OCPERT-154: Change urls to Atlassian Cloud#930
OCPERT-154: Change urls to Atlassian Cloud#930tomasdavidorg wants to merge 1 commit intoopenshift:masterfrom
Conversation
openshift-ci-robot
added
the
jira/valid-reference
openshift-ci
bot
added
the
do-not-merge/work-in-progress
|
@tomasdavidorg: This pull request references OCPERT-154 which is a valid jira issue. Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the task to target the "4.22.0" version, but no target version was set. DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository. |
|
Skipping CI for Draft Pull Request. |
WalkthroughThis pull request migrates Jira URL references from the legacy Red Hat instance (issues.redhat.com) to the new Atlassian Cloud instance (redhat.atlassian.net). The changes update documentation, core logic, test expectations, and dashboard components consistently throughout the codebase. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes 🚥 Pre-merge checks | ✅ 3 | ❌ 2❌ Failed checks (2 warnings)
✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
@tomasdavidorg: This pull request references OCPERT-154 which is a valid jira issue. Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the task to target the "4.22.0" version, but no target version was set. DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository. |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
♻️ Duplicate comments (3)
tests/test_notification.py (1)
162-162: Ensure Slack message Jira URL matches the actual link builder.This expectation uses
/browsewithout/jira. If the production builder now uses/jira/browse, the test will drift. Please verify the builder and keep tests aligned with the canonical base.#!/bin/bash # Locate Jira link builders and message formatting to confirm the base URL used. rg -n "get_jira_link|redhat\.atlassian\.net" -g'*.py' rg -n "get_slack_message_for_unverified_cve_issues_to_managers" -C3🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@tests/test_notification.py` at line 162, The test expectation contains a Jira link using "/browse" but the canonical link builder (functions like get_jira_link and the message builder get_slack_message_for_unverified_cve_issues_to_managers) uses "/jira/browse"; verify which base the production builder returns and update the test expectation to match that canonical base (or change the builder if the test is canonical). Ensure the test string uses the exact URL path returned by get_jira_link (e.g., include "/jira/browse" if that is what the builder produces) so the Slack message assertion stays in sync.oar/core/shipment.py (1)
1159-1160: Keep bug-removal source matching in sync with issue extraction.Same normalization concern as above; this should use the same normalized/allowed source set to avoid partial removals.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@oar/core/shipment.py` around lines 1159 - 1160, The conditional that checks for the specific string 'source: redhat.atlassian.net/jira' (inside the loop that uses lines and lines_to_remove) is brittle and must be changed to use the same normalization/allowed-source logic used by the issue extraction code; replace the hardcoded substring test with a membership check against the canonical allowed_sources (or call the same normalize_source()/extract_issue_sources helper used elsewhere) after normalizing lines[i+1], so the bug-removal path and issue-extraction path use the identical normalized source set.tools/release_progress_dashboard/release_progress_dashboard.py (1)
521-521: Keep Jira URL base consistent across link builders.This renders Jira links with
/browse(no/jira) while other helpers now build/jira/browse. Please confirm the canonical base and keep all builders/tests in sync.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@tools/release_progress_dashboard/release_progress_dashboard.py` at line 521, The Jira link is being built as jira_url = f"https://redhat.atlassian.net/browse/{jira_ticket}" which is inconsistent with other link builders using the "/jira/browse" path; update this to use the canonical base (e.g., jira_url = f"https://redhat.atlassian.net/jira/browse/{jira_ticket}") and ensure any other builders/tests that construct Jira URLs use the exact same "/jira/browse" base so all helpers and tests remain in sync.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@oar/core/util.py`:
- Around line 107-108: get_jira_link currently builds URLs with an extra "/jira"
segment so it returns https://redhat.atlassian.net/jira/browse/{key} which is
invalid; update the function get_jira_link to use the correct base host
"https://redhat.atlassian.net" (i.e. remove the "/jira" suffix) so it returns
https://redhat.atlassian.net/browse/{key} for all callers.
In `@oar/notificator/jira_notificator.py`:
- Line 709: The JIRA client is being instantiated with an incorrect server URL
suffix: update the JIRA(...) call that assigns jira (the JIRA =
JIRA(server="https://redhat.atlassian.net/jira", token_auth=jira_token)
instantiation) to use the base Atlassian Cloud URL without the "/jira" suffix
(https://redhat.atlassian.net) so the library can append REST paths correctly;
also replace the hardcoded URL by reading from configuration like
oar/core/jira.py's get_jira_server() (or call get_jira_server()) to centralize
the server setting and avoid future hardcoding.
