Re: mount -t aufs is slow with current aufs-utils
Hello Erik, Erik Braun: when using the newest Debian (testing) kernel 3.16.7 and its aufs.ko 3.16-20140908 and the current aufs-utils origin/aufs3.14, the mounting takes about 15 seconds for each AUFS-mount: I am afraid such slowness is caused by a mismatching of configuration. If CONFIG_AUFS_FHSM is disabled when building the kernel module but BuildFHSM=yes is set when compiling aufs-util, then such slowness will happen. Read README file in aufs-util.git and try building aufs-util with BuildFHSM=no. J. R. Okajima -- Dive into the World of Parallel Programming The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net/
mount -t aufs is slow with current aufs-utils
Hi,
when using the newest Debian (testing) kernel 3.16.7 and its aufs.ko
3.16-20140908 and the current aufs-utils origin/aufs3.14, the mounting
takes about 15 seconds for each AUFS-mount:
# mkdir -p /tmp/aufstest/ro /tmp/aufstest/rw /tmp/aufstest/dest
# time mount -t aufs -o br=/tmp/aufstest/rw=rw:/tmp/aufstest/ro=ro aufs
/tmp/aufstest/dest
real0m14.929s
user0m0.160s
sys 0m1.916s
strace shows, that this happens because many repeating (more than 45000)
sleeps:
...
ioctl(4, AGPIOC_ALLOCATE, 0) = -1 EOPNOTSUPP (Operation not supported)
nanosleep({0, 10}, NULL) = 0
ioctl(4, AGPIOC_ALLOCATE, 0) = -1 EOPNOTSUPP (Operation not supported)
2174 nanosleep({0, 10}, NULL) = 0
...
I have no idea, what this means.
This does not happen, when using Debian's aufs-tools which a version
number that seems outdated to me: 1:3.2+20130722-1.1
Should I remove the self-compiled aufs-utils and use Debians aufs-tools?
Best regards, Erik
--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
Re: mount -t aufs is slow with current aufs-utils
On Tue, 10 Mar 2015, sf...@users.sourceforge.net wrote: 3.16-20140908 and the current aufs-utils origin/aufs3.14, the mounting takes about 15 seconds for each AUFS-mount: I am afraid such slowness is caused by a mismatching of configuration. If CONFIG_AUFS_FHSM is disabled when building the kernel module but BuildFHSM=yes is set when compiling aufs-util, then such slowness will happen. Read README file in aufs-util.git and try building aufs-util with BuildFHSM=no. You are right, with BuildFHSM=no the mount process is as fast as it used to be. Thank you! Erik -- Dive into the World of Parallel Programming The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net/
Cannot install aufs-utils due to wrong strip program being used
In a previous message (sent about 5 minutes ago), I described a problem not being able to run âmake installâ for aufs-utils. Right after I sent the message I figured out the problem. My host machineâs strip was being run to strip the aufs library, but it was built with a cross-compiler. I ended up having to replace /usr/bin/strip on my host with a symlink to my toolchainâs strip temporarily, install aufs-utils, then put the original back. Is this a bug in the build system for aufs-utils? -- Want fast and easy access to all the code in your enterprise? Index and search up to 200,000 lines of code with a free copy of Black Duck Code Sight - the same software that powers the world's largest code search on Ohloh, the Black Duck Open Hub! Try it now. http://p.sf.net/sfu/bds
make install for aufs-utils fails stripping shared lib due to unrecognized format
Greetings. âsudo make installâ for aufs-utils is failing with the message âstrip: Unabletorecognisetheformatofthe input file `/s/src/aufs/stage/usr/lib/libau.so.2.7'â. The output of âfile /s/src/aufs/stage/usr/lib/libau.so.2.7â is âregular fileâ, not âshared objectâ like I would expect. I am cross-compiling. Here is the whole story: Iâm trying to build aufs for my embedded ARM system. Iâm using a kernel from TI which is linux-3.12.10-ti2013.12.01. I retrieved the standalone git repo and checked out origin/aufs3.12.x, patched my kernel, built it (not as a module), and booted it up without issue. Now Iâm trying to build the utils, which I was able to build without error. I used branch origin/aufs3.9. Here is the top portion of the Makefile I used: HOSTCC ?