aufs3 and aufs4 GIT release

o bugfix - aufs mmap: bugfix, concurrent msync triggers NULL pointer dereference, reported by Xavier Chantry and fixed by Ben Hutchings. The patch was applied to all supported versions (after aufs3.14). o news - for linux-4.2-rc1, there was rather major changes around symbolic links. Now au

Re: aufs development stops for a while

Hello all, I am coming back to my aufs development slowly and step by step. I think I need some kind of rehab since I was far from linux kernel for a few months. Today I've release aufs for linux-4.2-rc1. Not linux-4.2. I've just come back to the status a few months ago. In a few weeks, I will r

aufs3 and aufs4 GIT release

o news - a new branch aufs4.2 for linux-4.2. - now aufs4.x-rcN is for linux-4.3-rcN. o misc - in linux-4.3-rcN, export a new function path_noexec(). J. R. Okajima - aufs3-linux.git aufs: tiny, simplify the return value of au_seq_path() - aufs3-st

Re: aufs3.14 ... aufs3.17 will end

sf...@users.sourceforge.net: > > Hello all, > > I am going to stop maintaining aufs3.14 ... aufs3.17 versions, and make > aufs3.18 a new base version. The reason to choose aufs3.18 is linux-3.18 > is marked as longterm on http://www.kernel.org. > > The shift will happen after aufs4.2 or aufs4.3 is

aufs3 and aufs4 GIT release

This is a less important release, just to follow the mainline linux-4.3-rc3. A new branch aufs-util.git#aufs3.18 is created since my development base is aufs3.18 now. J. R. Okajima - aufs3-linux.git none - aufs3-standalone.git none - aufs4-linux.

Re: [fix] Undeletable Dirs with Layered ro+wh Mounts

Hello Karl, Karl Semich: > - 3 layers of +wh under the rw layer on top > - Bottom layer has folder A with files inside it > - Second-to-bottom layer has a whiteout deleting folder A on bottom layer > - Second-to-top layer has folder A recreated with files inside it > - Go into rw top mount, remov

Re: [fix] Undeletable Dirs with Layered ro+wh Mounts

Hello Yokogawa, Kazuya Yokogawa: > I have similar problem, and I hope that it is the same cause. > > I created a test script that reproduces my problem. Thank you for the script. Now I think I could understand the problem. I will apply Karl Semich's fix and release on next Monday. J. R. Okajim

Re: aufs_migratepage:803...This operation is not supported

Hello B Buhlig, Your mail wasn't delivered to me, but I've just found it in the archive on SourceForge. I don't know why it was not delivered but I remember such accident has ever happened several times. B Buhlig: > For my mythtv backend, I followed the below instructions at > http://zackreed.m

aufs3 and aufs4 GIT release

o bugfix - whiteout on the deeper branch, reported by Karl Semich and Kazuya Yokogawa, fixed by Karl Semich. This is one of the most long lived bug. J. R. Okajima -- - aufs3-linux.git aufs: tiny, correct some spellin

aufs3 and aufs4 GIT release

o bugfix - possible bugfix, maintain the internal array if AUFS_MVDOWN_KUPPER - possible bugfix, stop updating the inode generaion if AUFS_MVDOWN_KUPPER - possible bugfix, dbrange_test() may fail if UDBA_NONE - testing no ->migratepage() J. R. Okajima

Re: AUFS fs unstable once top layer is full

Hello Udo, Udo Giacomozzi: > Problem: When the tmpfs fills up (eg. `cat /dev/zero > /DUMMY`), then > the system consistently becomes pratically unusable, meaning that simple > commands like `find` or `reboot` can't be run anymore ("command not found"). > > I've also noticed some kernel errors (

Re: AUFS fs unstable once top layer is full

Udo Giacomozzi: > I don't have /dev/shm in "init" (when AUFS is set up) so I simply > mounted a tmpfs just for AUFS's "xino" file and it works like a charm! > Great! Glad to hear that! > After a `find / -mount` my debugfs files show these contents: > > xib: 16x4096 8192 > xigen: 280x4096 1606

