Open Bug 1661711 Opened 4 years ago Updated 10 months ago

Handle the remaining cases that if the origin stored in the metadata file mismatches the origin directory name

Categories

(Core :: Storage: Quota Manager, enhancement, P3)

enhancement

Tracking

()

People

(Reporter: tt, Unassigned)

References

Details

(Keywords: leave-open, Whiteboard: [storage-jv])

Attachments

(2 files)

No description provided.
Status: NEW → ASSIGNED

If the origin name that stored in the directory metadata file doesn't match
its origin directory name, it would cause temporary storage initialization to
fail.

We have handle the origins that with a trailing dot in bug 1659053.
Here, we want to handle the remaining unknown cases.

The policy is:

  • Try to restore the origin directory name base on the origin name that stored
    in the directory metadata file.
    • If the target origin exists, we put the origin directory name into a deny
      list.
    • If the target origin doesn't exists, we should be okay.

Once the origin directory name is put into the list, that means

  • Such directories would be ignored by normal accesses
    (e.g. InitStorageAndOrigin, ListOrigin, and GetOriginUsage)
  • Such directories wouldn't be ignored by ClearOp

Depends on D89704

Attachment #9174850 - Attachment description: Bug Bug 1661711 - Use Result<V, E> for GetDirectoryOrigin and use *_TRY_VAR for its callsites; → Bug 1661711 - Use Result<V, E> for GetDirectoryOrigin and use *_TRY_VAR for its callsites;
Attachment #9174850 - Attachment description: Bug 1661711 - Use Result<V, E> for GetDirectoryOrigin and use *_TRY_VAR for its callsites; → Bug 1661711 - Use Result<V, E> for GetDirectoryOrigin and use *_TRY_VAR for its callsites;
Keywords: leave-open
Pushed by ttung@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/b05355f2f5f7
Use Result<V, E> for GetDirectoryOrigin and use *_TRY_VAR for its callsites; r=sg,dom-workers-and-storage-reviewers

The leave-open keyword is there and there is no activity for 6 months.
:jstutte, maybe it's time to close this bug?

Flags: needinfo?(jstutte)

I think the priority of D89705 depends on the result from the telemetry of QM_TRY.

The patch might still have value but we probably don't have time to get into that if there are no related failures to that.

Flags: needinfo?(jstutte) → needinfo?(jvarga)

The leave-open keyword is there and there is no activity for 6 months.
:jstutte, maybe it's time to close this bug?

Flags: needinfo?(jstutte)
Flags: needinfo?(jstutte)

The bug assignee didn't login in Bugzilla in the last 7 months.
:jstutte, could you have a look please?
For more information, please visit auto_nag documentation.

Assignee: shes050117 → nobody
Status: ASSIGNED → NEW
Flags: needinfo?(jstutte)
Assignee: nobody → jvarga
Status: NEW → ASSIGNED
Flags: needinfo?(jstutte)
Assignee: jvarga → nobody
Status: ASSIGNED → NEW
Flags: needinfo?(jvarga)
Whiteboard: [storage-jv]

The leave-open keyword is there and there is no activity for 6 months.
:edenchuang, maybe it's time to close this bug?
For more information, please visit auto_nag documentation.

Flags: needinfo?(echuang)
Assignee: nobody → jvarga
Status: NEW → ASSIGNED

Hi Jan, as the patch is assigned to you, the bug will stick with you, too. Should we find someone else to work on this?

Flags: needinfo?(echuang) → needinfo?(jvarga)

(In reply to Jens Stutte [:jstutte] from comment #11)

Hi Jan, as the patch is assigned to you, the bug will stick with you, too. Should we find someone else to work on this?

Yeah, we could try to improve this area of code, but I think it's rather a low priority at the moment.

Assignee: jvarga → nobody
Severity: -- → S3
Status: ASSIGNED → NEW
Flags: needinfo?(jvarga)
Priority: -- → P3
You need to log in before you can comment on or make changes to this bug.