= cc CC = /home/steve/ti-sdk-am335x-evm-07.00.00.00/linux-devkit/sysroots/i686-arago-l inux/usr/bin/arm-linux-gnueabihf-gcc CFLAGS += -I./libau CFLAGS += -O -Wall CPPFLAGS += -I /s/src/aufs/linux-3.12.10-ti2013.12.01/usr/include -I /s/src/aufs/aufs3-standalone.git/usr/include CFLAGS += -I /s/src/aufs/linux-3.12.10-ti2013.12.01/usr/include -I /s/src/aufs/aufs3-standalone.git/usr/include DESTDIR = /s/src/aufs/stage I noticed the documentation at [1]http://aufs.sourceforge.net/ talked about adding those include directories to CPPFLAGS, but it looks like the source is C so I added them to CFLAGS as well (shouldnât hurt). I ran âsudo make installâ and got the error above. Help? Thanks, Steve References 1. http://aufs.sourceforge.net/ -- Want fast and easy access to all the code in your enterprise? Index and search up to 200,000 lines of code with a free copy of Black Duck Code Sight - the same software that powers the world's largest code search on Ohloh, the Black Duck Open Hub! Try it now. http://p.sf.net/sfu/bds
Re: Cannot install aufs-utils due to wrong strip program being used
Hello Steve, Steve Ziuchkovski: Right after I sent the message I figured out the problem. My host machine's strip was being run to strip the aufs library, but it was built with a cross-compiler. I ended up having to replace /usr/bin/strip on my host with a symlink to my toolchain's strip temporarily, install aufs-utils, then put the original back. Is this a bug in the build system for aufs-utils? How about setting make LD=/usr/toolchain/ld? J. R. Okajima -- Want fast and easy access to all the code in your enterprise? Index and search up to 200,000 lines of code with a free copy of Black Duck Code Sight - the same software that powers the world's largest code search on Ohloh, the Black Duck Open Hub! Try it now. http://p.sf.net/sfu/bds
Re: aufs-utils
Thank you. Maybe I am doing something wrong, but 'make headers_install' only installs /usr/include/linux/aufs_type.h which doesn't contain any text like EAU* I'm compiling for kernel 3.10.14 Tomas M On Thu, Oct 3, 2013 at 2:36 AM, sf...@users.sourceforge.net wrote: Hi Tomas, Tomas M: I have some strange errors while compiling aufs utils. perror.c complains that EAU_Last is undeclared, as well as the other constants starting with EAU I've git-checked aufs utils as origin/aufs3.9 Where are those supposed to be defined? They are defined in linux/include/uapi/linux/aufs_type.h. Please specify the include path by CPPFLAGS after make install_headers. J. R. Okajima -- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60134791iu=/4140/ostg.clktrk
Re: aufs-utils
Tomas M: Thank you. Maybe I am doing something wrong, but 'make headers_install' only installs /usr/include/linux/aufs_type.h which doesn't contain any text like EAU* Would you post the output of these commands? $ fgrep EAU_ ./include/uapi/linux/aufs_type.h \ ./usr/include/linux/aufs_type.h /usr/include/linux/aufs_type.h $ diff -u ./include/uapi/linux/aufs_type.h ./usr/include/linux $ diff -u ./usr/include/linux/aufs_type.h /usr Note that there are THREE aufs_type.h files after 'make headers_install'. - ./include/uapi/linux/aufs_type.h (the source) - ./usr/include/linux/aufs_type.h (modified for installing) - /usr/include/linux/aufs_type.h(installed) J. R. Okajima -- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60134791iu=/4140/ostg.clktrk
aufs-utils
I have some strange errors while compiling aufs utils. perror.c complains that EAU_Last is undeclared, as well as the other constants starting with EAU I've git-checked aufs utils as origin/aufs3.9 Where are those supposed to be defined? Thank you Tomas M -- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60134791iu=/4140/ostg.clktrk
Re: aufs-utils
Hi Tomas, Tomas M: I have some strange errors while compiling aufs utils. perror.c complains that EAU_Last is undeclared, as well as the other constants starting with EAU I've git-checked aufs utils as origin/aufs3.9 Where are those supposed to be defined? They are defined in linux/include/uapi/linux/aufs_type.h. Please specify the include path by CPPFLAGS after make install_headers. J. R. Okajima -- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60134791iu=/4140/ostg.clktrk
Re: aufs utils ver test
Ah, actually I tried
git checkout origin/aufs3.10
which returned just error
error: pathspec 'origin/aufs3.10' did not match any files
So I thought I have to checkout 3.0.