Re: AUFS fs unstable once top layer is full

Udo Giacomozzi: > ==> /sys/kernel/debug/aufs/si_e6cc0b3f/xi0 <== > 1, 1360x4096 725944 ::: > ==> /sys/kernel/debug/aufs/si_e6cc0b3f/xi5 <== > 1, 120x4096 61280 ::: > So "xigen" is using 288 * 4096 = about 1,1 Megabytes? Exactly. > Based on all the files listed above, is the tota

Re: Making aufs work on CentOS 6

Hello Saurabh, Saurabh Jain: > I am in the situation of having to make aufs work on CentOS 6 / RHEL > 6. This is unfortunate because I'm not a fan of backporting stuff. My ::: Me either. I don't like backporting. > On this, I apply two changes. One is along the lines of > https://github

aufs3 and aufs4 GIT release

o bugfix - stop NULL param for symbol_put() o news - aufs4.3 branch is created. now aufs4-x-rcN is for aufs4.4-rcN o misc - enhancement for udba=none, no ->d_revalidate() J. R. Okajima -- - aufs3-linux.git aufs: bugfix

Re: Making aufs work on CentOS 6

Saurabh Jain: > I just tried replacing __putname() with free_page() but the panic > continues (appended at the end of this email). Is your change same to this commit in aufs3? 8d602a6 2012-11-09 aufs: for linux-3.7, replace __getname_gfp() by __get_free_page() > Would you send me a diff on ho

Re: AuFS Module not found under Ubuntu MATE 15.10 w/ Raspberry Pi 2

Hello Philipp, "Philipp Trulson": >The thing is: It worked flawlessly with UM 15.04 (Kernel 3.18) on RPi 2 and >it also works fine with UM 15.10 x64 on my desktop. But UM 15.10 on the RPi >2 always says "kernel module not found" and I don't know where to begin to >look for the pro

Re: More issues with aufs2 on 2.6.32

Saurabh Jain: > Thanks for your help last time. Aufs now compiles and works on the > latest 2.6.32 kernel from CentOS 6. However, I am back to experiencing > the problem I was seeing earlier. What happens is that different > processes on the system get hung due to aufs. The functions that seem > t

Re: More issues with aufs2 on 2.6.32

Saurabh Jain: > But S_BIAS is undefined. It is not declared anywhere in the aufs2 or > the kernel tree. It's just because your kernel is not vanilla. It is modified by CentOS. If you searched plain 2.6.32, you would find this in include/linux/fs.h. #define S_BIAS (1<<30) Ideally you should trac

aufs3 and aufs4 GIT release

o bugfix - mvdown and the plink maintenance mode o misc - udba=none has no ->getattr() - aufs-util, fhsm, support for EOPNOTSUPP J. R. Okajima - aufs3-linux.git aufs: bugfix, mvdown and the plink maintenance mode aufs: tiny, remove a meaning

Re: aufs and autofs

Hello Ken, Ken Gillett: > Will the aufs mount (i.e. merge) from fstab be sufficient to cause the NFS > shares to be automounted? > > If not, will they get automounted when someone accesses the aufs 'merged' > directory? I have not tried the branch which is automounted and auto-unmounted. Current

Re: Kernel memory allocation failures involving large aufs directories (4.2-20151102)

Hello OmegaPhil, OmegaPhil: > For some time now (across kernel versions, aufs versions, machines, etc) > I have noticed kernel memory allocation failures intermittently when > something lists a large aufs directory (the most recent example having > 20,000 files in it). My aufs usage is to have mu

aufs3 and aufs4 GIT release

This release is just to follow mainline linux-4.4-rc1. There is no bugfix, no new features. J. R. Okajima -- Go from Idea to Many App Stores Faster with Intel(R) XDK Give your users amazing mobile app experiences with I

Re: Kernel memory allocation failures involving large aufs directories (4.2-20151102)

OmegaPhil: ::: > repos, so I'm using this - currently the rsync init.d script has been > edited to export the right LD_PRELOAD and LIBAU values, and I've > confirmed the library has been loaded via: > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3

Re: aufs and autofs

sf...@users.sourceforge.net: > Ken Gillett: > > Will the aufs mount (i.e. merge) from fstab be sufficient to cause the NFS > > shares to be automounted? > > > > If not, will they get automounted when someone accesses the aufs 'merged' > > directory? > > I have not tried the branch which is automou

Re: Error with lockdep

Hello Karl, Karl Semich: > I'm getting the following when I mount my root aufs filesystem: > > [ 20.360553] BUG: looking up invalid subclass: 9 ::: It is due to the constant values defined in mainline is small. Several people have ever suggested making them customizable, but it doesn't

aufs3 and aufs4 GIT release

This release contains a few minor optimization. Also this is last release from me in this year. Have nice holidays J. R. Okajima - aufs3-linux.git aufs: minor optimization, replace au_array_free() by kvfree() aufs: minor optimization, refine

Re: aufs_destroy_inode() deadlock?

Hello Akihiro, Akihiro Suda: > Many people are reporting that AUFS hangs up when a Java process exits. > https://github.com/docker/docker/issues/18180 > (Interestingly, the problem seems particular to Java) Thanks for reporting. This is new to me and ML. Here is a debug patch. Please apply, repro

Re: aufs_do_xino_fwrite() EINTR loop (WAS: aufs_destroy_inode() deadlock?)

Akihiro Suda: > aufs do_xino_fwrite:85:java[1077]: err -4 > aufs au_xino_do_write:439:java[1077]: I/O Error, write failed (-4) Thanks for testing. > I'll continue to find the source of this EINTR loop. > Maybe this is a bug of Linux kernel itself (kernel/pid_namespace.c?) rather > than AUFS,

Re: aufs_do_xino_fwrite() EINTR loop (WAS: aufs_destroy_inode() deadlock?)

> > I'll continue to find the source of this EINTR loop. > > Maybe this is a bug of Linux kernel itself (kernel/pid_namespace.c?) rather > > than AUFS, but I'm still not sure. > > Or aufs should support the case of PF_EXITING flag set in > current->flags... Here is my current and UNTESTED solution

Re: aufs_do_xino_fwrite() EINTR loop (WAS: aufs_destroy_inode() deadlock?)

Akihiro Suda: > However, the bug is a regression caused by commit 296291cd ("mm: make > sendfile(2) killable") to the upstream of Linux kernel. > https://github.com/torvalds/linux/commit/296291cd > This produces infinite -EINTR loop in mm/filemap.c:generic_perform_write(). > Perhaps it can also af

Re: aufs_do_xino_fwrite() EINTR loop (WAS: aufs_destroy_inode() deadlock?)

Akihiro Suda: > However, the bug is a regression caused by commit 296291cd ("mm: make > sendfile(2) killable") to the upstream of Linux kernel. > https://github.com/torvalds/linux/commit/296291cd > This produces infinite -EINTR loop in mm/filemap.c:generic_perform_write(). > Perhaps it can also af

aufs3 and aufs4 GIT release

o news - add a new patch into aufs[34]-standalone.git lockdep-debug.patch is not for a generic use. It is just for those who wants to debug aufs by LOCKDEP. Suggested-by: Karl Semich <0xl...@gmail.com> - aufs4.1.13+ branch is released a commit in linux-4.3 has an effect to aufs4.3. It is

try reproducing this problem?

Hello all, Have you heard about the latest overlayfs security problem? http://www.halfdog.net/Security/2015/UserNamespaceOverlayfsSetuidWriteExec/ The problem is already fixed in mainline. I am afraid that aufs might have a similar problem, paricularly when the module paramter 'allow_userns' is

Re: try reproducing this problem?

Woops! Forgot attaching. sf...@users.sourceforge.net: > > Hello all, > > Have you heard about the latest overlayfs security problem? > > http://www.halfdog.net/Security/2015/UserNamespaceOverlayfsSetuidWriteExec/ > > The problem is already fixed in mainline. > I am afraid that aufs might have a si

Re: try reproducing this problem?

