|
- `CUDA_CACHE_*` variables are now set for GPU jobs such that compiled targets
|
|
|
|
|
pass-through to the host. This allows subsequent container invocations within
|
|
|
|
|
the same node the ability to reuse cached PTX JIT targets.
|
2016-11-12 08:08:58 +03:00
|
|
|
- `batch_shipyard`:`storage_entity_prefix` is now optional and defaults to
|
|
|
|
|
`shipyard` if not specified.
|
2016-11-12 23:35:56 +03:00
|
|
|
- Major internal configuration/settings refactor
|
2016-11-07 21:23:43 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
2016-11-08 05:34:35 +03:00
|
|
|
- Pool resize down with wait
|
2016-11-07 21:23:43 +03:00
|
|
|
- More Python2/3 compatibility issues
|
2016-11-08 05:34:35 +03:00
|
|
|
- Ensure pools that deploy GlusterFS volumes have more than 1 node
|
2016-11-07 21:23:43 +03:00
|
|
|
|
2016-11-02 18:26:39 +03:00
|
|
|
## [2.0.0rc2] - 2016-11-02
|
2016-10-30 00:37:28 +03:00
|
|
|
### Added
|
|
|
|
|
- `install.sh` install/setup helper script
|
|
|
|
|
- `shipyard` execution helper script created via `install.sh`
|
2016-10-31 18:45:05 +03:00
|
|
|
- `generated_sas_expiry_days` json property to config json for the ability to
|
|
|
|
|
override the default number of days generated SAS keys are valid for.
|
2016-11-01 09:12:14 +03:00
|
|
|
- New options on commands/subcommands:
|
|
|
|
|
- `jobs add`: `--recreate` recreate any jobs which have completed and use
|
|
|
|
|
the same id
|
|
|
|
|
- `jobs termtasks`: `--force` force docker kill to tasks even if they are
|
|
|
|
|
in completed state
|
|
|
|
|
- `pool resize`: `--wait` wait for completion of resize
|
2016-11-02 18:26:39 +03:00
|
|
|
- HPCG-Infiniband-IntelMPI and HPLinpack-Infiniband-IntelMPI recipes
|
2016-10-30 00:37:28 +03:00
|
|
|
|
2016-10-29 21:00:31 +03:00
|
|
|
### Changed
|
2016-10-31 18:45:05 +03:00
|
|
|
- Default SAS expiry time used for resource files and data movement changed
|
|
|
|
|
from 7 to 30 days.
|
2016-10-29 21:00:31 +03:00
|
|
|
- Pools failing to start will now automatically retrieve stdout.txt and
|
2016-10-31 18:45:05 +03:00
|
|
|
stderr.txt to the current working directory under
|
|
|
|
|
`poolid/<node ids>/std{out,err}.txt`. These files can be inspected
|
|
|
|
|
locally and submitted as context for GitHub issues if pertinent.
|
2016-11-02 03:10:03 +03:00
|
|
|
- Pool resizing will now attempt to add an SSH user on the new nodes if
|
|
|
|
|
an SSH public key is referenced or found in the invocation directory
|
2016-10-30 00:37:28 +03:00
|
|
|
- Improve installation doc
|
2016-10-29 21:00:31 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
2016-10-31 18:45:05 +03:00
|
|
|
- Improve Python2/3 compatibility
|
2016-10-29 21:00:31 +03:00
|
|
|
- Unicode literals warning with Click
|
|
|
|
|
- Config file loading issue in some contexts
|
|
|
|
|
- Documentation typos
|
2016-10-28 20:16:57 +03:00
|
|
|
|
|
|
|
|
## [2.0.0rc1] - 2016-10-28
|
2016-10-06 21:03:10 +03:00
|
|
|
### Added
|
2016-10-20 20:48:25 +03:00
|
|
|
- Comprehensive data movement support. Please see the data movement guide
|
|
|
|
|
and configuration doc for more information.