If you modify the error messages as you suggested, it won't help me.
For example if I get error message that aufs util is for aufs 3.0 but
my aufs is 3.10, I still do not know what to checkout, which version
of aufs utils is available for me.
I would suggest to modify the error message in ver.c to show the user
a command which he can use to list all available checkpoints. I do not
use git much so I do not know what is that command. Something like:
git show me all checkpoints which I can checkout to
- that will make it easy for the user to see which nearest version of
aufs tools is available for his aufs version.
Thank you for consideration
Tomas M
On Thu, Aug 22, 2013 at 9:35 PM, sf...@users.sourceforge.net wrote:
Hi,
Tomas M:
while compiling aufs utils for Linux 3.10.9, I get error that aufs is
for 3.0 and later, but aufs is 3.10-...
For aufs3.10 module, aufs-util.git#aufs3.9 branch is recommended.
If your aufs3.10 module is not so new, please try aufs-util.git#aufs3.2
branch since latest aufs-util.git#aufs3.9 branch refers to some newest
symbols in aufs3.{9,10,x-rcN} module.
aufs-util.git#aufs3.0 branch is for aufs3.[01] module only which are not
maintained now.
It seems that the message
Wrong version!
aufs3.x, but aufs is aufs3.y.
makes people confused.
I will modify as
- for aufs-util.git#aufs3.0
Wrong version!
aufs3.0 is for aufs3.[01], but aufs is aufs3.y.
- for aufs-util.git#aufs3.2
Wrong version!
aufs3.2 is for aufs3.[2-8], but aufs is aufs3.y.
- for aufs-util.git#aufs3.9
Wrong version!
aufs3.9 is for aufs3.9 and later, but aufs is aufs3.y.
- for aufs-util.git#aufs3.x-rcN
Wrong version!
aufs3.x-rcN is for aufs3.x-rcN, but aufs is aufs3.y.
Do they look good?
J. R. Okajima
--
Introducing Performance Central, a new site from SourceForge and
AppDynamics. Performance Central is your source for news, insights,
analysis and resources for efficient Application Performance Management.
Visit us today!
http://pubads.g.doubleclick.net/gampad/clk?id=48897511iu=/4140/ostg.clktrk
Re: aufs utils ver test
Tomas M: Ah, actually I tried git checkout origin/aufs3.10 which returned just error error: pathspec 'origin/aufs3.10' did not match any files So I thought I have to checkout 3.0. I see. The detail is described in README file actually. (from aufs-util.git:README) -- You will find GIT branches whose name is in form of aufs3.x where x represents the linux kernel version, linux-3.x. For instance, aufs3.0 is for linux-3.0. You may not be able to find the GIT branch in aufs-util for your version. In this case, you should git-checkout the branch for the nearest lower number. If you are using linux-3.10 and aufs3.10 (which are not released yet), but the aufs3.10 branch doesn't exit in this repository, then aufs3.9, aufs3.8, ... or something is the branch for you. Also you can view all branches by $ git branch -a -- How about this message? Wrong version! aufs3.x, but aufs is aufs3.y. See README in detail and try git branch -a. J. R. Okajima -- Introducing Performance Central, a new site from SourceForge and AppDynamics. Performance Central is your source for news, insights, analysis and resources for efficient Application Performance Management. Visit us today! http://pubads.g.doubleclick.net/gampad/clk?id=48897511iu=/4140/ostg.clktrk
Re: aufs utils ver test
How about this message? Wrong version! aufs3.x, but aufs is aufs3.y. See README in detail and try git branch -a. That would be excellent :) Thank you Tomas M -- Introducing Performance Central, a new site from SourceForge and AppDynamics. Performance Central is your source for news, insights, analysis and resources for efficient Application Performance Management. Visit us today! http://pubads.g.doubleclick.net/gampad/clk?id=48897511iu=/4140/ostg.clktrk
aufs utils ver test
Hello, while compiling aufs utils for Linux 3.10.9, I get error that aufs is for 3.0 and later, but aufs is 3.10-... Attached patch fixes that. Thanks for consideration Tomas M a.diff Description: Binary data -- Introducing Performance Central, a new site from SourceForge and AppDynamics. Performance Central is your source for news, insights, analysis and resources for efficient Application Performance Management. Visit us today! http://pubads.g.doubleclick.net/gampad/clk?id=48897511iu=/4140/ostg.clktrk
Re: aufs utils ver test
Hi,
Tomas M:
while compiling aufs utils for Linux 3.10.9, I get error that aufs is
for 3.0 and later, but aufs is 3.10-...