KATOH Yasufumi: > (Vanilla kernel does not allow root in a user namespace to mount > overlayfs. So I guess it is only the case of ubuntu kernel.) I see. That must be the point I missed. > I also don't understand it well. But I ran the test program. > * I can reproduce by overlayfs > * I can

Re: try reproducing this problem?

Hi! James B: > Tested on 3.18.7. I'm not sure where to set allow_userns=0 or =1; but this > kernel has userns enabled. Thank you very much. allow_userns is aufs's module parameter. You can set it by # modprobe aufs allow_userns=1 or append this to your boot loader (the kernel co

Re: try reproducing this problem?

James B: > Looks okay to me as long as I don't get a root shell. > The kernel is vanilla + aufs patch only. Thank you very much for your repeated tests! J. R. Okajima -- Site24x7 APM Insight: Get Deep Visibility into A

aufs3 and aufs4 GIT release

o news - aufs4.4 linux-v4.4 is released and aufs4.4 branch is born. - aufs4.2, the end of life linux-v4.2 is marked as EOL. So aufs4.2 transits "less tested" stage. o misc - minor optimization, au_iigen() returns the flags only J. R. Okajima - aufs3

Re: Kernel memory allocation failures involving large aufs directories (4.2-20151102)

OmegaPhil: > It has now been some time since I got the kernel memory allocation > failures, so clearly the libau hack has fixed it - thanks. Glad to hear that! (Honestly speaking, I totally forgot about this issue) > In the manpage, please can you change 'If you have a directory which has > mil

Re: Kernel memory allocation failures involving large aufs directories (4.2-20151102)

Sorry for my late reply. OmegaPhil: > Suddenly the VM doesn't error anymore in the particular test I set up, > so back on the server, I fiddled with the rsync init.d script and ran > the daemon via 'strace -fv'. One EINVAL hit in the resulting file, here > is it with some context: Thanks. The tr

Re: Fwd: aufs - Text file busy error

Hello James, James Burry: > Hello everyone. I am attempting to use aufs with Ubuntu 14.04.3. When I run > the line below in an aufs filesystem, I will intermittently get an error - > "text file is busy" > > chmod +x script.sh && ./script.sh I didn't know such problem. Your kernel, Ubuntu 14.04.3

Re: Kernel memory allocation failures involving large aufs directories (4.2-20151102)

OmegaPhil: > Here is the raw ioctl trace: ::: > [pid 10412] ioctl(0x6, 0xc0404100, 0x74878360) =3D -1 (errno 22) ::: Ok, the value 0xc0404100 is correct. Just to make sure, I'd ask you to check your /sys/fs/aufs/config. Does it show you 'CONFIG_AUFS_RDU=y' line? J. R. Okajim

Re: Kernel memory allocation failures involving large aufs directories (4.2-20151102)

OmegaPhil: > Well... that was it. What had happened was that I had started the libau ::: I'm glad that your problem is solved! > I've just donated, sorry for wasting your time. Oh, thank you very much. Donation will surely make the life of aufs (and me) longer. I will add your name to

Re: Ubuntu: unable to mount RW branches

Hello Lionel, Lionel Beard: > Ok I found by myself. > The issue has nothing to do with Ubuntu. > But the server needs to have write access to NFS share in order to create > .wh..wh stuff. And as mount is made by root, server needs to have root > access to NFS exports, which is not acceptable in m

Re: Ubuntu: unable to mount RW branches

Lionel Beard: > I would like to avoid giving root access, even for specific client, but I > don't know how to do... Hmm, then you may be able to make the non-root users to allow mouting, by specifying 'user' in your /etc/fstab. But I am not sure. Just an idea to try. > BTW, what are these hidde

Re: Fwd: aufs - Text file busy error

James Burry: > I tried the aufs-4.4 Linux tree from git ( > git://github.com/sfjro/aufs4-linux.git > <https://github.com/sfjro/aufs4-linux.git>) > > I still see these "text file busy" errors. The errors happen in a chroot > environment of the aufs mountp

