Fetching contributors…
Cannot retrieve contributors at this time
287 lines (154 sloc) 7.36 KB
ssbench (0.3.9)
* Fix #123: ssbench should detect closed connections and not reuse them.
* Fix #124: update the README instructions for using ssbench with
* virtualenv.
* Fix #117: ssbench cleanup-containers needs to be aware of the policy
suffix (otherwise no containers are cleaned up).
* Fix #96: when reporting average time per request, if no time elapsed
(happens if --noop was specified), ssbench should not print a
DivideByZero exception. The behavior is changed to print "inf" in such
ssbench (0.3.8)
* Added IPv6 support for the master and workers.
* Added errors, retries and retry_rate to cvs output.
* Removed the statlib dependency.
* Updated README to encourage using virtualenv.
* Updated AUHORS file.
ssbench (0.2.23)
* Misc test fixes/improvements including Travis CI integration.
* Improved Python 2.6 support.
* Fixed bug in batch job generation when job decoration fails.
* Worker job exceptions no longer count toward request counts.
* Updated AUTHORS file.
ssbench (0.2.22)
* Always print the distribution of requests per worker-ID in the report.
* Allow ssbench-master to override the default work block size used for
GET/PUT operations (-b BYTES or --block-size BYTES)
* Document the container_base key in the scenario file and allow a
non-default value for that to be passed into the cleanup-containers
ssbench (0.2.21)
* Updated AUTHORS file (sorry guys!)
* Fixed #66: Add cleanup-containers command to ssbench-master to delete all
ssbench-created containers and objects.
* Fixed #65: Only create initial files if they're not already present (try
to HEAD them first)
ssbench (0.2.20)
* Pick a storage URL for each request (better multiple -S support)
* Lowered default ssbench-worker concurrency level to 64
ssbench (0.2.19)
* Calculate some requests-per-worker-id stats (printed at DEBUG level)
* Juggle requirements files around a bit to be more standard.
ssbench (0.2.18)
* Bind to with --workers and no --bind-ip specified.
ssbench (0.2.17)
* Added retries/errors in reporting [Kota Tsuyuzaki]
* Fixed #40: Allow load-balancing via multiple -S options.
* Fixed #48: Don't crash in reporting when zero jobs succeeded.
ssbench (0.2.16)
* Fixed another bug in ssbench-master kill-workers
ssbench (0.2.15)
* Fixed bug in ssbench-master kill-workers
ssbench (0.2.14)
* Fixed #62: Support benchmark results >> size of RAM.
* Fixed #64: Include more parameters in default results filename.
* Fixed #56: Allow specification of run-time instead of op count.
* Fixed #50: Make ssbench output include version number.
* Fixed #69: Fixes SSL issues with ssbench. [Sergio Rubio]
* Fixed worker issues when using SSL and --insecure [Sergio Rubio]
* Added experimental --csv option (CSV reporting output)
ssbench (0.2.13)
* Fixed a few bugs/typos.
ssbench (0.2.12)
* Fixed #46: Use blocking calls where possible.
* Fixed #47: Allow batching of work jobs.
ssbench (0.2.11)
* Fixed #26: Add socket timeouts.
ssbench (0.2.10)
* Fixed #42: Support Keystone v2.0 auth.
* Fixed #39: Never set nofile less than 1024.
* Fixed #34: Never use concurrency < 1 for workers.
* Fixed #38: Allow -S while still authenticating.
* Fixed #32: Save results gzipped.
ssbench (0.2.9)
* Fixed #25: ssbench-worker will re-auth if necessary.
* Fixed #21; added ability to kill workers.
* Fixed #16: raise file descriptor limit if run as root.
* Fixed bug when using the scenario's user count.
ssbench (0.2.8)
* Actually install the sample scenario files, CHANGELOG, AUTHORS, and
LICENSE as "data files"
* Tweak CentOS install instructions to install gevent-zeormq as its own
ssbench (0.2.4)
* Fixed #29: Added support for Python 2.6 and cleaned up CentOS install
* Fixed #22: Added -k/--keep-objects option to keep all created objects
from a benchmark run.
ssbench (0.2.3)
* Update install docs with stupid Ubuntu install work-around.
ssbench (0.2.2)
* Try pinning pyzmq to 2.2.0 to fix reported install problems
ssbench (0.2.1)
* Fixed install bug by removing ssbench-qdrain from
ssbench (0.2.0)
* Fixed #18: Replaced beanstalkd with ZeroMQ sockets
* Fixed #10: Added convenience spawning of ssbench-workers
* Fixed #20: Included a (simple) multi-box invocation example in README.
ssbench (0.1.1)
* Fixed up README installation instructions.
ssbench (0.1.0)
* Fixed #23: Allow per-size CRUD profile in scenarios.
* Fixed object uploading bug when size > several chunk sizes with retries
* Fixed #27: Replaced eventlet with gevent to improve HTTPS support
* Fixed bug where "." was printed during a run for results with no
first_byte_latency (i.e. CREATE or UPDATE) instead of "_".
ssbench (0.0.12)
* Fixed #15: added CHANGELOG and AUTHORS; added "Contributing to ssbench"
section to README.
* Cleaned up PEP8 compliance and added .pep8 to ignore E501: line too long
ssbench (0.0.11)
* Added ssbench-master scenario op count override.
* Fixed up docs (README.rst).
* Fixed #13: can set ssbench-worker concurrency (again)
* Removed stale dependency on PyYAML.
ssbench (0.0.10)
* Fixed #11: added --noop mode.
* Fixed #5: replaced YAML with msgpack.
* Fixed #12: addded ability to profile master and worker.
* Got rid of some result-sinking code which was slowing the master down
quite a bit.
* Fixed #4: replace median with configurable percentile.
ssbench (0.0.9)
* Fixed #1: handle zero responses per CRUD type.
* Fixed #2: add start/stop times & duration to report.
* Fixed #3: add timestamp to default output stats filename.
ssbench (0.0.8)
* Fixed bug in reporting where per-size stats per operation type were using
the "all sizes" data.
* Fixed bug in GET where last_byte_latency was being calculated before
actually reading the data out of the socket.
* Some operations cannot get a meaningful first-byte-latency, so they
don't; this is now correctly reported as "N/A".
* Each result's Swift X-Trans-Id header is now included in the result.
* Added "Swift TX ID for worst latench" column to report.
* Increased precision in report data (now visible down to milliseconds).
ssbench (0.0.7)
* Added .requirements.txt to
* Added an optional -S and -T args to specify the storage URL and a token.
* ssbench-master will use ST_* env vars as default values for -A, -U, -K.
* Added -v/--verbose to ssbench-master/worker.
* ssbench-master can override the scenario-specified container count (-c #)
and user count (-u #) on the command-line.
* Removed ssbench-worker concurrency option and just hard-coded it to 256.
* Added back in real-time, per-result single-character progress to STDERR.
* Updated and improved README.rst
ssbench (0.0.4)
* Fixed exclusion of tests when packaging.
ssbench (0.0.3)
* Fixed bug in ssbench-master when default results dir already exists.
* Updated README.rst
ssbench (0.0.2)
* Converted README to reST format.
ssbench (0.0.1)
* Initial public release after rewrite of ssbench guts