For aufs3.10 module, aufs-util.git#aufs3.9 branch is recommended.
If your aufs3.10 module is not so new, please try aufs-util.git#aufs3.2
branch since latest aufs-util.git#aufs3.9 branch refers to some newest
symbols in aufs3.{9,10,x-rcN} module.
aufs-util.git#aufs3.0 branch is for aufs3.[01] module only which are not
maintained now.
It seems that the message
Wrong version!
aufs3.x, but aufs is aufs3.y.
makes people confused.
I will modify as
- for aufs-util.git#aufs3.0
Wrong version!
aufs3.0 is for aufs3.[01], but aufs is aufs3.y.
- for aufs-util.git#aufs3.2
Wrong version!
aufs3.2 is for aufs3.[2-8], but aufs is aufs3.y.
- for aufs-util.git#aufs3.9
Wrong version!
aufs3.9 is for aufs3.9 and later, but aufs is aufs3.y.
- for aufs-util.git#aufs3.x-rcN
Wrong version!
aufs3.x-rcN is for aufs3.x-rcN, but aufs is aufs3.y.
Do they look good?
J. R. Okajima
--
Introducing Performance Central, a new site from SourceForge and
AppDynamics. Performance Central is your source for news, insights,
analysis and resources for efficient Application Performance Management.
Visit us today!
http://pubads.g.doubleclick.net/gampad/clk?id=48897511iu=/4140/ostg.clktrk
Re: Error while compiling aufs utils
Thank you for confirming the issue. I fixed it locally. I just wanted to assure that what I did is the correct way. Tomas M On Sun, Mar 10, 2013 at 4:46 AM, sf...@users.sourceforge.net wrote: Tomas M: The problem is that 'make headers_install' does not install aufs_type.h anywhere, as far as I know. Probably the aufs3-kbuild.patch should add one line to $LinuxSrc/include/uapi/linux/Kbuild as follows? header-y += aufs_type.h May be you are right. I think there is a bug in my release script which generates aufs3-kbuild.patch automatically by git-diff using my local branches. I am sorry. It will be fixed in next release. Until then will you fix it locally? J. R. Okajima -- Symantec Endpoint Protection 12 positioned as A LEADER in The Forrester Wave(TM): Endpoint Security, Q1 2013 and remains a good choice in the endpoint security space. For insight on selecting the right partner to tackle endpoint security challenges, access the full report. http://p.sf.net/sfu/symantec-dev2dev
Re: Error while compiling aufs utils
Tomas M: I've compiled aufs (standalone) for Linux 3.8.2, and I've copied file aufs_type.h from Linux tree to /usr/include/linux/aufs_type.h When that was done, I wanted to compile aufs utils, but 'make' returned an error: cc -I./libau -O -Wall -DMOUNT_CMD_PATH=\\ ver.c -o ver In file included from ver.c:19:0: /usr/include/linux/aufs_type.h:180:27: error: expected :, ,, ;, } or __attribute__ before * token make: *** [ver] Error 1 The released aufs_type.h file is for kernel-space only and you cannot use it for user-space. To use it in user-space (eg. aufs-util), - run make headers_install in your kernel source tree or run make install_header in your aufs3-standalone source tree, then new $PWD/usr/include/linux/aufs_type.h file will be generated. - when compiling aufs-util, you need to specify the path to aufs_type.h, for example, CPPFLAGS=-I $LinuxSrc/usr/include -I $AufsStdalone/usr/include J. R. Okajima -- Symantec Endpoint Protection 12 positioned as A LEADER in The Forrester Wave(TM): Endpoint Security, Q1 2013 and remains a good choice in the endpoint security space. For insight on selecting the right partner to tackle endpoint security challenges, access the full report. http://p.sf.net/sfu/symantec-dev2dev
Re: Error while compiling aufs utils
kirk w: Recently I've had to do a make headers_install in the kernel source. Then c= d into aufs3-standalone.git and do make install_headers, then copy those he= aders which include aufs_type.h into the kernel headers.=0A=0A=0A=0A___= Yes, I guess the bug in my release script affects only make headers_install in the patched kernel source, instead of make install_header in the aufs3-standalone source tree. J. R. Okajima -- Symantec Endpoint Protection 12 positioned as A LEADER in The Forrester Wave(TM): Endpoint Security, Q1 2013 and remains a good choice in the endpoint security space. For insight on selecting the right partner to tackle endpoint security challenges, access the full report. http://p.sf.net/sfu/symantec-dev2dev
Error while compiling aufs utils
Thank you for AUFS for 3.8 kernel.