Re: Fwd: aufs - Text file busy error

James Burry: > root@U1204-32-BS:~# cat /sys/fs/aufs/si_3532123f/br* > /proj/5.113.79.8/.scratch=rw ::: > /proj/5.113.79.8/.isoMounts/114_ThirdPartyLinuxDev.iso=ro ::: > All of the ro branches are mounted from an NFS share. Maybe the extra > latency associated with the NFS share ha

Re: Fwd: aufs - Text file busy error

James Burry: > The .rw branch (/proj/5.113.79.8/.scratch ) is > ext2. http?? Let me make sure once again. Your RW branch is local ext2 as /dev/sda1, right? J. R. Okajima -- Site24x7 APM I

Re: Fwd: aufs - Text file busy error

James Burry: > All of the ro branches are mounted from an NFS share. Maybe the extra > latency associated with the NFS share has something to do with the issue. > The file which is causing the "text file busy" error for me is coming form > /proj/5.113.79.8/.isoMounts/114_ThirdPartyLinuxDev.iso=ro,

Re: Fwd: aufs - Text file busy error

James Burry: > Here is the kernel message produced : Thanx for the call trace. Here is my current guess. Brief scenario - chmod sets "this file is being changed" mark to the file inode. - exec(./a.sh) checks the mark, and rejects as "the file should not be changed during the execution." Rather

Re: Fwd: aufs - Text file busy error

Thanks for testing. James Burry: > I applied this patch, and it appears to be working. The build software is > no longer reporting "text file busy". Do you think that this patch will > make it into the official AUFS tree ? Yes, it will. I will refine the patch and apply to every versions aufs3.1

aufs3 and aufs4 GIT release

Today is not Monday, but I release latest aufs here. I'm just very busy now, and I'm afraid I won't afford to develop aufs next weeks. aufs4.x-rcN branch in this release is for linux-4.5-rc3. o news - aufs3.18.25+ branch is created, based upon the request from Bernhard Kaindl. Linux-3.18.25

Re: AUFS does not work with `fcntl(fd, F_SETFL, O_APPEND)`

Hi, Akihiro Suda: > A recent Docker issue reports that AUFS does not work with `fcntl(fd, > F_SETFL, O_APPEND)`. > How to reproduce the situation is also attached in the issue ticket. > https://github.com/docker/docker/issues/20199 Thanx for reporting. > I found a workaround which disables `au

aufs3 and aufs4 GIT release

This release is not a regular one. It is rather emergency because I've received personally a possible security problem. Until the security problem is fixed or confirmed that it was a false-positive, aufs rejects FUSE as a branch. I hope FUSE branch is not so popular. J. R. Okajima

aufs3 and aufs4 GIT release

This is not a regular release, and it contains two security bugfixes. Now FUSE branch comes back. Also this release supports linux-v4.5-rc4. o security bugfix - test mnt_ns in open(2) for fuse branch, reported by halfdog. - copy-up resets ACL, reported by halfdog. o bugfix - skipped an unlock f

Re: aufs3 and aufs4 GIT release

Hi, Philip M ller: > please check for EXPORT_SYMBOL_GPL(fput). > Seems you have added this by mistake. It is actually still > EXPORT_SYMBOL(fput)! Did I make something wrong? Which tree, which branch? And which patch? J. R. Okajima -

Re: aufs3 and aufs4 GIT release

sf...@users.sourceforge.net: > This is not a regular release, and it contains two security bugfixes. > Now FUSE branch comes back. ::: > aufs: tiny, fix an editing failure I made another similar mistake, and many copy-up will fail with this version. I will release the fix as soon as

aufs3 and aufs4 GIT release

With this release, the security bugfix series is completed. And aufs4.x-rcN branch in this release is for v4.5-rc5. o bugfix - EOPNOTSUPP from vfs_removexattr() - permission bits under copying-up J. R. Okajima - aufs3-linux.git aufs: fix an editing

Re: Simple fsck.aufs python script

Hello Alexandre, Alexandre Levavasseur: > The result is here : https://gist.github.com/Alex131089/560a1386dd756059e21b > . > > It does the same as fsck.mergerfs for the permissions (but using another > method to find branches), and it checks for useless whiteouts (hoping I'm > not misunderstandi

