Bug 48863
| Summary: | Write access to VFAT-mounted directories is broken | ||
|---|---|---|---|
| Product: | [Retired] Red Hat Linux | Reporter: | djh <djh> |
| Component: | kernel | Assignee: | Arjan van de Ven <arjanv> |
| Status: | CLOSED WONTFIX | QA Contact: | David Lawrence <dkl> |
| Severity: | medium | Docs Contact: | |
| Priority: | low | ||
| Version: | 7.1 | CC: | teg |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | i386 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2003-06-06 14:39:23 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
Using truncate to increase file size is not allowed in the 2.4 kernel, and is not a bug in samba... reassigning to kernel (but I doubt they'll say much more than "your application is broken", which is true enough) |
Description of Problem: When sharing a directory on a VFAT-mounted fs via samba with any 2.4.x kernel, writes to that shared drive will usually fail. Changes introduced into the kernel somewhere after 2.4.0-test12 break growing ftruncate()s on VFAT-mounted filesystems. How Reproducible: 100% reproducible Steps to Reproduce: 1. share a local VFAT directory via samba (with appropriate permissions) 2. copy a large-ish file from a win98se client to the shared directory 3. watch the permission denied message appear Actual Results: It doesn't work. ("Cannot create or replace <file> Access is denied. Make sure the disk is not full or write-protected and that the file is not currently in use.") Expected Results: It should work, or at least be documented as not working. It does with 2.2.x (last time I tried), and early 2.4.0-test kernels. Additional Information: I meant to report this about 7 months ago (whoops). I'm still not sure whether this should be filed under kernel or samba, and I'm even less sure that anyone really cares about this problem. But IMVHO, one of the following things should be done: - it should be documented somewhere that it's broken - growing truncates should be fixed on FAT filesystems (IIRC, Al Viro made an experimental patch a long time ago) Others have suggested that the application needs fixing, as it's relying on undocumented behaviour. Another suggestion was to use a decent fs instead. :) This is related to bug #39547. If it's not going to be fixed, it should at least be documented as not working.