Short answer: Django 4.2 LTS reached end of life on April 7, 2026 and no longer gets security fixes. Upgrade through 5.0 and 5.1 deprecations to Django 5.2 LTS, supported until April 2028.
Django 4.2 LTS reached end of life on April 7, 2026, when its extended support period closed, according to the official Django download and support page. It no longer receives security fixes from the project. The recommended target is Django 5.2 LTS, which gets security and data-loss fixes until April 2028, reached by stepping through 5.0 and 5.1 with deprecation warnings switched on at every stop, exactly as the official upgrade guide describes.
Dates from the official calendar, checked November 21, 2026. Released dates are the first PyPI upload of each series. On that calendar, mainstream support covers security, data-loss, crash and major bugs, while extended support narrows to security and data-loss fixes. Django 5.2 is already in its extended phase, which is normal for an LTS: it still gets every security release until April 2028. The roadmap lists 6.2 as an LTS planned for April 2027 with extended support to April 2030; treat that as a plan, not a release you can install today. The official page also says that, starting with Django 2028, feature releases switch to year-based version numbers released each January, so 6.2 is the final release on the old numbering.
Why 5.2 and not 6.1? Version 6.1 is newer, but its support ends in December 2027, before 5.2's does. Moving from one LTS to the next keeps the upgrade count low, and 5.2 gives you a stable base until a 6.2 LTS is actually out.
Nothing in your application stopped working on April 7. The change is upstream: when the Django security team publishes a fix now, it ships for supported series only. The most recent security releases on the official weblog went to 6.1.3, 6.0.10 and 5.2.19 on November 3, 2026, with no 4.2 build alongside them. Each advisory published from now on is a question your team must answer by hand: does this bug exist in 4.2, and can we backport the patch ourselves?
If a platform vendor or hosting provider offers its own long-term patches for older Django, check your provider directly. We have not verified any third-party offer for this page.
Django rarely lives in one place. Monoliths, internal admin tools, a background worker that imports the ORM, and a reporting service that someone started from the same cookiecutter template can all pin different versions. Search for the pin, then confirm what is actually installed.
To speed up the file sweep, paste a requirements.txt, pyproject.toml, lockfile or Dockerfile into our EOL deadline radar, a free checker that flags each pin past its official cutoff, with the date, in one report.
The official guide, "How to upgrade Django to a newer version", recommends going one feature release at a time on the latest patch of each, and reading every set of release notes in between. From 4.2 that means the 5.0, 5.1 and 5.2 notes. These are the changes that most often surface in a 4.2 codebase.
Django 5.0 flipped the default of USE_TZ from False to True and removed support for pytz timezones, along with the USE_DEPRECATED_PYTZ setting and the is_dst argument on timezone.make_aware() and the Trunc functions. A project that never set USE_TZ explicitly will start handling aware datetimes the moment it upgrades. Set the value explicitly in settings before the jump, then audit code that calls pytz.localize() or compares naive and aware values.
The USE_L10N setting no longer exists. Also gone in 5.0: django.utils.timezone.utc, the PickleSerializer for sessions, CryptPasswordHasher, and the old table-based default form templates, since form rendering is now div-based. Custom templates that override django/forms/default.html need attention, and any session backend configured for pickling must switch to JSON before deploy.
LogoutView and logout_then_login() no longer log a user out on a GET request. A plain "Log out" link in a navigation template silently stops working. Replace it with a small form that POSTs with the CSRF token, and check any single sign-on flow that redirects to the logout URL.
The 5.1 notes list removals that were deprecated back in 4.2: Meta.index_together (replace it and generate the migration while still on 4.2), the length_is template filter, SHA1PasswordHasher and the unsalted SHA1 and MD5 hashers, the case-insensitive CICharField, CIEmailField and CITextField (still allowed in historical migrations), and the DEFAULT_FILE_STORAGE and STATICFILES_STORAGE settings. Test helpers changed as well: assertFormsetError() and assertQuerysetEqual() are gone, so old test suites can fail on import alone.
Django 5.1 requires MariaDB 10.5 or later and PostgreSQL 13 or later; 5.2 raises PostgreSQL to 14 and drops PostGIS 3.0 and GDAL 3.0 for GeoDjango users. SQLite must be 3.31.0 or newer. Check every environment, including developer laptops and CI service containers, since an old test database will fail long before production does.
Per the 5.2 notes, MySQL connections now use utf8mb4 instead of utf8, which was an alias for the deprecated utf8mb3. Legacy schemas still on utf8mb3 can request it in the OPTIONS block of DATABASES, but the better path is converting tables, especially if you are also preparing the database itself for a newer server. Our MySQL 8.0 end of life page covers that side.