|
2016-10-14 23:59:19 +03:00
|
|
|
- Ingress from local machine with `files` in global configuration
|
|
|
|
|
- To GlusterFS shared volume
|
|
|
|
|
- To Azure Blob Storage
|
2016-10-15 23:56:56 +03:00
|
|
|
- To Azure File Storage
|
2016-10-21 07:18:31 +03:00
|
|
|
- Ingress from Azure Blob Storage, Azure File Storage, or another Azure
|
|
|
|
|
Batch Task with `input_data` in pool and jobs configuration
|
2016-10-14 23:59:19 +03:00
|
|
|
- Pool-level: to compute nodes
|
2016-10-17 00:42:25 +03:00
|
|
|
- Job-level: to compute nodes prior to running the specified job
|
|
|
|
|
- Task-level: to compute nodes prior to running a task of a job
|
2016-10-14 23:59:19 +03:00
|
|
|
- Egress to local machine as actions
|
|
|
|
|
- Single file from compute node
|
2016-10-16 00:30:44 +03:00
|
|
|
- Entire task-level directories from compute node
|
2016-11-02 03:10:03 +03:00
|
|
|
- Entire node-level directories from compute node
|
2016-10-17 00:42:25 +03:00
|
|
|
- Egress to Azure Blob of File Storage with `output_data` in jobs
|
|
|
|
|
configuration
|
2016-10-20 07:10:40 +03:00
|
|
|
- Task-level: to Azure Blob or File Storage on successful completion of a
|
2016-10-17 00:42:25 +03:00
|
|
|
task
|
2016-10-20 20:48:25 +03:00
|
|
|
- Credential encryption support. Please see the credential encryption guide
|
|
|
|
|
and configuration doc for more information.
|
2017-02-01 22:06:26 +03:00
|
|
|
- Experimental support for OpenSSH with HPN patches on Ubuntu
|
2016-10-24 19:12:24 +03:00
|
|
|
- Support pool resize up with GlusterFS
|
2016-10-25 08:12:58 +03:00
|
|
|
- Support GlusterFS volume options
|
2016-10-28 06:51:36 +03:00
|
|
|
- Configurable path to place files generated by `pool add` or `pool asu`
|
|
|
|
|
commands
|
2016-10-25 08:12:58 +03:00
|
|
|
- MXNet-CPU and Torch-CPU as options in the quickstart guide
|
2016-10-23 05:27:48 +03:00
|
|
|
- Update CNTK recipes for 1.7.2 and switch multinode/multigpu samples to
|
|
|
|
|
MNIST
|
2016-10-25 08:12:58 +03:00
|
|
|
- MXNet-CPU and MXNet-GPU recipes
|
2016-10-06 21:03:10 +03:00
|
|
|
|
|
|
|
|
### Changed
|
2016-10-27 07:39:00 +03:00
|
|
|
- **Breaking Change:** All new CLI experience with proper multilevel commands.
|
|
|
|
|
Please see usage doc for more information.
|
2016-10-28 06:51:36 +03:00
|
|
|
- Added new commands: `cert`, `data`
|
|
|
|
|
- Added many new convenience subcommands
|
2016-10-28 20:16:57 +03:00
|
|
|
- `--filespec` is now delimited by `,` instead of `:`
|
2016-10-06 21:03:10 +03:00
|
|
|
- **Breaking Change:** `ssh_docker_tunnel` in the `pool_specification` has
|
2016-10-16 03:54:16 +03:00
|
|
|
been replaced by the `ssh` property. `generate_tunnel_script` has been renamed
|
|
|
|
|
to `generate_docker_tunnel_script`. Please see the configuration doc for
|
2016-10-06 21:03:10 +03:00
|
|
|
more information.
|
2016-10-28 09:41:47 +03:00
|
|
|
- The `name` property of a task json object in the jobs specification is no
|
|
|
|
|
longer required for multi-instance tasks. If not specified, `name` defaults
|
|
|
|
|
to `id` for all task types.
