Bug 1146152

Summary:	fallocate fails with keep size mode unsupported when fs doesn't support FALLOC_FL_PUNCH_HOLE
Product:	Red Hat Enterprise Linux 6	Reporter:	Federico Simoncelli <fsimonce>
Component:	util-linux-ng	Assignee:	Karel Zak <kzak>
Status:	CLOSED NEXTRELEASE	QA Contact:	qe-baseos-daemons
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	6.6
Target Milestone:	rc
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2015-08-25 13:35:17 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Federico Simoncelli 2014-09-24 15:09:50 UTC

Description of problem:
fallocate fails with keep size mode unsupported when fs doesn't support FALLOC_FL_PUNCH_HOLE:

$ truncate -s 1024 test1
$ fallocate -p -o 0 -l 1024 test1
fallocate: keep size mode (-n option) unsupported

The error should be the same as for regular fallocate:

$ fallocate -o 0 -l 1024 test1 
fallocate: test1: fallocate failed: Operation not supported

Version-Release number of selected component (if applicable):
util-linux-ng-2.17.2-12.14.el6_5.x86_64

How reproducible:
100%

Steps to Reproduce:
1. try to run fallocate -p on a fs that doesn't support fallocate (e.g. nfs v3/4)

Actual results:
Fail with a wrong message: "fallocate: keep size mode (-n option) unsupported"

Expected results:
Fail with correct message: "fallocate failed: Operation not supported"

Comment 2 Karel Zak 2015-08-25 13:35:17 UTC

fallocate syscall differentiate between unsupported fallocate (=ENOSYS) and unsupported FALLOC_FL_KEEP_SIZE flag (=EOPNOTSUPP).

I think we want to make a difference between these two errors, but "(-n option)" in the error message is mistake as the KEEP_SIZE flag is generic and used in more situation than only for "-n".

I'm going to fix the error message (remove "-n ..") in the upstream tree. I don't think this is so critical issue that we have to fix it in RHEL update (reopen if do not agree).

Thanks for your report.