FileField now raises FieldError when saving a file with no name. The admin's delete_selected action no longer sends pre_save and post_save signals for LogEntry when deleting many objects, which matters if you hook audit trails to those signals. In 5.2, single-argument aggregates such as Count and Sum raise TypeError when called with the wrong number of arguments, and the debug() context processor is no longer in the default project template.
The deprecation timeline lists what the 6.x line removes, including positional arguments to Model.save(), the check keyword on CheckConstraint, the transitional div form renderers, and the change of the forms.URLField default scheme to https. Fixing those warnings on 5.2 now makes the later hop to the next LTS much smaller.
The method is the same for every service: one feature release per step, warnings loud, tests green, then deploy. What varies is how many services you have and how much third-party code sits between you and Django.
List every service with its Django pin, Python version, database engine and version, and its top third-party packages (Django REST Framework, Celery integrations, django-storages, allauth, admin themes). For each package, record the release that first supports 5.2. Capture a baseline: test suite pass rate, coverage on views and models, and production error and latency numbers, so you can show nothing regressed.
On 4.2, run python -Wa manage.py test (or PYTHONWARNINGS=always pytest tests --capture=no) and fix every deprecation warning that comes from your own code. Then build a fresh virtualenv, install the latest 5.0 patch with python -m pip install -U "Django>=5.0,<5.1", run the suite with -Wa again, fix, and commit. Repeat for 5.1 and for 5.2. Run makemigrations at each stop so model changes such as the index_together conversion land as reviewed migrations.
Deploy 5.2 to staging with production-like data, then route a small share of traffic to a few 5.2 web workers behind your load balancer. Web nodes on 4.2 and 5.2 can share a database during the canary only if you hold back migrations that the old code cannot tolerate, so ship schema changes in a separate, backward-compatible release. Watch login, logout, timezone-sensitive reports and file uploads closely, since those match the changes above.
Roll the remaining web nodes, then workers and scheduled jobs. Clear the cache right after deploy: the official guide warns that pickled objects, including responses stored by cache_page(), are not guaranteed to be compatible across Django versions. Keep the previous image tagged and avoid destructive migrations in the same release so rolling back is a redeploy, not a restore.
Remove compatibility shims, delete the 4.2 CI jobs and old images, tighten the pin to Django>=5.2,<5.3, and turn deprecation warnings into test failures for your own modules so new 6.x warnings cannot pile up again.
A 4.2 to 5.2 move usually needs a senior Django engineer who has done LTS jumps before and knows where the ORM and auth changes hide, a backend engineer to raise test coverage around timezones, forms and file handling, and a platform engineer for images, CI matrices and the rolling deploy. If several services share the work, a tech lead keeps the package support matrix and sequencing honest. See our Django developers and Python developers, or Python development services.
Many teams begin with a pod of 3 to 5 senior engineers (a tech lead plus engineers), and a full pod is about seven. Ryz engineers on your team start from $7,000 per month (our published range is $7,000–$15,000 per engineer per month, mostly mid-level to senior), quoted per team. Once you say go, engineers can typically start in under two weeks. Ryz engineers work on your team, reporting to your leads. To talk specifics, scope your team with us.
It runs, but every new Django security advisory now ships without a 4.2 build, so your exposure grows with each one. Short term, stay on the final 4.2 patch you can get, keep the admin and other sensitive URLs off the public internet where possible, and start the 5.0 step now rather than waiting for a specific advisory to force it.
Django 5.2 LTS for nearly every team, since its security support runs until April 2028, longer than 6.0 or 6.1. Get there through 5.0 and 5.1 on their latest patches. A 6.2 LTS is on the official roadmap for April 2027, and fixing 6.x deprecation warnings on 5.2 keeps that next move small.
Five drivers set the pace: the number of services that pin Django, how strong your test coverage is on auth, forms and datetime handling, how many third-party packages need new releases, whether Python or the database must move first, and how much custom admin or template code overrides built-ins. Codebases that already run clean with deprecation warnings enabled move fastest.
The official Django support page describes mainstream and extended support periods for each series and lists no paid option to prolong 4.2. Some platform and hosting vendors sell their own long-term patching for older open source stacks. Check your platform vendor's current terms directly, and treat any such offer as a bridge to 5.2, not a destination.
Multiply headcount by months by a monthly rate: our range is $7,000 per engineer at the low end up to $15,000 for senior people, with team leads quoted above that. You receive one quote for the whole team, together with a plan and the engineers named up front.
Questions we didn't answer? Email info@ryzlabs.com.
Thanks — your message has been sent. We’ll get back to you soon.
Something went wrong while sending your message. Please try again or email info@ryzlabs.com.