I've compiled aufs (standalone) for Linux 3.8.2, and I've copied file
aufs_type.h from Linux tree to /usr/include/linux/aufs_type.h
When that was done, I wanted to compile aufs utils, but 'make' returned an
error:
cc -I./libau -O -Wall -DMOUNT_CMD_PATH=\\ ver.c -o ver
In file included from ver.c:19:0:
/usr/include/linux/aufs_type.h:180:27: error: expected :, ,, ;, } or
__attribute__ before * token
make: *** [ver] Error 1
After some trial and error, I found out that if I use the attached patch for
aufs_type.h, then aufs utils compile properly (but I am not sure if that
means it is correct).
--- aufs_type.h 2013-03-08 11:56:34.0 -0600
+++ aufs_type.h 2013-03-08 13:58:12.949177844 -0600
@@ -177,7 +177,7 @@
}
union au_rdu_ent_ul {
- struct au_rdu_ent __user *e;
+ struct au_rdu_ent *e;
uint64_t ul;
};
I really do not know what I am doing, I just tried to remove the __user part
and it fixed the compilation error, but that doesn't mean it is the right
thing to do. I have no idea. Does aufs utils compile for you? Am I doing
anything wrong?
Thank you very much
Tomas M
--
Symantec Endpoint Protection 12 positioned as A LEADER in The Forrester
Wave(TM): Endpoint Security, Q1 2013 and remains a good choice in the
endpoint security space. For insight on selecting the right partner to
tackle endpoint security challenges, access the full report.
http://p.sf.net/sfu/symantec-dev2dev
Re: Aufs utils
aufs2-util is essentially necessary, especially these files. (from the aufs2-util README) (...) They are always invoked at mount/remount/unmount time. The main purpose of /sbin/mount.aufs is, - maintain /etc/mtab - maintain the aufs pseudo-link (by calling /sbin/auplink) If you don't install /sbin/mount.aufs, then the aufs line in your /etc/mtab will be incorrect, and you may loose the aufs pseudo-link. /sbin/auplink is called by /sbin/umount.aufs too. If you don't know what the aufs pseudo-link is, then refer to the aufs manual in detail. Additional information, (from linux/Documentation/filesystems/aufs/design/02struct.txt) Pseudo-link -- Assume fileA exists on the lower readonly branch only and it is hardlinked to fileB on the branch. When you write something to fileA, aufs copies-up it to the upper writable branch. Additionally aufs creates a hardlink under the Pseudo-link Directory of the writable branch. The inode of a pseudo-link is kept in aufs super_block as a simple list. If fileB is read after unlinking fileA, aufs returns filedata from the pseudo-link instead of the lower readonly branch. Because the pseudo-link is based upon the inode, to keep the inode number by xino (see above) is important. All the hardlinks under the Pseudo-link Directory of the writable branch should be restored in a proper location later. Aufs provides a utility to do this. The userspace helpers executed at remounting and unmounting aufs by default. -- And http://www.mail-archive.com/aufs-users@lists.sourceforge.net/msg02639.html If you don't care about /etc/mtab and the pseudo-link, you can uninstall aufs2-util. Otherwise, just install it. Generally readdir(3) in libau.so runs faster than readdir in aufs kernel-space, particulary for the huge dir. Thank you very much for the explanation. Now it is much more clear :) -- This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
Re: Aufs utils
I've read documentation for aufs and there is some note that aufs2 utilities are always necessary. That's not true. I use a basic aufs2 system on my device and don't use or even have installed aufs utilities. It can all be done with mount -o I guess you'll need the utilities only for advanced stuff :-) -- Homepage: http://www.holgerschurig.de -- This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
Re: Aufs utils
