I'm still pretty new to Python as well as Django so I have a situation I'm not sure how to resolve.
Main issue is that on deploy of my code to dev, deployment fails, to stage or prod, it passes.
I worked on an issue where I had to drop some columns in a table in our app.
After making the changes, I deployed to dev and asked for a code review.
In code review, it was suggested I change the name of the migration file to something more descriptive rather than just leaving it 0018_auto_.
I made that change and deployed to dev and stage. Dev failed (when I expected it to succeed) because the new name was seen and django tried to drop columns that no longer exist. In stage, the name was never changed and the columns were dropped for the first time using that new name of the file.
So stage deploys just fine.
How do I resolve this error on dev so it recognizes this migration already took place?
Thanks!
If you are 100% sure that the only change done for 0018 was the file name, and that it is the last migration for your app, you can do:
python manage.py migrate myapp 0017 --fake # Move to 17 without removing changes done by 18
python manage.py migrate myapp 0018 --fake # 18 is already applied, so just migrate with the new filename
This will keep the changes that are present in 0018, and just apply the changes to the filename.
Django constructs a table named djang_migrations that keeps track of the migrations that have been applied. If you later change the name of a migration file, then it will appear as if that migration did not run.
You can rename the migration in the django_migrations table of your development environment with:
UPDATE django_migrations
SET name = '0018_new_name'
WHERE app = '%%%name_of_the_app'
AND name = '0018_auto_'
where 0018_new_name is the new name of the migration file.
By updating this name, Django should now consider the new migration file as done.
All these were great answers and will no doubt help others now and in the future!
I tried all these answers but only one thing worked. The long way!
For full visibility (this worked locally and I understand --fake now in practice, thanks #brian-destura!):
python manage.py migrate myapp 0017 --fake # Move to 17 without removing changes done by 18
python manage.py migrate myapp 0018 --fake # 18 is already applied, so just migrate with the new filename
In deployment through gitlab, our django / zappa app just didn't want to deploy successfully though (even with many different solutions tried).
In the end, the only way to fix the issue was to update the 0018 migration file to add back all those fields, rollback the changes in the models and serializers, and re-deploy to dev.
After that, revert all those changes including the migration file, and re-deploy again to dev.
There's probably some easier way to do this but with all those with more backend knowledge out for the holidays (I'm a front end originally with more backend knowledge/projects through the last year), I was very happy to resolve and release my changes.
Thanks to you guys for some things to try, ideas, and def knowledge I could use to create this, my own long winded solution!
I have an application in Django 1.6.5. I have a model where I removed one field, I added another field, and the third upgraded. And when we now turn to the model in the admin panel, I get the message:
ProgrammingError at /admin/app/subscription/
column app_subscription.enabled does not exist
The command python manage.py syncdb does not work.
Django (hopefully) doesn't modify your database schema if you don't explicitely ask for it. The syncdb command works perfectly, but (as documented) it will only create tables that don't yet exists (and are not marked as being managed externally in your models).
So you have mostly three options here:
manually drop your table and re-run syncdb. This mean you will loose all our data, so it's hardly a "solution"
manually alter your database schema. You won't loose your data, but you'll have to repeat the same (manual) operation everywhere your app is deployed... If it's only installed on your local workstation that might be ok, else it's not a reliable professional production-level option.
Use South (which seems to be installed since you do have a migrate command available.
Note that solution #3 imply that you do create the migration files for your app, as documented here : http://south.readthedocs.org/en/latest/tutorial/part1.html#the-first-migration
It just happened that I faced the same issue with django 1.9.x, where I added a new field in my django app which triggered the same error as you mentioned above.
I logged into the dbshell environment using
python manage.p dbshell # I know some use ./manage.py
and dropped all of my tables by running the following command within the dbshell to drop the tables
your_psql=# drop schema public cascade;
This will drop all of your tables (be careful as you may lose your data, there away to keep the data!) and you will get a message right after executing this command tells you that all dropped. Right after that run the following command to create the schema again, otherwise your server will not run:
your_psql=# create schema public;
Then just do the
python manage.py makemigrations # you might not need this, and
python manage.py migrate
And you're ready to go.
