Bug 592500 - Review Request: libeatmydata - Disable fsync() and friends
Review Request: libeatmydata - Disable fsync() and friends
Product: Fedora
Classification: Fedora
Component: Package Review (Show other bugs)
All Linux
medium Severity medium
: ---
: ---
Assigned To: Nobody's working on this, feel free to take it
Fedora Extras Quality Assurance
Depends On:
  Show dependency treegraph
Reported: 2010-05-14 21:39 EDT by Zing
Modified: 2015-01-23 11:58 EST (History)
5 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2012-06-29 18:27:01 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)

  None (edit)
Description Zing 2010-05-14 21:39:02 EDT
Spec URL: http://shishz.fedorapeople.org/libeatmydata.spec
SRPM URL: http://shishz.fedorapeople.org/libeatmydata-15-1.fc12.src.rpm

LD_PRELOAD library that disables all forms of writing data safely to
disk. fsync() becomes a NO-OP, O_SYNC is removed etc. The idea is to
use in testing to get faster test runs where real durability is not

DO NOT use libeatmydata on software where you care about what it stores.
It's called libEAT-MY-DATA for a reason.
Comment 1 Jason Tibbitts 2010-11-19 12:54:01 EST
I think there's a newer upstream release available.

Here's the rpmlint output; you should provide this yourself when submitting packages:

  libeatmydata.x86_64: E: no-ldconfig-symlink /usr/lib64/libeatmydata.so.1.0
Not sure how problematic this is, honestly.

  libeatmydata.x86_64: W: shared-lib-calls-exit
   /usr/lib64/libeatmydata.so.1.0 _exit@GLIBC_2.2.5
Generally libraries shouldn't call exit.  Especially this one.  Any idea why it does so?

  libeatmydata.x86_64: W: no-manual-page-for-binary eatmydata
It's nice to have manual pages but not essential.

  libeatmydata-debuginfo.x86_64: E: debuginfo-without-sources
This indicates something wrong with the build process, either because something strips them or because the compiler flags aren't correct.  In this case, it's the compiler flags.  You need to somehow make the package build process use %{optflags} (or $RPM_OPT_FLAGS; they're the same).
Comment 2 Zing 2010-12-06 22:54:22 EST
Although I submitted this, if anyone wants to pick this up and maintain this for fedora, they are welcome to.
Comment 3 Rudd-O DragonFear 2012-07-16 21:35:09 EDT
Will someone pick this up?
Comment 4 Rudd-O DragonFear 2012-07-16 21:57:47 EDT
I have a better and upstream-up-to-date spec file


De nada.
Comment 5 Dave Love 2015-01-23 11:58:37 EST
For what it's worth, there's a currently-current srpm with several updates at
and a copr build under http://copr.fedoraproject.org/coprs/loveshack/livhpc

Note You need to log in before you can comment on or make changes to this bug.