In `@tests/test_jira_notificator.py`:
- Line 16: The JIRA client instantiation uses an incorrect server URL and auth
method: update the JIRA(...) call that assigns self.jira (and any other JIRA
instantiations such as in jira_notificator.py and oar/core/jira.py where
ConfigStore may supply the URL) to use server="https://redhat.atlassian.net"
(remove the "/jira" suffix) and switch from token_auth to basic_auth using the
email and API token (basic_auth=(os.environ.get("JIRA_EMAIL"), jira_token)), and
ensure all other JIRA client constructions follow the same pattern for Atlassian
Cloud compatibility.
---
Duplicate comments:
In `@oar/core/shipment.py`:
- Around line 1159-1160: The conditional that checks for the specific string
'source: redhat.atlassian.net/jira' (inside the loop that uses lines and
lines_to_remove) is brittle and must be changed to use the same
normalization/allowed-source logic used by the issue extraction code; replace
the hardcoded substring test with a membership check against the canonical
allowed_sources (or call the same normalize_source()/extract_issue_sources
helper used elsewhere) after normalizing lines[i+1], so the bug-removal path and
issue-extraction path use the identical normalized source set.
In `@tests/test_notification.py`:
- Line 162: The test expectation contains a Jira link using "/browse" but the
canonical link builder (functions like get_jira_link and the message builder
get_slack_message_for_unverified_cve_issues_to_managers) uses "/jira/browse";
verify which base the production builder returns and update the test expectation
to match that canonical base (or change the builder if the test is canonical).
Ensure the test string uses the exact URL path returned by get_jira_link (e.g.,
include "/jira/browse" if that is what the builder produces) so the Slack
message assertion stays in sync.
In `@tools/release_progress_dashboard/release_progress_dashboard.py`:
- Line 521: The Jira link is being built as jira_url =
f"https://redhat.atlassian.net/browse/{jira_ticket}" which is inconsistent with
other link builders using the "/jira/browse" path; update this to use the
canonical base (e.g., jira_url =
f"https://redhat.atlassian.net/jira/browse/{jira_ticket}") and ensure any other
builders/tests that construct Jira URLs use the exact same "/jira/browse" base
so all helpers and tests remain in sync.
ℹ️ Review info
Configuration used: Repository: openshift/coderabbit/.coderabbit.yaml
Review profile: CHILL
Plan: Pro
Cache: Disabled due to data retention organization setting
Knowledge base: Disabled due to data retention organization setting
📒 Files selected for processing (9)
AGENTS.mddocs/JIRA_NOTIFICATOR.mdoar/core/shipment.pyoar/core/util.pyoar/notificator/jira_notificator.pyprow/job/job.pytests/test_jira_notificator.pytests/test_notification.pytools/release_progress_dashboard/release_progress_dashboard.py
| def get_jira_link(key): | ||
| return "%s/browse/%s" % ("https://issues.redhat.com", key) | ||
| return "%s/browse/%s" % ("https://redhat.atlassian.net/jira", key) |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
# First, find and examine the get_jira_link function in context
grep -n "get_jira_link" oar/core/util.py | head -20Repository: openshift/release-tests
Length of output: 94
🏁 Script executed:
# Search for all usages of get_jira_link in the codebase
rg "get_jira_link" --type py -B 2 -A 2Repository: openshift/release-tests
Length of output: 5370
🏁 Script executed:
# Search for other Jira URL patterns or references in the codebase
rg "atlassian|jira" --type py -i | head -30Repository: openshift/release-tests
Length of output: 2968
🏁 Script executed:
# Check if there are tests for this function
fd -e py -type f | xargs grep -l "get_jira_link\|test.*jira" 2>/dev/null | head -10Repository: openshift/release-tests
Length of output: 237
🌐 Web query:
Jira Cloud browse URL format standard path atlassian.net
💡 Result:
The standard Jira Cloud “browse issue” URL on an atlassian.net site is:
https://{site}.atlassian.net/browse/{ISSUE-KEY}(example:https://mysite.atlassian.net/browse/PROJ-123) [1][2]
Notes:
{site}is your Jira Cloud site subdomain (the part before.atlassian.net). [2]{ISSUE-KEY}is the issue key likePROJ-123. [1][2]
Sources: Atlassian developer docs example for Jira issue URLs [1], Atlassian Developer Community confirmation of the format [2].
