-
Notifications
You must be signed in to change notification settings - Fork 21
Storage docs #121
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Storage docs #121
Conversation
preview available: https://docs.tds.cscs.ch/121 |
1 similar comment
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
… storage-refactor
preview available: https://docs.tds.cscs.ch/121 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Superficial nitpicks. I hope someone who knows a bit more about filesystems can do a deeper review.
Potential restructuring idea already mentioned to @bcumming offline: Add separate sections about filesystem types, i.e. Lustre and VAST, and link to those from each of the filesystem sections. This means that one can avoid repeating e.g. warnings about performance degradation when the fileystem is close to full. It may become more complicated to navigate though, but worth considering.
| | [backups][ref-storage-backups] | [snapshot][ref-storage-snapshots] | [cleanup][ref-storage-cleanup] | access | | ||
| --------- | ---------- | ---------- | ----------- | --------- | | ||
| [Home][ref-storage-home] | yes | yes | no | user | | ||
| [Scratch][ref-storage-scratch]| no | no | yes | user | |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Scratch default access is as far as I know user and primary group, so access should be primary group
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The point here is that every user gets a scratch path in /<scratch>/$USER
.
/capstor/scratch
contains a single path cscs
, which contains all users... from what I can tell.
```console | ||
$ ls $HOME/.snapshot | ||
big_catalog_2025-05-21_08_49_34_UTC | ||
big_catalog_2025-05-21_09_19_34_UTC |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What is big_catalog_*
? It is unclear why it is there, and what to expect from it.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
No idea. Maybe the Storage guys can chip in with an explanation.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We are investigating - it looks like it is an "implementation artifact" of the VAST file system.
We have reached out to them for clarification.
I will add it to the docs when they reply (and if the information is useful)
## Common Questions | ||
|
||
??? question "My files are gone, but the directories are still there" | ||
When the [cleanup policy][ref-storage-cleanup] is applied on LUSTRE file systems, the files are removed, but the directories remain. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, you describe the observation, but you do not provide the reason for this behaviour. Why is it a good idea to keep the directories, and not the files?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I have no idea why we keep the directories - and I don't think explaining the why changes the observation (which is strange the first time you see it)
If I had to guess, it looks like the following is run:
find /capstor/scratch/cscs/$username -type f -atime +30 -delete
Co-authored-by: Mikael Simberg <[email protected]>
preview available: https://docs.tds.cscs.ch/121 |
Co-authored-by: Mikael Simberg <[email protected]>
preview available: https://docs.tds.cscs.ch/121 |
Co-authored-by: Mikael Simberg <[email protected]>
Co-authored-by: Mikael Simberg <[email protected]>
preview available: https://docs.tds.cscs.ch/121 |
2 similar comments
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
preview available: https://docs.tds.cscs.ch/121 |
This is a draft of the new storage documentation.
do not merge
https://docs.tds.cscs.ch/121/storage/filesystems/