|
|
|
|
|
- `data stream` no longer has an arbitrary max streaming time; the action will
|
2016-10-09 21:37:29 +03:00
|
|
|
stream the file indefinitely until the task completes
|
2016-10-20 07:10:40 +03:00
|
|
|
- Validate container with `storage_entity_prefix` for length issues
|
2016-11-02 03:10:03 +03:00
|
|
|
- `pool del` action now cleans up and deletes some storage containers
|
2016-10-13 20:53:53 +03:00
|
|
|
immediately afterwards (with confirmation prompts)
|
2016-10-25 08:12:58 +03:00
|
|
|
- `/opt/intel` is no longer automatically mounted for infiniband-enabled
|
2016-11-02 03:10:03 +03:00
|
|
|
containers on SUSE SLES-HPC hosts. Please see the configuration doc
|
2016-10-25 08:12:58 +03:00
|
|
|
on how to manually map this directory if required. OpenLogic CentOS-HPC
|
|
|
|
|
hosts remain unchanged.
|
2016-10-16 03:54:16 +03:00
|
|
|
- Modularized code base
|
2016-10-06 21:03:10 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- GlusterFS mount ownership/permissions fixed such that SSH users can
|
|
|
|
|
read/write
|
2016-10-13 20:53:53 +03:00
|
|
|
- Azure File shared volume setup when invoked from Windows
|
2016-10-20 07:10:40 +03:00
|
|
|
- Python2 compatibility issues with file encoding
|
2016-10-21 07:18:31 +03:00
|
|
|
- Allow shipyard.py to be invoked outside of the root of the GitHub cloned
|
|
|
|
|
base directory
|
|
|
|
|
- TensorFlow-Distributed recipe issues
|
2016-10-05 19:20:00 +03:00
|
|
|
|
|
|
|
|
## [1.1.0] - 2016-10-05
|
2016-09-26 21:17:50 +03:00
|
|
|
### Added
|
2016-10-04 06:02:29 +03:00
|
|
|
- Transparent Infiniband assist for SUSE SLES-HPC 12-SP1 image
|
2016-10-05 19:20:00 +03:00
|
|
|
- Add version for shipyard.py script
|
2016-10-04 06:02:29 +03:00
|
|
|
- NAMD-GPU, OpenFOAM-Infiniband-IntelMPI, Torch-CPU, Torch-GPU recipes
|
2016-09-29 06:50:53 +03:00
|
|
|
|
|
|
|
|
### Changed
|
|
|
|
|
- GlusterFS mountpoint is now within `$AZ_BATCH_NODE_SHARED_DIR` so files can
|
|
|
|
|
be viewed/downloaded with Batch APIs
|
|
|
|
|
- NAMD-Infiniband-IntelMPI recipe now contains a real Docker image link
|
2016-09-22 23:39:18 +03:00
|
|
|
|
2016-09-30 05:15:38 +03:00
|
|
|
### Fixed
|
|
|
|
|
- GlusterFS not properly starting on Ubuntu
|
|
|
|
|
|
2016-09-22 23:39:18 +03:00
|
|
|
## [1.0.0] - 2016-09-22
|
2016-09-09 06:15:11 +03:00
|
|
|
### Added
|
2016-09-15 22:47:43 +03:00
|
|
|
- Automated GlusterFS support
|
2016-09-17 00:02:25 +03:00
|
|
|
- Added `configdir` argument for convenience in loading configuration files,
|
|
|
|
|
please see the usage documentation for more details
|
2016-09-09 06:15:11 +03:00
|
|
|
- Ability to retrieve files from live compute nodes in addition to streaming
|
2016-09-17 08:59:50 +03:00
|
|
|
- Added `filespec` argument for non-interactive `streamfile` and `gettaskfile`
|
|
|
|
|
actions
|
2016-09-09 06:15:11 +03:00
|
|
|
- Added .gitattributes to designate Unix line-endings for text files
|
2016-09-09 23:42:33 +03:00
|
|
|
- Sample configuration files for each recipe
|
2016-09-22 23:39:18 +03:00
|
|
|
- Caffe-CPU, OpenFOAM-TCP-OpenMPI, TensorFlow-CPU, TensorFlow-Distributed
|
|
|
|
|
recipes
|
2016-09-09 06:15:11 +03:00
|
|
|
|
|
|
|
|
### Changed
|
2016-09-15 22:47:43 +03:00
|
|
|
- Updated configuration docs to detail which properties are required vs. those
|
|
|
|
|
that are optional
|
2016-09-13 21:39:05 +03:00
|
|
|
- SSH tunnel user is now added with a default expiry time of 7 days which can
|
|
|
|
|
be modified through the pool configuration file
|
2016-09-16 21:38:09 +03:00
|
|
|
- Configuration is not output to console by default, `-v` flag added for
|
|
|
|
|
verbose output
|
2016-09-17 08:20:07 +03:00
|
|
|
- Determinstic remote login settings output (node, ip, port) that can be
|
|
|
|
|
easily parsed
|
2016-09-21 18:30:58 +03:00
|
|
|
- Update Azurefile Docker Volume Driver plugin to 0.5.1
|
2016-08-31 19:38:33 +03:00
|
|
|
|
2016-09-09 23:42:33 +03:00
|
|
|
### Fixed
|
2016-09-16 21:38:09 +03:00
|
|
|
- Cascade (container-only) start issue with no private registry
|
2016-09-09 23:42:33 +03:00
|
|
|
- Non-shipyard docker image node prep with new azure-storage package
|
2016-09-17 03:54:49 +03:00
|
|
|
- Inter-node communication not specified key error on addpool
|
2016-09-16 23:00:05 +03:00
|
|
|
- Cross-platform fixes:
|
|
|
|
|
- Temp file creation used for environment variables
|
|
|
|
|
- SSH tunnel creation disabled on Windows if public key is not supplied
|
2016-09-17 08:59:50 +03:00
|
|
|
- Batch Shipyard Docker container not getting cleaned up if peer-to-peer is
|
|
|
|
|
disabled
|
2016-09-09 23:42:33 +03:00
|
|
|
|
2016-09-09 07:06:21 +03:00
|
|
|
### Removed
|
2016-09-09 23:42:33 +03:00
|
|
|
- `gpu`:`nvidia_driver`:`version` property removed from pool configuration
|
2016-09-13 21:39:05 +03:00
|
|
|
and is no longer required as the version is now automatically detected
|
2016-09-09 07:06:21 +03:00
|
|
|
|
2016-09-08 07:16:27 +03:00
|
|
|
## [0.2.0] - 2016-09-08
|
2016-09-03 01:02:07 +03:00
|
|
|
### Added
|
|
|
|
|
- Transparent GPU support for Azure N-Series VMs
|
2016-09-08 07:16:27 +03:00
|
|
|
- New recipes added: Caffe-GPU, CNTK-CPU-OpenMPI, CNTK-GPU-OpenMPI,
|
2016-09-08 21:03:40 +03:00
|
|
|
FFmpeg-GPU, NAMD-Infiniband-IntelMPI, NAMD-TCP, TensorFlow-GPU
|
2016-09-03 01:02:07 +03:00
|
|
|
|
|
|
|
|
### Changed
|
2016-09-06 20:32:42 +03:00
|
|
|
- Multi-instance tasks now automatically complete their job by default. This
|
|
|
|
|
removes the need to run the `cleanmijobs` action in the shipyard tool.
|
|
|
|
|
Please refer to the
|
|
|
|
|
[multi-instance documentation](docs/80-batch-shipyard-multi-instance-tasks.md)
|
|
|
|
|
for more information and limitations.