🏁 Script executed:
# Search for tests of get_jira_link
find . -name "*.py" -type f | xargs grep -l "test.*get_jira_link\|get_jira_link.*test" 2>/dev/null | head -10Repository: openshift/release-tests
Length of output: 49
🏁 Script executed:
# Search for any test files that might test util.py
find . -name "*test*.py" -type f | xargs grep -l "util\|get_jira_link" 2>/dev/null | head -10Repository: openshift/release-tests
Length of output: 365
🏁 Script executed:
# Check test_jira.py for any URL format tests
cat tests/test_jira.py | head -100Repository: openshift/release-tests
Length of output: 3979
🏁 Script executed:
# Check test_notification.py for get_jira_link validation
rg "get_jira_link|browse" tests/test_notification.py -B 2 -A 2Repository: openshift/release-tests
Length of output: 225
🏁 Script executed:
# Check more context around that test assertion
rg "redhat.atlassian.net/browse" tests/test_notification.py -B 5 -A 5Repository: openshift/release-tests
Length of output: 512
🏁 Script executed:
# Verify the exact line in util.py to confirm the /jira path
sed -n '107,108p' oar/core/util.pyRepository: openshift/release-tests
Length of output: 161
🏁 Script executed:
# Check the full test_notification.py to see how the URL is constructed in the test
rg "unverified_cve" tests/test_notification.py -B 15 | head -50Repository: openshift/release-tests
Length of output: 1901
🏁 Script executed:
# Search for where the notification message is actually created to see if get_jira_link is called
rg "unverified_cve\|get_jira_link" oar/core/notification.py -B 3 -A 3Repository: openshift/release-tests
Length of output: 49
🏁 Script executed:
# Search to see if the test actually calls get_jira_link or just checks the output string
rg "get_slack_message_for_unverified_cve_issues_to_managers" oar/core/notification.py -B 5 -A 20 | head -60Repository: openshift/release-tests
Length of output: 2523
🏁 Script executed:
# Check if get_jira_link is called within that method
ast-grep --pattern 'def get_slack_message_for_unverified_cve_issues_to_managers {
$$$
}'Repository: openshift/release-tests
Length of output: 49
🏁 Script executed:
# Just search for the method implementation more directly
rg "def get_slack_message_for_unverified_cve_issues_to_managers" oar/core/notification.py -A 30Repository: openshift/release-tests
Length of output: 1613
Fix hardcoded Jira URL—it includes erroneous /jira path that breaks links.
get_jira_link() generates https://redhat.atlassian.net/jira/browse/{key}, but Jira Cloud browse URLs use the format https://redhat.atlassian.net/browse/{key} without the /jira suffix. Remove /jira from the hardcoded base URL. All links generated through this function (used in 13+ places across notifications and worksheets) are currently malformed.
Fix
def get_jira_link(key):
- return "%s/browse/%s" % ("https://redhat.atlassian.net/jira", key)
+ return "%s/browse/%s" % ("https://redhat.atlassian.net", key)🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@oar/core/util.py` around lines 107 - 108, get_jira_link currently builds URLs
with an extra "/jira" segment so it returns
https://redhat.atlassian.net/jira/browse/{key} which is invalid; update the
function get_jira_link to use the correct base host
"https://redhat.atlassian.net" (i.e. remove the "/jira" suffix) so it returns
https://redhat.atlassian.net/browse/{key} for all callers.