Re: AUFS does not work with `fcntl(fd, F_SETFL, O_APPEND)`

Akihiro Suda: > A recent Docker issue reports that AUFS does not work with `fcntl(fd, > F_SETFL, O_APPEND)`. > How to reproduce the situation is also attached in the issue ticket. > https://github.com/docker/docker/issues/20199 Here is my current solution. But I really don't like it. Any comment (

Re: Compile error

Hello Steve, Steve Fisher: > I have tried to compile aufs against a V4 kernel and have hit an issue. ::: > fs/exec.c:100:15: error: =E2=80=98path_noexec=E2=80=99 undeclared here (not= > in a function) > > EXPORT_SYMBOL(path_noexec); > >^ I don't think aufs is related to

Re: aufs3 and aufs4 GIT release

Ben Hutchings: > My reasoning was that if a function is not already exported, it must be > intended as "an internal implementation issue, and not really an > interface" which is precisely where EXPORT_SYMBOL_GPL is appropriate >

aufs3 and aufs4 GIT release

o news - Support for fcntl(F_SETFL), reported by Akihiro Suda. In order to support fcntl(F_SETFL), aufs adds a new member into struct file_operations. But I don't think it a good approach. Simply I just don't know other good solution. o misc - replace all EXPORT_SYMBOL(), which are added by

Re: Deduplication on stat change?

Hello Karl, Karl Semich: > I was wondering if there could be some way to deduplicate in this > scenario? Such that when the ctime needs to be updated, just some > small metafile is copied up, and not all the contents which could be > many gigabytes? Many years ago, I had thought about somethink

Re: Issue with aufs, boot2docker, Percona and O_DIRECT

Hell Christie, Christie Wilson: > We have run into an issue running Percona with O_DIRECT when using > boot2docker (which uses aufs). We have been able to work around this > problem by forcing docker not to use aufs in this case, but thought you > might be interested in case there is an underlyin

Re: Deduplication on stat change?

Karl Semich: > Would it be reasonable for aufs to refrain from copying up if only the > ctime is changed? Perhaps a mount option to ignore isolated ctime > changes? Hmm, I think I can understand your point. But it won't be so easy to decide because it is against a generic filesystem behaviour.

Re: Issue with aufs, boot2docker, Percona and O_DIRECT

sf...@users.sourceforge.net: > While I do know nothing about Percona and boot2docker, I will try > looking at them. > I am not using docker either. I am afraid I have to build another new > test environment which may take some time. Unfortunately it becomes clear that I cannot (almost impossible)

aufs3 and aufs4 GIT release

o news - new aufs4.5 branch is created o misc - tiny, convert a few pr_warn() to ..._once() - aufs manual, describe git work-tree and aufs J. R. Okajima - aufs3-linux.git aufs: tiny, convert a few pr_warn() to ..._once() - aufs3-standalone.git d

aufs3 and aufs4 GIT release

o news - linux-4.3 becomes EOL, so do aufs4.3. o misc - minor, convert radix_tree for si_pid to bitmap array with hoping a better performance under some unusual environment. J. R. Okajima - aufs3-linux.git aufs: minor, convert radix_tree for si_

Re: Regarding problem with the building aufs-util.

Hello Atul Singh, Atul Singh Mandla: > Is it possible to compile the aufs-util without using "make headers > install"? I am afraid not. > In this case I got a different error and I am not able to get to the root > cause of the same. ::: > error: expected =E2=80=98:=E2=80=99, =E2=80=98,

aufs3 and aufs4 GIT release

The branch aufs4.x-rcN in this release is for linux-4.6-rc6. o minor update - minor optimization, embed plink list - tiny optimization, convert si_list to hlist - lockdep-debug.patch, reduce MAX_LOCKDEP_KEYS_BITS o misc - update the donators J. R. Okajima -

Re: system boot hangs with the latest systemd

Hello Santiago, Santiago Gimeno: > Sorry for resurrecting this old thread, but I'm experiencing this very same > issue. Whao, it is back in Nov 2012... Incredebly old thread! > Also, regarding the disable of CONFIG_FHANDLE, how would I do that? Should I > set CONFIG_FHANDLE=n and recompile the