I know this answer might be very late but I hope it will help someone.
Cheers
I'm in the process of preparing a Django application for its initial production release, and I have deployed development instances of it in a few different environments. One thing that I can't quite get happening as smoothly as I'd like is the initial database migration. Given a fresh installation of Django, a deployment of my application from version control, and a clean database, manage.py migrate will handle the initial creation of all tables (both Django's and my models'). That's great, but it doesn't actually create the initial migration files for my apps. This leads to a problem down the road when I need to deploy code changes that require a new database migration, because there's no basis for Django to compute the deltas.
I've tried running manage.py makemigrations as the first step in the deployment, in the hopes that it would create the migration files, but it reports that there are no changes to migrate. The only way I've found to get the baseline state that I need is to run manage.py makemigrations [appname] for each of my apps. Shouldn't makemigrations, called without a specific app name, pick up all the installed apps and create their migrations? Where am I going wrong?
You're going wrong at the very end -- yes, you do need to call manage.py makemigrations <appname> for each of your apps once. It's not automatically done for all apps.
Presumably that is because Django has no way of knowing if that is what you want to do (especially if some apps were downloaded from PyPI, etc). And a single command per app can't really be an extreme amount of work, right?
As the title says, I can't seem to get migrations working.
The app was originally under 1.6, so I understand that migrations won't be there initially, and indeed if I run python manage.py migrate I get:
Operations to perform:
Synchronize unmigrated apps: myapp
Apply all migrations: admin, contenttypes, auth, sessions
Synchronizing apps without migrations:
Creating tables...
Installing custom SQL...
Installing indexes...
Running migrations:
No migrations to apply.
If I make a change to any models in myapp, it still says unmigrated, as expected.
But if I run python manage.py makemigrations myapp I get:
No changes detected in app 'myapp'
Doesn't seem to matter what or how I run the command, it's never detecting the app as having changes, nor is it adding any migration files to the app.
Is there any way to force an app onto migrations and essentially say "This is my base to work with" or anything? Or am I missing something?
My database is a PostgreSQL one if that helps at all.
If you're changing over from an existing app you made in django 1.6, then you need to do one pre-step (as I found out) listed in the documentation:
python manage.py makemigrations your_app_label
The documentation does not make it obvious that you need to add the app label to the command, as the first thing it tells you to do is python manage.py makemigrations which will fail. The initial migration is done when you create your app in version 1.7, but if you came from 1.6 it wouldn't have been carried out. See the 'Adding migration to apps' in the documentation for more details.
This may happen due to the following reasons:
You did not add the app in INSTALLED_APPS list in settings.py
(You have to add either the app name or the dotted path to the subclass of AppConfig in apps.py in the app folder, depending on the version of django you are using). Refer documentation: INSTALLED_APPS
You don't have migrations folder inside those apps. (Solution: just create that folder).
You don't have __init__.py file inside migrations folder of those apps. (Solution: Just create an empty file with name __init__.py)
You don't have an __init__.py file inside the app folder. (Solution: Just create an empty file with name __init__.py)
You don't have a models.py file in the app
Your Python class (supposed to be a model) in models.py doesn't inherit django.db.models.Model
You have some semantic mistake in definition of models in models.py
Note:
A common mistake is to add migrations folder in .gitignore file. When cloned from remote repo, migrations folder and/or __init__.py files will be missing in local repo. This causes problem.
Migration files are supposed to be included in the repo. read here. If your team frequently faces migration issues you may consider ignoring migration files as follows:
I suggest to gitignore migration files by adding the following lines to .gitignore file
*/migrations/*
!*/migrations/__init__.py
Remember, it is not recommended to gitignore migration files as per django documentation
Ok, looks like I missed an obvious step, but posting this in case anyone else does the same.
