I am looking for an example to get the user list, add a user, delete a user from Sumologic

After getting the proper endpoint from _get_endpoint. The if statement checking for a forward slash at the end of the URL is checking the wrong endpoint variable. It's checking the endpoint variable passed in when initializing the class as opposed to self.endpoint.

When trying to get the appropriate API URL, the get statement in the _get_endpoint() function would not follow redirects even though it the default setting is to allow_redirects was set to true,

Only seen this on a windows 10 host behind a corporate proxy with SSL offloading. This does not happen on macOS with no custom ssl ca bundles or behind any proxies.

Hi there,

I want to send logs using Sumologic-python-sdk.

I see there are two function in the code, which are put and post.

Can anyone one explain me how to use them, what are the parameters it accepts and how I should send payload to particular collector source.

When the API doesn't return 200, you're using raise_for_status to throw exceptions:

  File "/usr/local/lib/python2.7/site-packages/sumologic/sumologic.py", line 77, in search_job_records
    r = self.get('/search/jobs/' + str(search_job['id']) + '/records', params)
  File "/usr/local/lib/python2.7/site-packages/sumologic/sumologic.py", line 43, in get
    r.raise_for_status()
  File "/usr/local/lib/python2.7/site-packages/requests/models.py", line 825, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 400 Client Error: Bad Request

However, the problem with this is that, particular with 400 Client Error, the return JSON from Sumo indicates what exactly in the request is wrong. However, with this message, we are left with no idea what may be bad within a given request.

For example, if the start time isn't properly formatted when doing a search, the library shows this 400 Client Error but no information that the reason for failure is the timezone formatting.

It would be helpful to all developers, especially those new to Sumo, to add this information in the exception.

While using the search-job.py script the records returned are always:

{'fields': [{'name': '_count', 'fieldType': 'int', 'keyField': False}], 'records': []}

for query * | count
Using the UI I get many results.

Any suggestions?

Need to look into whether and how we can change the name and if that will break anything.

Heyo,

The line from sumologic import * in init.py (https://github.com/SumoLogic/sumologic-python-sdk/blob/2ca5e97e4f7514a98cdf8ddd372a9ff00ea8b8a1/sumologic/__init__.py) doesn't work correctly in Python 3.6 because it uses an absolute import, and not a relative import. It ends up just importing its own (currently empty) package. As a result, from sumologic import SumoLogic doesn't work, and you have to import sumologic.sumologic to access the class:

$ python
Python 3.6.0 (default, Dec 24 2016, 00:01:50)
[GCC 4.2.1 Compatible Apple LLVM 8.0.0 (clang-800.0.42.1)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import sumologic
>>> dir(sumologic)
['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', 'sumologic']
>>>
>>> from sumologic import SumoLogic
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: cannot import name 'SumoLogic'
>>>
>>> from sumologic.sumologic import SumoLogic
>>> SumoLogic
<class 'sumologic.sumologic.SumoLogic'>

There are multiple APIs implemented server side that don't appear to be implemented in the SDK. There are also some methods in the SDK that don't appear to correspond to the documented API. Could the SDK be updated to true up support to current Sumo API standards?

Examples in API but not in client SDK:

• https://api.sumologic.com/docs/#operation/getItemByPath
• https://api.sumologic.com/docs/#operation/updateFolder

Examples in SDK but not found in API documentation:

• sync_folder()
  • POST /content/folders/%s/synchronize
  • https://github.com/SumoLogic/sumologic-python-sdk/blob/main/sumologic/sumologic.py#L217
• check_sync_folder()
  • GET /content/folders/%s/synchronize/%s/status
  • https://github.com/SumoLogic/sumologic-python-sdk/blob/main/sumologic/sumologic.py#L220

I think it would be helpful to have a higher-level, object-based abstraction around the current API.
Is this something that other people would find valuable?

Somewhat like Boto3 Service Resources for the AWS API
With that, there are Waiter and Paginator objects that can handle the polling, pagination, automatic retries after rate limiting.
Similarly, a heirarchy of objects and collections of child objects, which can lazy load as needed.
An application no longer needs to be cluttered with polling logic.

From @Segobiano - the change of the signature to the method in PR46 breaks positional argument usage. Looks like this might also break many of the examples listed.

The security services of Sumo API framework has been upgraded. API requests with multiple forward slashes (//) will receive a "500 Internal Server Error" response. Please remove any additional and unnecessary forward slashes from the API URL. Similar issue: #38

when I call the collectors endpoint with this sdk it fails with a 500. I believe this might be because there are // in the url for example: https://api.us2.sumologic.com/api/v1//collectors

if I try to get back a list of collectors or access a single collector I now get an error where this worked up till a month or so ago.

sumo.collectors()

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/rick.jury/.local/share/virtualenvs/sumologic-api-tools-f4YSK5be/lib/python3.6/site-packages/sumologic/sumologic.py", line 99, in collectors
    r = self.get('/collectors', params)
  File "/Users/rick.jury/.local/share/virtualenvs/sumologic-api-tools-f4YSK5be/lib/python3.6/site-packages/sumologic/sumologic.py", line 53, in get
    r.raise_for_status()
  File "/Users/rick.jury/.local/share/virtualenvs/sumologic-api-tools-f4YSK5be/lib/python3.6/site-packages/requests/models.py", line 940, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 500 Server Error: {"id":"YGMEC-MMS5N-UZJ1K","errors":[{"code":"internal.error","message":"Internal server error."}]} for url: https://api.us2.sumologic.com/api/v1//collectors