|
2016-09-03 01:02:07 +03:00
|
|
|
- Dumb back-off policy for DHT router convergence
|
2016-09-08 07:16:27 +03:00
|
|
|
- Optimzed Docker image storage location for Azure VMs
|
|
|
|
|
- Prompts added for destructive operations in the shipyard tool
|
2016-09-03 01:02:07 +03:00
|
|
|
|
|
|
|
|
### Fixed
|
|
|
|
|
- Incorrect file location of node prep finished
|
|
|
|
|
- Blocking wait for global resource on pool can now be disabled
|
2016-09-08 07:16:27 +03:00
|
|
|
- Incorrect process call to query for docker image size when peer-to-peer
|
|
|
|
|
transfer is disabled
|
|
|
|
|
- Use azure-storage 0.33.0 to fix Edm.Int64 overflow issue
|
2016-09-03 01:02:07 +03:00
|
|
|
|
2016-09-01 07:43:03 +03:00
|
|
|
## [0.1.0] - 2016-09-01
|
2016-08-31 19:38:33 +03:00
|
|
|
#### Added
|
|
|
|
|
- Initial release
|
|
|
|
|
|
2019-12-14 00:32:26 +03:00
|
|
|
[Unreleased]: https://github.com/Azure/batch-shipyard/compare/3.9.1...HEAD
|
|
|
|
|
[3.9.1]: https://github.com/Azure/batch-shipyard/compare/3.9.0...3.9.1
|
2019-11-15 22:08:52 +03:00
|
|
|
[3.9.0]: https://github.com/Azure/batch-shipyard/compare/3.8.2...3.9.0
|
2019-09-12 19:29:08 +03:00
|
|
|
[3.8.2]: https://github.com/Azure/batch-shipyard/compare/3.8.1...3.8.2
|
2019-08-19 19:24:33 +03:00
|
|
|
[3.8.1]: https://github.com/Azure/batch-shipyard/compare/3.8.0...3.8.1
|
2019-08-14 05:19:48 +03:00
|
|
|
[3.8.0]: https://github.com/Azure/batch-shipyard/compare/3.7.1...3.8.0
|
2019-07-24 05:49:38 +03:00
|
|
|
[3.7.1]: https://github.com/Azure/batch-shipyard/compare/3.7.0...3.7.1
|
2019-03-01 00:10:14 +03:00
|
|
|
[3.7.0]: https://github.com/Azure/batch-shipyard/compare/3.6.1...3.7.0
|
2018-12-03 20:03:24 +03:00
|
|
|
[3.6.1]: https://github.com/Azure/batch-shipyard/compare/3.6.0...3.6.1
|
2018-11-07 01:21:50 +03:00
|
|
|
[3.6.0]: https://github.com/Azure/batch-shipyard/compare/3.6.0b1...3.6.0
|
2018-09-20 22:45:01 +03:00
|
|
|
[3.6.0b1]: https://github.com/Azure/batch-shipyard/compare/3.6.0a1...3.6.0b1
|
2018-08-06 20:35:31 +03:00
|
|
|
[3.6.0a1]: https://github.com/Azure/batch-shipyard/compare/3.5.3...3.6.0a1
|
2018-07-31 20:28:30 +03:00
|
|
|
[3.5.3]: https://github.com/Azure/batch-shipyard/compare/3.5.2...3.5.3
|
2018-07-20 19:13:44 +03:00
|
|
|
[3.5.2]: https://github.com/Azure/batch-shipyard/compare/3.5.1...3.5.2
|
2018-07-17 23:16:23 +03:00
|
|
|
[3.5.1]: https://github.com/Azure/batch-shipyard/compare/3.5.0...3.5.1
|
2018-06-29 23:24:27 +03:00
|
|
|
[3.5.0]: https://github.com/Azure/batch-shipyard/compare/3.5.0b3...3.5.0
|
2018-06-13 23:25:15 +03:00
|
|
|
[3.5.0b3]: https://github.com/Azure/batch-shipyard/compare/3.5.0b2...3.5.0b3
|
2018-06-12 02:43:21 +03:00
|
|
|
[3.5.0b2]: https://github.com/Azure/batch-shipyard/compare/3.5.0b1...3.5.0b2
|
2018-05-02 17:48:54 +03:00
|
|
|
[3.5.0b1]: https://github.com/Azure/batch-shipyard/compare/3.4.0...3.5.0b1
|
2018-03-26 23:15:44 +03:00