| raise RuntimeError("JIRA token is missing or empty. Please set the JIRA_TOKEN environment variable.") | ||
|
|
||
| jira = JIRA(server="https://issues.redhat.com", token_auth=jira_token) | ||
| jira = JIRA(server="https://redhat.atlassian.net/jira", token_auth=jira_token) |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
# First, let's examine the file and the specific line
wc -l oar/notificator/jira_notificator.pyRepository: openshift/release-tests
Length of output: 106
🏁 Script executed:
# Read the context around line 709
sed -n '700,720p' oar/notificator/jira_notificator.pyRepository: openshift/release-tests
Length of output: 519
🏁 Script executed:
# Search for all JIRA() instantiations in the codebase
rg "JIRA\(" -A 2 -B 2Repository: openshift/release-tests
Length of output: 1164
🏁 Script executed:
# Look for any Jira API calls to understand how the client is used
rg "jira\." oar/notificator/jira_notificator.py | head -20Repository: openshift/release-tests
Length of output: 273
🏁 Script executed:
# Check for any configuration or comments about Jira URL
rg "redhat\.atlassian" .Repository: openshift/release-tests
Length of output: 1636
🏁 Script executed:
# Look for any tests related to Jira
fd -e test -e spec | xargs rg -l "jira" 2>/dev/null | head -10Repository: openshift/release-tests
Length of output: 315
🏁 Script executed:
# Find the get_jira_server() definition
rg "get_jira_server" -B 3 -A 5Repository: openshift/release-tests
Length of output: 1003
🏁 Script executed:
# Check oar/core/jira.py for full context
cat -n oar/core/jira.pyRepository: openshift/release-tests
Length of output: 22782
🏁 Script executed:
# Check oar/core/const.py for configuration
cat oar/core/const.pyRepository: openshift/release-tests
Length of output: 6644
🏁 Script executed:
# Find where jira_server is configured
rg "jira_server" -B 3 -A 3Repository: openshift/release-tests
Length of output: 935
🏁 Script executed:
# Look for ConfigStore initialization and configuration
rg "class ConfigStore" -A 20Repository: openshift/release-tests
Length of output: 3463
🌐 Web query:
jira-python library JIRA() server URL format Atlassian Cloud
💡 Result:
For Atlassian Cloud with the jira (jira-python / pycontribs) library, the JIRA() server URL should be your site base URL, e.g.:
https://<your-site>.atlassian.net(recommended; no extra path) [1]- A trailing slash is also commonly accepted:
https://<your-site>.atlassian.net/[2]
Example:
from jira import JIRA
jira = JIRA(server="https://yourcompany.atlassian.net",
basic_auth=("you@example.com", "your_api_token"))Don’t include a Server/DC-style context path like .../jira—Cloud sites typically don’t use that, and it can break requests/auth expectations. [1]
Sources: Atlassian Community example for Jira Cloud + jira-python [1]; pycontribs/jira issue showing Cloud server='https://XXX.atlassian.net/' usage [2].
🏁 Script executed:
# Check AGENTS.md to see the documented Jira testing URL
cat AGENTS.md | grep -A 2 -B 2 "redhat.atlassian.net"Repository: openshift/release-tests
Length of output: 393
🏁 Script executed:
# Check if there's any documentation about the config initialization
fd -name "*.md" -o -name "*.yaml" | xargs grep -l "jira_server" 2>/dev/nullRepository: openshift/release-tests
Length of output: 299
🏁 Script executed:
# Look at how the jira_notificator_jira.py function is structured
cat -n oar/notificator/jira_notificator.py | head -50Repository: openshift/release-tests
Length of output: 1793
Remove the /jira suffix from the JIRA server URL.
The JIRA() constructor expects the base URL for Atlassian Cloud (https://redhat.atlassian.net), not the REST API endpoint path. Including /jira will cause authentication and API call failures because the library appends REST paths automatically. Change line 709 to https://redhat.atlassian.net.
Additionally, consider moving the URL to configuration (similar to how oar/core/jira.py uses get_jira_server()) instead of hardcoding it to avoid similar issues in the future.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@oar/notificator/jira_notificator.py` at line 709, The JIRA client is being
instantiated with an incorrect server URL suffix: update the JIRA(...) call that
assigns jira (the JIRA = JIRA(server="https://redhat.atlassian.net/jira",
token_auth=jira_token) instantiation) to use the base Atlassian Cloud URL
without the "/jira" suffix (https://redhat.atlassian.net) so the library can
append REST paths correctly; also replace the hardcoded URL by reading from
configuration like oar/core/jira.py's get_jira_server() (or call
get_jira_server()) to centralize the server setting and avoid future hardcoding.