Re: system boot hangs with the latest systemd

Santiago Gimeno: > I'm still having issues with this. Apparently disabling CONFIG_FHANDLE > makes > CONFIG_EXPORTFS=m instead of CONFIG_EXPORTFS=y and I imagine that is > causing the error. > Am I doing something wrong? How are you configuring? If you set CONFIG_AUFS=y and CONFIG_EXPORTFS=m, then

Re: system boot hangs with the latest systemd

Santiago Gimeno: > I'm running make menuconfig. If I disable CONFIG_FHANDLE in > > General setup ---> open by fhandle syscalls > > The .config file generated contains CONFIG_EXPORTFS=m, And how did you get the aufs source files? If it is aufs[34]-linux.git instead of aufs[34]-standalone.git, the

Re: system boot hangs with the latest systemd

> Santiago Gimeno: > > I'm running make menuconfig. If I disable CONFIG_FHANDLE in > > > > General setup ---> open by fhandle syscalls > > > > The .config file generated contains CONFIG_EXPORTFS=m, > > And how did you get the aufs source files? > If it is aufs[34]-linux.git instead of aufs[34]-st

Re: system boot hangs with the latest systemd

Santiago Gimeno: > I've tried two ways: > > 1) Getting the 3.16.7 vanilla kernel from kernel.org and patching it with > aufs3-standalone.git Which branch? > 2) Getting the Debian Jessie kernel (3.16.7-ckt25-2) that already contains > aufs. Unfortunately I am not a debian kernel maintainer. J

Re: system boot hangs with the latest systemd

Santiago Gimeno: > git clone --depth 1 --single-branch --branch v3.16.7 git:// > git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git I am asking which branch in aufs3-standalone.git. And if you set CONFIG_AUFS=y and CONFIG_EXPORTFS=m, then you cannot set CONFIG_AUFS_EXPORT=y. But if

Re: system boot hangs with the latest systemd

> I am asking which branch in aufs3-standalone.git. > And if you set CONFIG_AUFS=y and CONFIG_EXPORTFS=m, then you cannot set > CONFIG_AUFS_EXPORT=y. > But if you set CONFIG_AUFS=m, then you can set CONFIG_EXPORTFS=m and > CONFIG_AUFS_EXPORT=y. Also, which aufs patch did you apply? J. R. Okajim

aufs3 and aufs4 GIT release

o bugfix - aufs mmap: bugfix, mainly for linux-4.5-rc5, remap_file_pages(2) emulation, reported by Colin Ian King. J. R. Okajima - aufs3-linux.git none - aufs3-standalone.git none - aufs4-linux.git aufs mmap: bugfix, mainly for linux-4.5-rc5

Re: system boot hangs with the latest systemd

Santiago Gimeno: > 2) This configuration that fails to build is reproducible for me using a > kernel 3.16.7 with aufs3.16 > > # CONFIG_FHANDLE is not set > CONFIG_EXPORTFS=m > CONFIG_AUFS_EXPORT=y Now I am testing next Monday aufs release which contains this commit commit b180f236a10e6fab2daa19d

Re: Module-only aufs

Hi Tomas, I hope you are doing well. Tomas M: > is it possible to compile current aufs as a kernel module, which > could be used with unpatched kernel? I can understand that patching and rebuilding the whole kernel takes a long time. But unfortunately it is impossible to skip patching. I guess t

aufs3 and aufs4 GIT release, aufs4.6 has come, linux-v4.7 will kill all aufs3 and aufs4.0

o news - linux-v4.6 is out and aufs4.6 too When linux-v4.7 comes, I will stop maintaining all aufs3 and aufs4.0 versions and my development base version will be aufs4.1. If you have any objection, please let me know. o bugfix - build bugfix, unsupport CONFIG_EXPORTFS=m and CONFIG_AUFS=y e

Re: test

Dan Kegel: > I got a bounce, too. Are the bounces benign? Your test message seems > to have made it to the archives. For many years, I have got such bounces. I think it is a problem of SF Mailman's. J. R. Okajima --

Re: The dreaded stale file handle with aufs, this time with ext4 and ubuntu 16.04?

