Support per-workspace filesystem quota on /workspace

[csweichel]

Description

This PR implements XFS based filesystem quota for each workspace in their resp. /workspace directory.

Related Issue(s)

Fixes #4938

How to test

1. Open a workspace
2. Run df -h and you should see a limit of 50gb on /workspace

Release Notes

[workspaces] Add support filesystem quota on /workspace

Caveats

• Before this PR can be merged we should drop the nodepool 4 commit. @wulfthimm can you confirm?
• For this to work the workspace content filesystem must be XFS and mounted with prjquota. In core-dev this means we need to deploy this ws-daemon first on a nodepool so that the raid-local-disks container can mount the XFS filesystem with prjquota

[codecov]

Codecov Report

Merging #5982 (fff2e9d) into main (c5c7c6f) will increase coverage by 3.09%.
The diff coverage is 22.28%.

@@            Coverage Diff             @@
##             main    #5982      +/-   ##
==========================================
+ Coverage   19.04%   22.13%   +3.09%     
==========================================
  Files           2       13      +11     
  Lines         168     2141    +1973     
==========================================
+ Hits           32      474     +442     
- Misses        134     1606    +1472     
- Partials        2       61      +59     

Flag	Coverage Δ
components-local-app-app-linux	?
components-local-app-app-windows	?
components-ws-daemon-app	22.13% <22.28%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
components/ws-daemon/pkg/content/hooks.go	0.00% <0.00%> (ø)
components/ws-daemon/pkg/content/initializer.go	0.00% <0.00%> (ø)
components/ws-daemon/pkg/content/service.go	0.00% <0.00%> (ø)
...onents/ws-daemon/pkg/internal/session/workspace.go	50.73% <ø> (ø)
components/ws-daemon/pkg/quota/xfs.go	45.05% <45.05%> (ø)
components/local-app/pkg/auth/pkce.go
components/local-app/pkg/auth/auth.go
components/ws-daemon/pkg/resources/dispatch.go	0.00% <0.00%> (ø)
components/ws-daemon/pkg/resources/limiter.go	77.77% <0.00%> (ø)
components/ws-daemon/pkg/resources/controller.go	31.06% <0.00%> (ø)
... and 8 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c5c7c6f...fff2e9d. Read the comment docs.

[csweichel]

/hold

until it's clear if we need to remove the nodepool commit (which we need for testing)

[aledbf]

/werft run

👍 started the job as gitpod-build-cw-fix-a-4938.15

[aledbf]

Works as advertised. I fill the disk using yes abcdefghijklmnopqrstuvwxyz0123456789 > until.full

/lgtm
/approve

[roboquat]

LGTM label has been added.

Git tree hash: 4cd92f9ce637a067d02f63f2968051e50e42f04f

[aledbf]

Also checked the project is removed when the workspace is stopped

from ws-daemon when /workspace is full

xfs_quota -x -c 'report -h' 
Project quota on /mnt/workingarea (/dev/md42)
                        Blocks              
Project ID   Used   Soft   Hard Warn/Grace   
---------- --------------------------------- 
#0            20K      0      0  00 [------]
#1000          4K    50G    50G  00 [------]
#1002       50.0G    50G    50G  00 [------]
#1003        6.0M    50G    50G  00 [------]
#1004         50G    50G    50G  00 [------]

after deleting the file

bash-5.1# xfs_quota -x -c 'report -h' 
Project quota on /mnt/workingarea (/dev/md42)
                        Blocks              
Project ID   Used   Soft   Hard Warn/Grace   
---------- --------------------------------- 
#0            24K      0      0  00 [------]
#1000          4K    50G    50G  00 [------]
#1001          4K    50G    50G  00 [------]
#1002       50.0G    50G    50G  00 [------]
#1003        6.0M    50G    50G  00 [------]
#1004      222.1M    50G    50G  00 [------]

[csweichel]

@wulfthimm what would be the consequence if we merged this PR with the nodepool change?

[fullmetalrooster]

@wulfthimm what would be the consequence if we merged this PR with the nodepool change?

In my first PR which introduced nodepool-2 and nodepool-3 which have 4 local ssds XFS support has been added. nodepool-4 has only been added to give you an own nodepool where no other was working. I would suggest to remove nodepool-4 and only use the previous nodepools. Otherwise you would reduce the number of nodepools for workspaces to one. That might lead to a nodepool being filled up with daemonsets.

[csweichel]

/hold cancel

[aledbf]

/lgtm

[roboquat]

LGTM label has been added.

Git tree hash: 55ace7f7294ff6c9122772eb602387a89130ca4e

[roboquat]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: aledbf

Associated issue: #4938

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [aledbf]
• components/content-service/OWNERS [aledbf]
• components/ws-daemon/OWNERS [aledbf]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment