Dont replace jira url

[dimanoga]

Bug summary

Hi there,

In our company, API access is provided through a white list system.
To "fix" IP addresses, we use Nginx as a proxy server for accessing Jira.
However, we have noticed that when we use the connection.issuel), in the dictresource method, the URL is replaced with one that Jira returns, rather than the one from which the request originated. Additionally, when updating issues using the issue.update method, the URL used is not the one that the authorization was initiated from, but rather the one provided by the Jira server.

Due to this, when attempting to update a task, we receive a 401 unauthorized error. We are hoping to resolve this issue and ensure that the correct URL is used for all API requests.

Could you please assist us in resolving this issue? We would greatly appreciate any assistance you can provide. Thank you in advance for your attention.

Is there an existing issue for this?

• I have searched the existing issues

Jira Instance type

Jira Server or Data Center (Self-hosted)

Jira instance version

8.20.0

jira-python version

3.5.2

Python Interpreter version

3.11

Which operating systems have you used?

• Linux
• macOS
• Windows

Reproduction steps

#1 connect to jira via nginx url
connection = JIRA(options={'server':"https://my-nginx.url/', 'verify': False}, auth=(login, pass))
#2 get issue
issue = connection.issue(key='JIRA-1')
#3 try to update issue
issue.update(fields={'description':'test'}) # the url is already used here = jira.company.com

Stack trace

File "/PycharmProjects/portal_back/venv/lib/python3.11/site-packages/jira/resources•py", Line 852, in update
super (). update(async_=async_, jira-jira, notify=notify, fields=data)
File "/PycharmProjects/portal_back/venv/lib/python3.11/site-packages/jira/resources-py", Line 328, in update
r = self._session.put (self.self + querystring, data-json.dumps (data))
Fetur se. euest Por, urt, vtadta, uergs) se-packages/requests/sesstons.py". Line 649, in put.
File "/PycharmProjects/portal_back/venv/lib/python3.11/site-packages/jira/resilientsession-py", line 247, in request elif raise_on_error (response, **processed_kwargs):
File "/PycharmProjects/portal_back/venv/lib/python3.11/site-packages/jira/resilientsession-py", Line 72, in raise_on_error
raise JIRAError(
jira.exceptions.JIRAError: JiraError HTTP 401 url: https://jira.company.com/rest/api/2/issue/15832118
text: You do not have the permission to see the specified issue., Login Required
response headers = {''}
response text = ("errorMessages": ["You do not have the permission to see the specified issue.", "Login Required"], "errors":}}
Process finished with exit code 1

Expected behaviour

The Jira URL has not been replaced, but the one that was used for authorization is still being used,

Additional Context

No response

[ichimukuren]

We’re experiencing a very similar issue in our setup! We also use Nginx as a proxy for Jira with IP whitelisting, and have noticed the URL handling seems inconsistent, which causes authentication errors like the 401 Unauthorized you mentioned.

It would be great if the Jira team could help clarify or address this behavior. Thanks for sharing your experience—it’s helpful to know we’re not alone with this issue!

[skiedude]

We also hit this, in our case. To reach jira from a server, you have to use a proxy address, but in some of the helper methods, they use the address returned on the issue object.

My specific use case was in the Issue().add_field_value which forwards to => Resource().update which uses the self.self value returned on the ticket from jira. (which appears to be the same place that @dimanoga referenced in their opening comment)

To get around it in our local testing we used the urllib.parse library to compare things.

import jira
user = 'username'
password = 'password'
j = jira.JIRA('https://jira.internal.net', basic_auth=(user, password))
issue = j.issue('ID-363772')
issue.add_field_value(field='labels', value='testpython')

The above code would fail due to the wrong jira address being used. To fix it, we added the following to https://github.com/pycontribs/jira/blob/main/jira/resources.py#L329

from urllib.parse import urlparse, urlunparse, ParseResult 
...
self_parsed = urlparse(self.self) #Get the ticket url from the ticket returned from jira, and run it through urlparse
server_parsed = urlparse(self._options['server']) #run the server address we configured in the JIRA object through urlparse
if self_parsed.netloc != server_parsed.netloc: #compare the two values
    self.self = urlunparse(
        ParseResult(
            scheme=self_parsed.scheme,
            netloc=server_parsed.netloc,  #use the server address we set in our config, everything else the same
            path=self_parsed.path, 
            params=self_parsed.params, 
            query=self_parsed.query, 
            fragment= self_parsed.fragment
        )
)
r = self._session.put(self.self + querystring, data=json.dumps(data))

This obviously very specific to the resource.py, but it may be adapted to other parts of the code.

[skiedude]

This can actually be achieved with the already included methods of the class Resource a bit more precise

        path = self._resource.format(self.key)
        self.self = self._get_url(path)

I'll put in a PR for discussion/review

[YanvarevDmitry]

We are now bypassing this by substituting the correct link through self.self, but this does not seem to be the right solution.

[studioj]

This can actually be achieved with the already included methods of the class Resource a bit more precise

        path = self._resource.format(self.key)
        self.self = self._get_url(path)

I'll put in a PR for discussion/review

if opening a pr , do know we'd like to have it tested.

This repo is not under active development and maintained by volunteers. As you can see prs get merged slowly. We cheer to contributors wanting to do the effort of creating a PR but ask for patience ... and a short bump once in a while :-) this is by far our main project

[studioj]

aha a PR was created, => we need to get deployment working first though.

https://github.com/pycontribs/jira/pull/1925/files

[skiedude]

aha a PR was created, => we need to get deployment working first though.

https://github.com/pycontribs/jira/pull/1925/files

Linting hopefully fixed 😃