| def setUp(self): | ||
| jira_token = os.environ.get("JIRA_TOKEN") | ||
| self.jira = JIRA(server="https://issues.redhat.com", token_auth=jira_token) | ||
| self.jira = JIRA(server="https://redhat.atlassian.net/jira", token_auth=jira_token) |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Search for JIRA server URL usages and token_auth patterns
echo "=== Files mentioning atlassian.net ==="
rg -l "atlassian\.net" --type=py
echo -e "\n=== JIRA server URL usages (context) ==="
rg -n --type=py "atlassian\.net" -A2 -B2
echo -e "\n=== token_auth usages in Python files ==="
rg -n --type=py "token_auth"
echo -e "\n=== JIRA client instantiations ==="
rg -n --type=py "JIRA\(" -A3Repository: openshift/release-tests
Length of output: 5706
🏁 Script executed:
#!/bin/bash
# Check for the main implementation file and related jira_notificator files
echo "=== Files related to jira_notificator ==="
find . -name "*jira*" -type f | grep -v __pycache__
echo -e "\n=== Check main implementation file (if exists) ==="
if [ -f "oar/notificator/jira_notificator.py" ]; then
echo "Found oar/notificator/jira_notificator.py"
grep -n "JIRA(" oar/notificator/jira_notificator.py | head -20
fiRepository: openshift/release-tests
Length of output: 400
🏁 Script executed:
#!/bin/bash
# Check oar/core/jira.py for the server configuration
echo "=== oar/core/jira.py content (first 50 lines) ==="
head -50 oar/core/jira.py
echo -e "\n=== Search for JIRA_SERVER or jira server config ==="
rg -n "get_jira_server|JIRA_SERVER" --type=py -B2 -A2
echo -e "\n=== Check config/settings files ==="
find . -name "config*" -o -name "settings*" -o -name "*constants*" | grep -v __pycache__ | head -20Repository: openshift/release-tests
Length of output: 2281
🏁 Script executed:
#!/bin/bash
# Check what get_jira_server returns in ConfigService
echo "=== Search for ConfigService and get_jira_server ==="
rg -n "class ConfigService|def get_jira_server" --type=py -A10Repository: openshift/release-tests
Length of output: 712
🏁 Script executed:
#!/bin/bash
# Look at actual usage in tests and check environment
echo "=== Check test_jira.py for any clues about server config ==="
head -100 tests/test_jira.py
echo -e "\n=== Check for any existing uses of basic_auth in codebase ==="
rg -n "basic_auth" --type=pyRepository: openshift/release-tests
Length of output: 4103
🏁 Script executed:
#!/bin/bash
# Check configstore.json for jira_server value
echo "=== oar/core/configstore.json ==="
cat oar/core/configstore.json | head -50
echo -e "\n=== Look for jira_server in the file ==="
rg -n "jira_server" oar/core/configstore.jsonRepository: openshift/release-tests
Length of output: 2357
🏁 Script executed:
#!/bin/bash
# Check for any README or documentation about Cloud migration
echo "=== Search for migration-related comments or docs ==="
rg -i "cloud|migration|atlassian" --type=md -B2 -A2 | head -50
echo -e "\n=== Check the PR/commit context if available ==="
git log --oneline -20 | head -20Repository: openshift/release-tests
Length of output: 2973
🏁 Script executed:
#!/bin/bash
# Let's verify what the jira library actually expects
# Search for any existing documentation or tests in the codebase
echo "=== Check if there are any docs about jira library usage ==="
find . -name "*.md" -o -name "*.txt" | xargs grep -l -i "jira" 2>/dev/null | head -10
echo -e "\n=== Check requirements or setup files for jira library version ==="
find . -name "requirements*" -o -name "setup.py" -o -name "pyproject.toml" | xargs grep -i jira 2>/dev/nullRepository: openshift/release-tests
Length of output: 579
🌐 Web query:
jira python library token_auth vs basic_auth Atlassian Cloud
💡 Result:
For Atlassian Cloud, use basic_auth (email + API token). In the jira (pycontribs) library docs, Cloud “API Token” auth is shown as JIRA(basic_auth=('email','API token')), while token_auth is for self‑hosted Jira (Server/Data Center) Personal Access Tokens (PATs), not Cloud. [1]
Atlassian’s Cloud REST API “basic auth” is email address + API token (passwords are deprecated). [2]
from jira import JIRA
jira = JIRA(
server="https://your-site.atlassian.net",
basic_auth=("you@company.com", "YOUR_ATLASSIAN_API_TOKEN"),
)Sources: [1] https://jira.readthedocs.io/examples.html [2] https://developer.atlassian.com/cloud/jira/platform/basic-auth-for-rest-apis/
🌐 Web query:
jira python library server URL format Atlassian Cloud /jira suffix
💡 Result:
For the pycontribs jira Python library, set server to the Jira site base URL (scheme + host, and only include a path if your Jira is deployed under a context path):
-
Atlassian Cloud (most common): use no
/jirasuffix
https://<your-site>.atlassian.net(then the library will call endpoints like/rest/api/3/...) [1], [2] -
Jira Server/Data Center with a context path: include it (this is where
/jiracomes from)
https://jira.example.com/jira(if that’s your instance’s configured base URL)
If you’re unsure what to use (especially with custom domains), query Jira for its canonical base URL and use the returned baseUrl:
GET https://<whatever-you-use-to-access-jira>/rest/api/3/serverInfo→ readbaseUrl[3], [4]
Example (Cloud):
from jira import JIRA
jira = JIRA(server="https://your-site.atlassian.net", basic_auth=("email@company.com", "api_token"))[1], [2]
Sources: [1] pycontribs/jira README, [2] Atlassian “Basic auth for REST APIs” (Feb 6, 2026), [3] Atlassian Jira Cloud REST API intro, [4] Jira Cloud /serverInfo baseUrl guidance.