When upgrading to 1.7, my models became unmanaged (managed = False) - I had them as True before but seems it got reverted.
Removing that line (To default to True) and then running makemigrations immediately made a migration module and now it's working. makemigrations will not work on unmanaged tables (Which is obvious in hindsight)
My solution was not covered here so I'm posting it. I had been using syncdb for a project–just to get it up and running. Then when I tried to start using Django migrations, it faked them at first then would say it was 'OK' but nothing was happening to the database.
My solution was to just delete all the migration files for my app, as well as the database records for the app migrations in the django_migrations table.
Then I just did an initial migration with:
./manage.py makemigrations my_app
followed by:
./manage.py migrate my_app
Now I can make migrations without a problem.
Agree with #furins. If everything seems to be in order and yet this problem arises, checkout if there is any property method with same title as the attribute which you are trying to add in the Model class.
Remove method with similar name as attribute you are adding.
manage.py makemigrations my_app
manage.py migrate my_app
Add the methods back.
This is kind of a stupid mistake to make, but having an extra comma at the end of the field declaration line in the model class, makes the line have no effect.
It happens when you copy paste the def. from the migration, which itself is defined as an array.
Though maybe this would help someone :-)
Maybe I am too late but did you try to have a migrations folder in your app with a __init__.py file in it?
Maybe this will help someone. I was using a nested app. project.appname and I actually had project and project.appname in INSTALLED_APPS. Removing project from INSTALLED_APPS allowed the changes to be detected.
The answer is on this stackoverflow post, by cdvv7788 Migrations in Django 1.7
If it is the first time you are migrating that app you have to use:
manage.py makemigrations myappname Once you do that you can do:
manage.py migrate If you had your app in database, modified its model
and its not updating the changes on makemigrations you probably havent
migrated it yet. Change your model back to its original form, run the
first command (with the app name) and migrate...it will fake it. Once
you do that put back the changes on your model, run makemigrations and
migrate again and it should work.
I was having the exact same trouble and doing the above worked perfectly.
I had moved my django app to cloud9 and for some reason I never caught the initial migration.
Following worked for me:
Add the app name to settings.py
use 'python manage.py makemigrations'
use 'python manage.py migrate'
Worked for me: Python 3.4, Django 1.10
People like me who don't like migrations can use steps below.
Remove changes what you want to sync.
Run python manage.py makemigrations app_label for the initial migration.
Run python manage.py migrate for creating tables before you make changes.
Paste changes which you remove at first step.
Run 2. and 3. steps.
If you confused any of these steps, read the migration files. Change them to correct your schema or remove unwanted files but don't forget to change next migration file's dependencies part ;)
I hope this will help someone in future.
You want to check the settings.py in the INSTALLED_APPS list and make sure all the apps with models are listed in there.
Running makemigrations in the project folder means it will look to update all the tables related to all the apps included in settings.py for the project. Once you include it, makemigrations will automatically include the app (this saves a lot of work so you don't have to run makemigrations app_name for every app in your project/site).
Just in case you have a specific field that does not get identified by makemigrations: check twice if you have a property with the same name.
example:
field = django.db.models.CharField(max_length=10, default = '', blank=True, null=True)
# ... later
#property
def field(self):
pass
the property will "overwrite" the field definition so changes will not get identified by makemigrations
Adding this answer because only this method helped me.
I deleted the migrations folder run makemigrations and migrate.
It still said: No migrations to apply.
I went to migrate folder and opened the last created file,
comment the migration I wanted(It was detected and entered there)
and run migrate again.
This basically editing the migrations file manually.
Do this only if you understand the file content.
Make sure your model is not abstract. I actually made that mistake and it took a while, so I thought I'd post it.
Did u use schemamigration my_app --initial after renaming old migration folder? Try it. Might work. If not - try to recreate the database and make syncdb+migrate. It worked for me...