Holger Schurig: I've read documentation for aufs and there is some note that aufs2 utilities are always necessary. That's not true. I use a basic aufs2 system on my device and don't use or even have installed aufs utilities. It can all be done with mount -o I guess you'll need the utilities only for advanced stuff :-) aufs2-util is essentially necessary, especially these files. (from the aufs2-util README) -- o /sbin/mount.aufs, /sbin/umount.aufs Helpers for util-linux-ng package. You should NOT invoke them manually. Just install them by make install. o /sbin/auplink Handles aufs pseudo-link at remount/unmount time. You can invoke it manually at anytime. -- They are always invoked at mount/remount/unmount time. The main purpose of /sbin/mount.aufs is, - maintain /etc/mtab - maintain the aufs pseudo-link (by calling /sbin/auplink) If you don't install /sbin/mount.aufs, then the aufs line in your /etc/mtab will be incorrect, and you may loose the aufs pseudo-link. /sbin/auplink is called by /sbin/umount.aufs too. If you don't know what the aufs pseudo-link is, then refer to the aufs manual in detail. Additional information, (from linux/Documentation/filesystems/aufs/design/02struct.txt) Pseudo-link -- Assume fileA exists on the lower readonly branch only and it is hardlinked to fileB on the branch. When you write something to fileA, aufs copies-up it to the upper writable branch. Additionally aufs creates a hardlink under the Pseudo-link Directory of the writable branch. The inode of a pseudo-link is kept in aufs super_block as a simple list. If fileB is read after unlinking fileA, aufs returns filedata from the pseudo-link instead of the lower readonly branch. Because the pseudo-link is based upon the inode, to keep the inode number by xino (see above) is important. All the hardlinks under the Pseudo-link Directory of the writable branch should be restored in a proper location later. Aufs provides a utility to do this. The userspace helpers executed at remounting and unmounting aufs by default. -- And http://www.mail-archive.com/aufs-users@lists.sourceforge.net/msg02639.html If you don't care about /etc/mtab and the pseudo-link, you can uninstall aufs2-util. Otherwise, just install it. aufs2-util contains other useful things. o /usr/bin/auchk Similar to generic fsck. Checks whether a branch is healthy or not from aufs's point of view. o /usr/bin/aubrsync Move files from the upper writable branch to the lower branch. If you use this script with aufs1, then you need to install aufs.shlib to /usr/lib/aufs.shlib. Currently only the 20080211 version is tested for aufs1. The development of this script is sponcered by ASUSTek Computer Inc. (http://www.asus.com/). Kindly they agreed that I keep my aufs work as free software as it has been. o /usr/lib/libau.so A dynamic link library for readdir(3) in user-space. If you need this , then run make libau.so install_ulib. And refer to the aufs manual in detail. Generally readdir(3) in libau.so runs faster than readdir in aufs kernel-space, particulary for the huge dir. J. R. Okajima -- This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
Re: Are aufs 'utils' mandatory ?
On Wed, Feb 11, 2009 at 11:13 AM, sf...@users.sourceforge.net wrote: [...] Or try aufs2 which include aufs utilities written in C language so that /usr is unnecessary. Junjiro, The aufs2 codebase contains kernel patches for only 2.6.27 kernel version, however the RHEL 5.2 kernel version is 2.6.18-xyz; These patches dont work on RHEL kernel . Do we have patches for earlier kernel versions ( something like the unionfs releases) for aufs2 ? Thanks, Nitin. -- Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM) software. With Adobe AIR, Ajax developers can use existing skills and code to build responsive, highly engaging applications that combine the power of local resources and data with the reach of the web. Download the Adobe AIR SDK and Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com
Re: Are aufs 'utils' mandatory ?
Nitin Bhardwaj: Do we have patches for earlier kernel versions ( something like the unionfs releases) for aufs2 ? For older kernels, I'd suggest you aufs1. (or change your mount/unmount order, or change the shutdown script as I suggested in the aufs manual). J. R. Okajima -- Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM) software. With Adobe AIR, Ajax developers can use existing skills and code to build responsive, highly engaging applications that combine the power of local resources and data with the reach of the web. Download the Adobe AIR SDK and Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com
