Skip to content

Latest commit

 

History

History
297 lines (210 loc) · 13.3 KB

CHANGELOG.md

File metadata and controls

297 lines (210 loc) · 13.3 KB
Raw

CHANGELOG

This file documents important changes to the SignalFx Python client library.

[1.1.12] - 2020-08-25: Fix some signalflow defaults

What it says on the tin. #111

[1.1.11] - 2020-08-24: Fix reconnection

Fix reconnection. #110

[1.1.10] - 2020-08-04: Fix get_aws_unique_id

Fix an exception in calling get_aws_unique_id.

[1.1.9] - 2020-07-21: Fix get_detector methods

get_detector_events and get_detector_incidents failed to correctly pass named arguments.

[1.1.8] - 2020-07-20: Fix tests and connection errors

  • Fix some broken tests and flake8 problems.
  • Handle ConnectionErrors caused by faulty urllib3 #104
  • Add python version classifiers #103
  • Adjust get_aws_unique_id to try ECS metadata before EC2. #71
  • Add additional parameters to the client. #105

[1.1.7] - 2020-06-17: Fix Get Data Link

The get_datalink function was just completely wrong and errored. Oops!

[1.1.6] - 2020-05-30: Fix JSON ingest client

Fix the JSON ingest client when using a Python 3.x interpreter. The zlib module expects a bytes object passed to the zlib.compress() function, so we need to encode our JSON payloads as UTF-8 byte strings before passing them to the _post() function.

[1.1.5] - 2020-04-27: Add datalink methods

Adds get_datalinks and get_datalink.

[1.1.4] - 2020-03-25: Add ingest error counters

Adds counters for errors during ingest and reset_error_counters to reset and return those counters.

[1.1.3] - 2020-01-16: Add new dashboard methods

Added get_dashboards, get_dashboard_group, and get_dashboard_groups.

[1.1.2] - 2019-11-14: Fix accessing some computation response fields

Fixed some bugs that tried to access a missing key

[1.1.1] - 2019-08-22: Added Detector Event method and Computation messages

  • Added get_detector_events method for getting events for a detector.
  • Added new properties for computation populated by job info messages

[1.1.0] - 2019-02-28: Expanded Detector and Incident methods

Added methods for accessing the API functionality of retrieving incidents, retrieving a detector by its ID, retrieving incidents for a detector by its ID, and clearing an incident by its ID.

Also added preliminary support for disable_all_metric_publishes flag when executing SignalFlow computations and removed an unsupported Property datapoint attribute.

[1.0.19] - 2018-05-03: Dimension support in Pyformance wrapper

This release enhances the SignalFx pyformance package and extends the Pyformance registry to support dimensional metadata. Please refer to the README and examples for more information on changes to the pyformance package.

[1.0.18] - 2018-03-15: Compression of datapoint payloads

The main change in 1.0.18 is that payloads of datapoints sent to SignalFx will now be compressed by default (using GZip compression and Content-Encoding: gzip). This can be disabled by specifying compress=False on the SignalFx client, or on the ingest sub-client directly.

[1.0.17] - 2018-03-02: Support for immediate SignalFlow results

Added support for the new immediate flag when executing SignalFlow computation. Setting this flag to true forces the system to shift the timerange of the computation by the maxDelay amount (either detected, or specified), to ensure that the computation returns and completes without additional delay to wait for late data.

Also added support in the library to access event metadata on events received from a SignalFlow computation.

Updated the default TCP timeout to 5 seconds to match our Java and Ruby libraries.

[1.0.16] - 2017-03-24: SignalFlow streaming performance

Added support for compressed SignalFlow WebSocket messages, which improves the streaming performance by reducing the bandwidth requirements of the client.

[1.0.15] - 2017-02-20: Preflight API

Added support for the detector preflighting API, allowing for the execution of a detector program in a mode that simply summarizes the events that would otherwise be generated, allowing for the quicker execution of that preflighting over longer spans of historical data.

This release also includes a bugfix to how the total number of input timeseries is calculated, as well as support for a new version of the binary data message encoding (not yet used).

[1.0.14] - 2016-12-07: SignalFlow client bug fixes and context managers

Fixes a bug in the SignalFlow streaming computation client library that would lead to an incomplete first data batch returned from the computation stream when the program being executed as multiple published streams.

All three sub-clients also now support Python context managers so they can be used in with blocks:

with signalfx.SignalFx().signalflow('MY_TOKEN') as flow:
    computation = flow.execute(program)
    for msg in computation.stream():
        # ...

[1.0.13] - 2016-12-05: More features from detector APIs

Added support for the /v2/detector/validate endpoint via rest.validate_detector(), and support for searching detectors by tags when using rest.get_detectors().

It is also now possible to pass ignore_not_found=True to REST delete operations to ignore failures on attempting to remove a non-existent resource for which the DELETE call would otherwise return a 404.

[1.0.12] - 2016-11-28: Detector APIs

Added support for managing SignalFlow V2 detectors via the REST client.

[1.0.11] - 2016-11-23: Long value support

long type metric values were previously unsupported. This release allows int64 values and property values as defined by the protocol buffer. Values greater than or equal to -(2**63) and less than or equal to (2**63)-1. Values exceeding the specified boundaries will raise a ValueError exception.

Boolean property values were previously dispatched as integer values. This release fixes this and emits boolean property values as a boolean type.

[1.0.10] - 2016-11-21: Unicode event properties fix

Unicode strings were previously unsupported for event properties. This release allows event properties to be assigned unicode strings.

[1.0.9] - 2016-10-26: Datapoints queue draining fix

In certain situations, it was possible for the ingest client to stop and let the program exit before the datapoints queue was fully drained to SignalFx. This release fixes this and ensures that the background sending thread does not prematurely exits before the queue is fully drained.

[1.0.8] - 2016-10-20: A missing field from events

Version 1.0.8 is a small point release to expose the EventTimeSeries ID from events received from a SignalFlow v2 computation. This field can then be used to lookup the metadata of that EventTimeSeries from the computation.

c = flow.execute(program)
for msg in c.stream():
    if isinstance(msg, signalfx.signalflow.messages.EventMessage):
        pprint.pprint(c.get_metadata(msg.tsid))

[1.0.7] - 2016-10-05: More Python 3 compatibility

Version 1.0.7 includes an updated version of the generated ProtocolBuffer code, generated with version 3 of the Protocol Buffer compiler and library, which produces Python 3 compatible Python source code.

[1.0.5] - 2016-09-29: Python 3 compatibility

Version 1.0.5 of the SignalFx Python client library provides compatibility for Python 3.x.

[1.0.1] - 2016-06-02: Support for SignalFlow API

In version 1.0.1 of this client, we introduced support for the SignalFlow API. This means you can use this client to programmatically stream analytical computations from SignalFx in real time, in addition to sending data in to SignalFx. Using SignalFlow, you can build your own applications that leverage SignalFx's streaming analytics outside the SignalFx UI. To read more about SignalFlow, click here: https://developers.signalfx.com/docs/signalflow-overview

Adding support for SignalFlow required upgrades to this client that are backwards-incompatible with previous versions. Customers who are upgrading from version 0.3.9 or earlier must change how the client is instantiated in application code, and how it is authorized.

1. Client instantiation

Each client's features are now divided among data transmission to SignalFx (ingest), metadata retrieval (rest), and data streaming from SignalFx to your client (signalflow). This means that when you instantiate a SignalFx client object, you must also choose which SignalFx API you will access using that object.

  • If you use the client to send data to SignalFx, use ingest.
  • If you use the client to retrieve metric names and metadata from SignalFx, use rest.
  • To use the new SignalFlow API to stream analytics to your client, use signalflow.

Before SignalFlow support, client instantiation used to look like this:

import signalfx

sfx = signalfx.SignalFx('ACCESS_TOKEN')

After SignalFlow support, client instantiation now looks like this:

import signalfx

sfx = signalfx.SignalFx()

# To send data from client to SignalFx using the ingest API
ingest = sfx.ingest('API_SESSION_TOKEN')

# To get and set properties and tags using the REST API
rest = sfx.rest('USER_SESSION_TOKEN')

# To stream data from SignalFx to client using the SignalFlow API
flow = sfx.signalflow('USER_SESSION_TOKEN')
2. Authenticating to SignalFx

As illustrated in the above example, instead of supplying an access token at the moment of instantiating a SignalFx client object, you must now supply it when you choose which API to access.

  • For ingest, supply your API session token. Obtain this token from within the SignalFx app.
  • For rest and signalflow, first authenticate with your SignalFx credentials, then supply your user session token. Click here to read about SignalFx authentication.

You can authenticate using cURL as in the following example:

curl -s -XPOST -HContent-Type:application/json https://api.signalfx.com/v2/session -d'{"email":"USERNAME","password":"PASSWORD"}' | jq -r '.accessToken'

Alternatively, you could authenticate to SignalFx programmatically from within this client:

import signalfx
sfx = signalfx.SignalFx()
token = sfx.login("USERNAME", "PASSWORD")