|
|
|
[3.4.0]: https://github.com/Azure/batch-shipyard/compare/3.3.0...3.4.0
|
2018-03-01 19:25:49 +03:00
|
|
|
[3.3.0]: https://github.com/Azure/batch-shipyard/compare/3.2.0...3.3.0
|
2018-02-21 19:26:13 +03:00
|
|
|
[3.2.0]: https://github.com/Azure/batch-shipyard/compare/3.1.0...3.2.0
|
2018-01-30 22:21:22 +03:00
|
|
|
[3.1.0]: https://github.com/Azure/batch-shipyard/compare/3.0.3...3.1.0
|
2018-01-22 19:39:12 +03:00
|
|
|
[3.0.3]: https://github.com/Azure/batch-shipyard/compare/3.0.2...3.0.3
|
2018-01-12 20:06:57 +03:00
|
|
|
[3.0.2]: https://github.com/Azure/batch-shipyard/compare/3.0.1...3.0.2
|
2017-11-22 19:30:13 +03:00
|
|
|
[3.0.1]: https://github.com/Azure/batch-shipyard/compare/3.0.0...3.0.1
|
2017-11-13 09:45:19 +03:00
|
|
|
[3.0.0]: https://github.com/Azure/batch-shipyard/compare/3.0.0rc1...3.0.0
|
2017-11-09 00:33:51 +03:00
|
|
|
[3.0.0rc1]: https://github.com/Azure/batch-shipyard/compare/3.0.0b1...3.0.0rc1
|
2017-11-06 02:58:29 +03:00
|
|
|
[3.0.0b1]: https://github.com/Azure/batch-shipyard/compare/3.0.0a2...3.0.0b1
|
2017-10-27 21:35:48 +03:00
|
|
|
[3.0.0a2]: https://github.com/Azure/batch-shipyard/compare/3.0.0a1...3.0.0a2
|
2017-10-03 19:12:24 +03:00
|
|
|
[3.0.0a1]: https://github.com/Azure/batch-shipyard/compare/2.9.6...3.0.0a1
|
2017-10-03 18:03:47 +03:00
|
|
|
[2.9.6]: https://github.com/Azure/batch-shipyard/compare/2.9.5...2.9.6
|
2017-09-25 05:50:14 +03:00
|
|
|
[2.9.5]: https://github.com/Azure/batch-shipyard/compare/2.9.4...2.9.5
|
2017-09-12 18:56:03 +03:00
|
|
|
[2.9.4]: https://github.com/Azure/batch-shipyard/compare/2.9.3...2.9.4
|
2017-08-29 18:05:34 +03:00
|
|
|
[2.9.3]: https://github.com/Azure/batch-shipyard/compare/2.9.2...2.9.3
|
2017-08-16 18:59:53 +03:00
|
|
|
[2.9.2]: https://github.com/Azure/batch-shipyard/compare/2.9.0rc1...2.9.2
|
2017-08-09 18:57:33 +03:00
|
|
|
[2.9.0rc1]: https://github.com/Azure/batch-shipyard/compare/2.9.0b2...2.9.0rc1
|
2017-08-05 00:51:27 +03:00
|
|
|
[2.9.0b2]: https://github.com/Azure/batch-shipyard/compare/2.9.0b1...2.9.0b2
|
2017-08-01 01:05:58 +03:00
|
|
|
[2.9.0b1]: https://github.com/Azure/batch-shipyard/compare/2.8.0...2.9.0b1
|
2017-07-06 21:12:24 +03:00
|
|
|
[2.8.0]: https://github.com/Azure/batch-shipyard/compare/2.8.0rc2...2.8.0
|
2017-06-30 20:20:19 +03:00
|
|
|
[2.8.0rc2]: https://github.com/Azure/batch-shipyard/compare/2.8.0rc1...2.8.0rc2
|
2017-06-27 21:38:36 +03:00
|
|
|
[2.8.0rc1]: https://github.com/Azure/batch-shipyard/compare/2.8.0b1...2.8.0rc1
|
2017-06-07 18:30:04 +03:00
|
|
|
[2.8.0b1]: https://github.com/Azure/batch-shipyard/compare/2.7.0...2.8.0b1
|
2017-05-31 17:43:14 +03:00
|
|
|
[2.7.0]: https://github.com/Azure/batch-shipyard/compare/2.7.0rc1...2.7.0
|
2017-05-24 17:43:19 +03:00
|
|
|
[2.7.0rc1]: https://github.com/Azure/batch-shipyard/compare/2.7.0b2...2.7.0rc1
|
2017-05-18 19:10:16 +03:00
|
|
|
