Pipenv stuck "⠋ Locking…"
Question:
Why is my pipenv stuck in the “Locking…” stage when installing [numpy|opencv|pandas]?
When running pipenv install pandas or pipenv update it hangs for a really long time with a message and loading screen that says it’s still locking. Why? What do I need to do?
Answers:
Your package(s) are being installed and your wheel is being built
Perhaps better terminology to describe this state would be ‘Building and Locking…’ or something similar.
This is especially likely to happen if you’re installing numpy, opencv, pandas, or other large packages.
What’s going on in the background is that pipenv is downloading your package and maybe building the wheel.
The remedy in this case is often a strong dose of patience.
What is Locking?
To understand more about what "Locking" is in the pipenv context you can read more here: https://pipenv.kennethreitz.org/en/latest/basics/#pipenv-lock
$ pipenv lock is used to create a Pipfile.lock, which declares all dependencies (and sub-dependencies) of your project, their latest available versions, and the current hashes for the downloaded files. This ensures repeatable, and most importantly deterministic, builds.
However, there are times when it is not just a slow/large install, but is instead an issue with your Pipfile[.lock]. If you’re fairly certain that this is the problem try pipenv lock --clear and rerun your pipenv install command, also check this thread for more information.
This is an open issue with pipenv
https://github.com/pypa/pipenv/issues/3827
I suggest go back to pip
try doing pipenv --rm – removes virtual environment
then pipenv shell – this will again initiate virtual env
then pipenv install installs all the packages again
worked for me
For folks trying to use pipenv with an existing requirements.txt file in the working dir, you may find this
Github post helpful.
Note: I also used pipenv --rm before attempting what I show.
HTH 😉
P.S. Here’s a shout out to Zebradil’s script to create a requirements.txt, in case you’re collaborating with others who don’t use pipenv.
I had this happen to me just now. Pipenv got stuck locking forever, 20+ minutes with no end in sight, and pipenv --rm didn’t help.
In the end, the problem was that I had run pipenv install "boto3~=1.21.14" to upgrade boto3 from boto3 = "==1.17.105". But I had other conflicting requirements (in my case, botocore = "==1.20.105" and s3transfer = "==0.4.2") which are boto3 dependencies. The new version of boto3 required newer versions of these two packages, but the == requirements didn’t allow that. Pipenv didn’t explain this, and just spun "Locking…" forever.
So if you run into this, I would advise you to carefully look at your Pipenv packages, see if there are any obvious conflicts, and loosen or remove package requirements where possible.
In my case, I was able to just remove the s3transfer and botocore packages from the list entirely, and rely on boto3 to fetch the necessary versions.
Why is my pipenv stuck in the “Locking…” stage when installing [numpy|opencv|pandas]?
When running pipenv install pandas or pipenv update it hangs for a really long time with a message and loading screen that says it’s still locking. Why? What do I need to do?
Your package(s) are being installed and your wheel is being built
Perhaps better terminology to describe this state would be ‘Building and Locking…’ or something similar.
This is especially likely to happen if you’re installing numpy, opencv, pandas, or other large packages.
What’s going on in the background is that pipenv is downloading your package and maybe building the wheel.
The remedy in this case is often a strong dose of patience.
What is Locking?
To understand more about what "Locking" is in the pipenv context you can read more here: https://pipenv.kennethreitz.org/en/latest/basics/#pipenv-lock
$ pipenv lock is used to create a Pipfile.lock, which declares all dependencies (and sub-dependencies) of your project, their latest available versions, and the current hashes for the downloaded files. This ensures repeatable, and most importantly deterministic, builds.
However, there are times when it is not just a slow/large install, but is instead an issue with your Pipfile[.lock]. If you’re fairly certain that this is the problem try pipenv lock --clear and rerun your pipenv install command, also check this thread for more information.
This is an open issue with pipenv
https://github.com/pypa/pipenv/issues/3827
I suggest go back to pip
try doing pipenv --rm – removes virtual environment
then pipenv shell – this will again initiate virtual env
then pipenv install installs all the packages again
worked for me
For folks trying to use pipenv with an existing requirements.txt file in the working dir, you may find this
Github post helpful.
Note: I also used pipenv --rm before attempting what I show.
HTH 😉
P.S. Here’s a shout out to Zebradil’s script to create a requirements.txt, in case you’re collaborating with others who don’t use pipenv.
I had this happen to me just now. Pipenv got stuck locking forever, 20+ minutes with no end in sight, and pipenv --rm didn’t help.
In the end, the problem was that I had run pipenv install "boto3~=1.21.14" to upgrade boto3 from boto3 = "==1.17.105". But I had other conflicting requirements (in my case, botocore = "==1.20.105" and s3transfer = "==0.4.2") which are boto3 dependencies. The new version of boto3 required newer versions of these two packages, but the == requirements didn’t allow that. Pipenv didn’t explain this, and just spun "Locking…" forever.
So if you run into this, I would advise you to carefully look at your Pipenv packages, see if there are any obvious conflicts, and loosen or remove package requirements where possible.
In my case, I was able to just remove the s3transfer and botocore packages from the list entirely, and rely on boto3 to fetch the necessary versions.