🌐 Web query:
jira 3.4.1 python library Atlassian Cloud authentication
💡 Result:
For Atlassian Jira Cloud with the jira (pycontribs) Python library v3.4.1, use Basic Auth with your Atlassian account email + an API token (not your password). Atlassian documents this as the supported Basic Auth approach for Jira Cloud REST APIs. [1]
from jira import JIRA
jira = JIRA(
server="https://YOUR_SITE.atlassian.net",
basic_auth=("you@company.com", "YOUR_ATLASSIAN_API_TOKEN"),
)Notes:
- The
jiralibrary’sJIRA(...)client supports authentication viabasic_auth(and alsotoken_auth, which is for bearer/PAT-style tokens, typically used in non-Cloud scenarios). [2] - For integrations, Atlassian recommends OAuth 2.0 (3LO) over Basic Auth for most non-personal-script use cases; Basic Auth is mainly for simple scripts/bots. [3]
Sources
- Atlassian: “Basic auth for REST APIs” (email address + API token; passwords deprecated) [1]
jiralibrary docs:JIRA(...)authentication viabasic_auth/token_auth[2]- Atlassian: “Security for other integrations” (OAuth 2.0 (3LO) recommended; Basic Auth for scripts) [3]
Correct server URL and authentication method for Atlassian Cloud compatibility.
Two required fixes for Atlassian Cloud:
-
Server URL: Remove the
/jirasuffix. Atlassian Cloud useshttps://redhat.atlassian.net(the/jirapath segment is specific to self-hosted Server/Data Center deployments). Thejiralibrary appends REST paths directly to the server value, so including/jirawould produce malformed endpoints like.../jira/rest/api/3/...instead of.../rest/api/3/.... -
Authentication: Change from
token_authtobasic_auth=(email, api_token). Thetoken_authparameter is for self-hosted Jira Personal Access Tokens (PATs), not Atlassian Cloud. Cloud requires HTTP Basic Auth with the user's email and API token, as documented in Atlassian's official guidance.
Correct Cloud-compatible pattern:
self.jira = JIRA(
server="https://redhat.atlassian.net",
basic_auth=(os.environ.get("JIRA_EMAIL"), jira_token),
)Note: The same issues exist in oar/notificator/jira_notificator.py line 709 and potentially in oar/core/jira.py if ConfigStore returns the Cloud URL. Ensure consistency across all JIRA client instantiations.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@tests/test_jira_notificator.py` at line 16, The JIRA client instantiation
uses an incorrect server URL and auth method: update the JIRA(...) call that
assigns self.jira (and any other JIRA instantiations such as in
jira_notificator.py and oar/core/jira.py where ConfigStore may supply the URL)
to use server="https://redhat.atlassian.net" (remove the "/jira" suffix) and
switch from token_auth to basic_auth using the email and API token
(basic_auth=(os.environ.get("JIRA_EMAIL"), jira_token)), and ensure all other
JIRA client constructions follow the same pattern for Atlassian Cloud
compatibility.
2 participants
This is draft PR. The URLs have to be checked when the Atlassian Cloud migration is done.
Summary by CodeRabbit
Bug Fixes
Documentation