[2.7.0b2]: https://github.com/Azure/batch-shipyard/compare/2.7.0b1...2.7.0b2
|
2017-05-13 00:40:06 +03:00
|
|
|
[2.7.0b1]: https://github.com/Azure/batch-shipyard/compare/2.6.2...2.7.0b1
|
2017-05-05 18:42:53 +03:00
|
|
|
[2.6.2]: https://github.com/Azure/batch-shipyard/compare/2.6.1...2.6.2
|
2017-05-02 08:16:48 +03:00
|
|
|
[2.6.1]: https://github.com/Azure/batch-shipyard/compare/2.6.0...2.6.1
|
2017-04-20 19:45:37 +03:00
|
|
|
[2.6.0]: https://github.com/Azure/batch-shipyard/compare/2.6.0rc1...2.6.0
|
2017-04-14 22:54:43 +03:00
|
|
|
[2.6.0rc1]: https://github.com/Azure/batch-shipyard/compare/2.6.0b3...2.6.0rc1
|
2017-04-03 19:29:23 +03:00
|
|
|
[2.6.0b3]: https://github.com/Azure/batch-shipyard/compare/2.6.0b2...2.6.0b3
|
2017-03-22 19:53:30 +03:00
|
|
|
[2.6.0b2]: https://github.com/Azure/batch-shipyard/compare/2.6.0b1...2.6.0b2
|
2017-03-17 03:25:40 +03:00
|
|
|
[2.6.0b1]: https://github.com/Azure/batch-shipyard/compare/2.5.4...2.6.0b1
|
2017-03-08 20:09:34 +03:00
|
|
|
[2.5.4]: https://github.com/Azure/batch-shipyard/compare/2.5.3...2.5.4
|
2017-02-28 20:41:05 +03:00
|
|
|
[2.5.3]: https://github.com/Azure/batch-shipyard/compare/2.5.2...2.5.3
|
2017-02-23 19:06:27 +03:00
|
|
|
[2.5.2]: https://github.com/Azure/batch-shipyard/compare/2.5.1...2.5.2
|
2017-02-01 22:06:26 +03:00
|
|
|
[2.5.1]: https://github.com/Azure/batch-shipyard/compare/2.5.0...2.5.1
|
2017-01-19 21:15:32 +03:00
|
|
|
[2.5.0]: https://github.com/Azure/batch-shipyard/compare/2.4.0...2.5.0
|
2017-01-11 20:14:24 +03:00
|
|
|
[2.4.0]: https://github.com/Azure/batch-shipyard/compare/2.3.1...2.4.0
|
2017-01-03 21:04:40 +03:00
|
|
|
[2.3.1]: https://github.com/Azure/batch-shipyard/compare/2.3.0...2.3.1
|
2016-12-15 06:01:50 +03:00
|
|
|
[2.3.0]: https://github.com/Azure/batch-shipyard/compare/2.2.0...2.3.0
|
2016-12-09 22:32:05 +03:00
|
|
|
[2.2.0]: https://github.com/Azure/batch-shipyard/compare/2.1.0...2.2.0
|
2016-11-30 19:12:07 +03:00
|
|
|
[2.1.0]: https://github.com/Azure/batch-shipyard/compare/2.0.0...2.1.0
|
2016-11-23 20:06:37 +03:00
|
|
|
[2.0.0]: https://github.com/Azure/batch-shipyard/compare/2.0.0rc3...2.0.0
|
2016-11-14 21:54:23 +03:00
|
|
|
[2.0.0rc3]: https://github.com/Azure/batch-shipyard/compare/2.0.0rc2...2.0.0rc3
|
2016-11-02 18:26:39 +03:00
|
|
|
[2.0.0rc2]: https://github.com/Azure/batch-shipyard/compare/2.0.0rc1...2.0.0rc2
|
2016-10-28 20:16:57 +03:00
|
|
|
[2.0.0rc1]: https://github.com/Azure/batch-shipyard/compare/1.1.0...2.0.0rc1
|
2016-10-05 19:20:00 +03:00
|
|
|
[1.1.0]: https://github.com/Azure/batch-shipyard/compare/1.0.0...1.1.0
|
2016-09-22 23:39:18 +03:00
|
|
|
[1.0.0]: https://github.com/Azure/batch-shipyard/compare/0.2.0...1.0.0
|
2016-09-03 01:02:07 +03:00
|
|
|
[0.2.0]: https://github.com/Azure/batch-shipyard/compare/0.1.0...0.2.0
|
2016-09-01 19:40:24 +03:00
|
|
|
[0.1.0]: https://github.com/Azure/batch-shipyard/compare/ab1fa4d...0.1.0
|
