Hi all,
while the cryosparc update v3.0.0 on 5 of 6 computers worked fine, one update wiped the database (please backup as recommended) and somewhat killed my anaconda environment. Warning: conda environment not found; this indicates that a cryoSPARC installation is either incomplete or in progress
Nothing conda related in .bashrc, and everything conda related deactivated using:
Still giving me a Warning: conda environment not found; this indicates that a cryoSPARC installation is either incomplete or in progress
So I did a clean install and noticed that the flags now act up ./install.sh --standalone --license $LICENSE_ID --worker_path /my/worker/path--cudapath /usr/local/cuda --ssdpath /my/ssd/path/ --initial_name "Daniel Lastname" --initial_email "my@email" --initial_password "my_password"
results in a Unknown option: --initial_name
When I work around it using the direct assignments from install.sh
–initial_firstname
–initial_lastname
–initial_username
The installation runs through, however, upon the first start CryoSparc gives me
Creating the initial cryoSPARC user…
Improper usage.
So I created the username manually using ./cryosparcm createuser --email "my@email" --password "my_password" --username "Daniel" --firstname "Daniel" --lastname "Lastname"
And I can log into the web interface.
Has the --initial_name flag changed in v3.0.0 or is my system now really messed up?
Hi @mannda, cryoSPARC v3 includes its own embedded conda installation and conda environment. Are you saying that updating to cryoSPARC affected your own previously-installed conda environment?
You may see the Warning: conda environment not found; this indicates that a cryoSPARC installation is either incomplete or in progress during the update but this should be gone after the install completes. The clean install should fix it, as you’ve done.
The interface for creating a new user did change but the latest docs aren’t live yet, I’ll post the updates shortly.
Hey @nfrasser, I’m not sure what happened to the conda env, it’s working fine now with the clean install. My main point was the --initial_name flag; thanks for the quick clarification and the update of the documentation.
Many hearts from Germany for including helical processing!
Cheers,
Dan
Same issue here. Although this workaround does help in creating a user, the error actually happens even before installing the worker node. And because of that, there are no lanes showing up when the job is queued. Tried reinstalling from scratch several times and have run into the same issue every time. Not sure what is going on.
We’ve updated the packages to include a fix for the createuser function- you should be able to re-download and re-do the single workstation installation.
Hi,
I am having a similar problem with “unknown option” when trying to run ./install.sh. I have pasted my installation commands below below.
$LICENSE_ID is defined -> echo $LICENSE_ID shows my license code. It is the code I have used for earlier versions. (A hard disk failure has required me to reinstall from scratch rather than update).
I get "Unkown option: --cudapath when I run the following:
I tried wrapping the LICENSE_ID in quotes but still the same problem. Perhaps a way to confirm a license issue would be to request a new license and see if this fixes the problem?