Open
Bug 191877
Opened 23 years ago
Updated 3 years ago
writing to the file stream is not returning the right error
Categories
(Core :: Networking, defect, P5)
Tracking
()
NEW
People
(Reporter: timonchristl, Unassigned)
References
Details
(Whiteboard: [necko-would-take])
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2.1) Gecko/20021130
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2.1) Gecko/20021130
When trying to save a large downloaded file to a directory on which there is a
quota set and if the hard limit of that quota is it, the following error message
appears: "%S could not be saved, because an unknown error occurred.\n\nSorry
about that. Try saving to a different location." (Defined in
mozilla/embedding/components/webbrowserpersist/locale/en-US/nsWebBrowserPersist.properties).
This error message is misleading because the actual reason is "disk full", for
which there is already a more descriptive error message.
Reproducible: Always
Steps to Reproduce:
1. Prepare a new user (let's call it "test") and give it a sufficiently small
quota (e.g soft limit 5MB, hard limit 10MB)
2. Download a file which will is larger than the user's hard limit and save it
into the home directory of user "test"
3. Wait until error message appears
Actual Results:
writeError message appears
Expected Results:
diskFull message should appear
I don't if it makes any difference, but the filesystem I used for testing this
was mounted over NFS.
![]() |
||
Comment 1•23 years ago
|
||
Sounds like we're not getting the right error code out of necko....
Assignee: blaker → dougt
Component: Download Manager → Networking: File
QA Contact: petersen → benc
bug 72892? (hard to find b/c it's in xpcom...)
I think this bug is a depends of that bug, because in theory, if we all doing
compotent-level file i/o tests, EVERY component that wrote to disk would have
this problem.
Assignee: dougt → law
Component: Networking: File → File Handling
QA Contact: benc → petersen
![]() |
||
Comment 3•23 years ago
|
||
Well, the core problem is that the error message returned from Write() is just
NS_ERROR_FAILURE, not an error that says there is no space on disk.
Do your component tests attempt to write to disk in space-limited-by-quota
conditions?
None of the file handling code is calling GetDiskSpaceAvailable (which is a
separate issue, but it wouldn't help much even if we did).
The point is, writing to the file stream is not returning the right error.
Assignee: law → dougt
Component: File Handling → Networking: File
QA Contact: petersen → benc
Updated•22 years ago
|
Summary: Error message when a download could not be saved is sometimes misleading → writing to the file stream is not returning the right error
Updated•22 years ago
|
Target Milestone: --- → Future
Comment 5•19 years ago
|
||
-> default owner
Assignee: darin → nobody
Component: Networking: File → Networking
QA Contact: benc → networking
Target Milestone: Future → ---
Updated•10 years ago
|
Whiteboard: [necko-would-take]
Comment 6•8 years ago
|
||
Bulk change to priority: https://bugzilla.mozilla.org/show_bug.cgi?id=1399258
Priority: -- → P5
Updated•3 years ago
|
Severity: normal → S3
You need to log in
before you can comment on or make changes to this bug.
Description
•