Is ez_setup really needed? pip installs break because of it. Here is traceback of it

$ pip install sumologic-sdk
You are using pip version 7.1.0, however version 8.0.2 is available.
You should consider upgrading via the 'pip install --upgrade pip' command.
Collecting sumologic-sdk
  Using cached sumologic-sdk-0.1.4.tar.gz
    Complete output from command python setup.py egg_info:
    Traceback (most recent call last):
      File "<string>", line 20, in <module>
      File "/private/var/folders/jp/djdh4dx56011p3hf0m71p2306sp5p0/T/pip-build-o342kT/sumologic-sdk/setup.py", line 1, in <module>
        import ez_setup
    ImportError: No module named ez_setup

    ----------------------------------------
Command "python setup.py egg_info" failed with error code 1 in /private/var/folders/jp/djdh4dx56011p3hf0m71p2306sp5p0/T/pip-build-o342kT/sumologic-sdk

I will submit a patch fixing this. I don't really think using ez_setup is needed if we can just specify that setuptools is required to be installed as a prerequisite.

instead of having this script as an example https://github.com/SumoLogic/sumologic-python-sdk/blob/master/scripts/search-job-messages.py I'd like to create API callback function so developers can use, any procedure for PR?

usage-report.py references a function called search_and_wait.

This fails because, the SumoLogic class has no instance method search_and_wait.

See https://github.com/SumoLogic/sumologic-python-sdk/search?utf8=%E2%9C%93&q=search_and_wait

Hi Team
sumo = SumoLogic(ACCESS_ID, ACCESS_KEY)
response = sumo.search_job(query, from_v, to, timeZone, True, autoParsingMode='Manual')
After calling the sumo.search_job_messages(response,8)
I am getting the below error though I am getting the job_id from search_job
raise HTTPError(http_error_msg, response=self) requests.exceptions.HTTPError: 400 Client Error: { "status" : 400, "id" : "2U5U6-YAO5P-DOFTB", "code" : "searchjob.generic", "message" : "Generic error." } for url: https://api.sumologic.com/api/v1/search/jobs/job_id/messages?limit=8&offset=0

I was trying with the below option:
auth = (ACCESS_ID, ACCESS_KEY)
headers = {
'Content-Type': 'application/json',
'Accept': 'application/json',
}
results_url = f'https://api.sumologic.com/api/v1/search/jobs/{response['id']}/messages'
resp = requests.get(results_url, auth=auth, headers=headers)

I am getting 401 error code.

How to solve this?

Collector Management API added a new features recently. Need to add methods to call the functionality. Release notes for reference (specifically 6/26/20190 https://help.sumologic.com/Release-Notes/Developer-Release-Notes

The API server implements several asynchronous APIs that require users to submit job, poll for status, and finally fetch results. It would be logical and would help fulfill the purpose of the Python SDK as a high level client to expose helper methods that support handling these asynchronous backend jobs smoothly. The goal would be for a single call to handle necessary functionality:

1. Submit job
2. Poll for completion
3. Return results

Support should allow for some specification by user of things like poll period, maximum number of tries/timeout, etc. The higher level methods could then signal situations like timeouts and failures with exceptions or similar.

The outcome would be higher level support in the SDK that smooths out the user experience and enables simpler client code without the need to add in so much polling and retry logic, etc.

Similar ideas have been proposed in #3 and #32.

Can't find wrapper for https://github.com/SumoLogic/sumo-api-doc/wiki/Collector-Management-API#creating

Hi guys,

I'm looking into this sdk to understand how it all works. In general I will need to get all data and push it to my system.

I'm executing script like this:

$ cat query.sumoql | python search-job-messages.py <secret_id> <secret_key> `date +%s000` $(echo | awk "{print 1000+`date +%s000`}") EEST                     Traceback (most recent call last):
  File "search-job-messages.py", line 31, in <module>
    sj = sumo.search_job(q, fromTime, toTime, timeZone)
  File "build/bdist.cygwin-2.6.0-x86_64/egg/sumologic/sumologic.py", line 67, in search_job
  File "build/bdist.cygwin-2.6.0-x86_64/egg/sumologic/sumologic.py", line 52, in post
  File "/usr/lib/python2.7/site-packages/requests/models.py", line 862, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 400 Client Error: Bad Request for url: https://api.us2.sumologic.com/api/v1/search/jobs

Ivan.

Job deletion seems to be missing.

The SumoLogic constructor calls an internal _get_endpoint method, which prints an "SDK Endpoint" to standard output; e.g. SDK Endpoint https://api.us2.sumologic.com/api. This makes it difficult to write utilities that are intended to be used in a unix-style pipeline, such as outputting JSON that can be parsed manipulated downstream by tools such as jq.

AutoParsing Mode is getting passed as null causing incomprehensible 500 errors. From reaching out to sumo support this is because it's passing that null.
This should either not be set if it's null (remove the key) or it should be set to either Manual or Automatic

Hi,

There's been a number of changes since the 0.1.7 release to pypi over a year ago in February 2017. Can you please tag and release a new version.

Thanks.

Hi, i'm trying to access a job that was successfully created, but I keep getting even though the create job returns 202 status

{
"status" : 404,
"id" : "D0FMU-CJDYR-H0VCO",
"code" : "searchjob.jobid.invalid",
"message" : "Job ID is invalid."
}

Thanks!

Example-
| Account_name as User
it renames field with lower case user

Hello, v0.1.9 has been tagged for some time in the repository but has yet to be published to PyPI.