mount user's volume #94
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #94 +/- ##
==========================================
- Coverage 56.20% 54.37% -1.84%
==========================================
Files 6 6
Lines 306 320 +14
==========================================
+ Hits 172 174 +2
- Misses 134 146 +12 ☔ View full report in Codecov by Sentry. |
| mount_base_dir = Path(os.environ['JUPYTERHUB_MOUNT_BASE_DIR']) | ||
| hub_secrets_dir = Path(os.environ['JUPYTERHUB_SECRETS_DIR']) | ||
|
|
||
| cdm_shared_dir = Path(os.environ['CDM_SHARED_DIR']) # Legacy data volume from JupyterLab |
There was a problem hiding this comment.
Should we just get rid of this? Seems like a good time to do so if we're switching infrastructure
There was a problem hiding this comment.
hmmm I am thinking we will need to mount a shared dir for people to put shared notebooks. Also, I am planning to use admin account to move existing notebooks to user's new workspace. For the later, there are other ways.
There was a problem hiding this comment.
So it's not really legacy then?
There was a problem hiding this comment.
hmmm yea. Maybe I will remove the comments or remove the mount once we are set stone of mounting volume.
| f'{mount_base_dir}/{user_home_dir}': f'{user_home_dir}', # Global users home directory | ||
| f'{mount_base_dir}/{hub_secrets_dir}': f'{hub_secrets_dir}', | ||
| f'{mount_base_dir}/{cdm_shared_dir}': f'{cdm_shared_dir}', # Legacy data volume from JupyterLab |
There was a problem hiding this comment.
Admin doesn't need the hive metastore dir?
Actually, didn't we say last time there shouldn't be a hive dir at all?
There was a problem hiding this comment.
hive is within cdm_shared_dir. So I didn't mount it here specifically.
There was a problem hiding this comment.
But didn't we say it's not needed in a recent pr?
There was a problem hiding this comment.
I think you are referring to postgres dir which is remote now so we don't need it. I think users need to read hive in order to read correctly.
There was a problem hiding this comment.
seems like when a namespace is created in postgres, an associated namespace.db file is created in the hive dir. So I am guessing it's needed. But we can remove it if we noticed it's no longer needed. I want to grant users broad privileges initially, then scale them back gradually based on our findings. That way we can move faster.
No description provided.