In my case I needed to add my model to the _init_.py file of the models folder where my model was defined:
from myapp.models.mymodel import MyModel
I had mistakely deleted folder of migrations from my project directory.
Solution is to create __init__.py file in the migrations folder, and then,
python manage.py makemigrations
python manage.py migrate
Had the same problem
Make sure whatever classes you have defined in models.py, you must have to inherit models.Model class.
class Product(models.Model):
title = models.TextField()
description = models.TextField()
price = models.TextField()
I had the same problem with having to run makemigrations twice and all sorts of strange behaviour. It turned out the root of the problem was that I was using a function to set default dates in my models so migrations was detecting a change every time I ran makemigrations. The answer to this question put me on the right track: Avoid makemigrations to re-create date field
I recently upgraded Django from 1.6 to 1.8 and had few apps and migrations for them. I used south and schemamigrations for creating migrations in Django 1.6, which is dropped in Django 1.8.
When I added new models after upgrade, the makemigrations command wasn't detecting any changes. And then I tried the solution suggested by #drojf (1st answer), it worked fine, but failed to apply fake initial migration (python manage.py --fake-initial). I was doing this as my tables (old tables) were already created.
Finally this worked for me, removed new models (or model changes) from models.py and then had to delete (or rename for safety backup) migrations folder of all apps and run python manage.py makemigrations for all apps, then did python manage.py migrate --fake-initial. This worked like a charm. Once initial migration is created for all apps and fake initial migrated, then added new models and followed regular process of makemigrations and migrate on that app. The changes were detected now and everything went fine.
I just thought of sharing it here, if someone faces same problem (having schemamigrations of south for their apps), it might help them :)
Maybe that can help someone, I had the same problem.
I've already created two tables with the serializer class and the views.
So when I wanted to updated, I had this error.
I followed this steps:
I made .\manage.py makemigrations app
I executed .\manage.py migrate
I erased both tables of my models.py
I erased all reference to my tables from serializer and view class.
I executed step 1 and 2.
I retrieved my changes just in the models.py
I executed again step 5.
I restored all my changes.
If you're working with Pycharm, local history is very helpfull.
Maybe this will help someone.
I've deleted my models.py and expected makemigrations to create DeleteModel statements.
Remember to delete *.pyc files!
./manage makemigrations
./manage migrate
Migrations track changes to DB so if youre changing from unmanaged to managed, you'll need to make sure that youre database table is up to date relating to the Model you're dealing with.
If you are still in dev mode, I personally decided to delete the migration files in my IDE as well as in the django_migrations table relating to my Model and rerun the above command.
REMEMBER: if you have a migration that ends with _001 in your IDE & _003 in your database. Django will only see if you have a migration ending with _004 for anything to update.
The 2 (code & db migrations) are linked and work in tandem.
Happy coding.
You may need to fake the initial migrations using the command below
python manage.py migrate --fake-initial
Remove changes what you want to sync.
Run python manage.py makemigrations app_label for the initial migration.
Run python manage.py migrate for creating tables before you make changes.
Paste changes which you remove at first step.
Run 2. and 3. steps
Added this answer because none of other available above worked for me.
In my case something even more weird was happening (Django 1.7 Version), In my models.py I had an "extra" line at the end of my file (it was a blank line) and when I executed the python manage.py makemigrations command the result was: "no changes detected".
To fix this I deleted this "blank line" that was at the end of my models.py file and I did run the command again, everything was fixed and all the changes made to models.py were detected!
First this solution is applicable to those who are facing the same issue during deployment on heroku server, I was facing same issue.
To deploy, there is a mandatory step which is to add django_heroku.settings(locals()) in settings.py file.
Changes:
When I changed the above line to django_heroku.settings(locals(), databases=False), it worked flawlessly.
I have encountered this issue, the command
python manage.py makemigrations
worked with me once I saved the changes that I made on the files.
One of the cause may be You didn't register your models in admin.py file .
First register your models in admin.py file then do the migrations.