Hello Dan, Dan Kegel: > I recently started trying to mount ephemeral lxc containers on tmpfs, > and fairly often, a script inside the container launched by > lxc-start-ephemeral will fail early with > > rm: failed to get attributes of '/': Stale file handle Unfortunately I don't know much about

aufs3 and aufs4 GIT release

o bugfix - possible bugfix, internal "fsnotify mark" ref-count o misc - minor, convert si_ninodes and si_nfiles into percpu_counter - convert nw_len into percpu_counter - minor, silence lockdep in internal exportfs_decode_fh() J. R. Okajima - aufs3-linu

aufs3 and aufs4 GIT release

o news - aufs4.x-rcN branch in this release is for linux-v4.7-rc2. And it didn't pass all of my local tests yet. o bugfix - half-bad-inode o misc - for linux-v4.0, loopback block device kthread name - for linux-v4.1, read_iter() for the loopback block dev - for linux-v4.4, Revert "aufs: for li

Re: inode changed

Hello, >I use aufs as root filesyetem.but I am in trouble. my application use >contant inode to mark important file, and don't change it. but all inode >changed when the computer reboot. I don't know how to keep contant inode >when I reboot machine. So you want the persistent in

aufs3 and aufs4 GIT release

aufs4.x-rcN branch in this release is for linux-v4.7-rc3, and it doesn't pass my local tests. I am still working on it. o bugfix - debug-print before dput() - missing null-test for inode before refreshing it o misc - aufs4.x-rcN for linux-v4.7-rc1 supports for the paralle readdir (and RDU).

aufs3 and aufs4 GIT release

This release is for linux-v4.7-rc4 plus alpha. I mean e7d6ef9 2016-06-20 fix idiotic braino in d_alloc_parallel() or 67016f6 2016-06-20 Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs A bug in VFS dcache was fixed, and aufs4.x-rcN branch

Re: copy file

Hello, >I use aufs filesystem as rootfs filesystem. the kylin user is a normal >user, tips information is wrong when I copy file to super user directory. >I use some command, like this. >kylin@kylin-os:~$ touch wyb >kylin@kylin-os:~$ cp wyb /etc >cp: cannot create regular

Re: copy file

> /sys/module/aufs: > kylin@kylin-Vostro-260s:/sys/module/aufs$ ls > coresize holders initsize initstate notes parameters refcnt sections > srcversion taint uevent version ::: Since you have /sys/module/aufs, you must have /sys/fs/aufs too, don't you? I need them. > kylin@ky

Re: copy file infomation error

> The info is corrent when I use linux kernel 4.7.0-rc5 > kylin@kylin-Inspiron-580s:~$ cat /sys/module/aufs/version > 4.x-rcN-20160627 > kylin@kylin-Inspiron-580s:~$ uname -r > 4.7.0-rc5 > kylin@kylin-Inspiron-580s:~$ cp new /etc/ > cp: cannot create regular file '/etc/new': Permission denied OK

aufs3 and aufs4 GIT release

aufs4.x-rcN branch in this release is for linux-v4.7-rc5. - for linux-v4.7-rc5, remove inode lock from i_op->set_acl() J. R. Okajima - aufs3-linux.git none - aufs3-standalone.git none - aufs4-linux.git#aufs4.0..aufs4.6 branch none - aufs4-linux

Re: nfs export as aufs branch.

All these three mails (except last one) were not delivered to me. Additionally the date of the first mail from Vaibhav Khanduja is 6/27/16, but Lionel Beard quoted it as 2016-07-04. Does it mean the mail delivery on SourceForge ML server is very slow? I don't know what is going on the ML server...

Re: Hitting up against max filename length in aufs

Hello OmegaPhil, OmegaPhil: > Reading the manpage, I think this is covered with: > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

aufs3 and aufs4 GIT release

o bugfix - aufs-util.git auplink, set the upper limit to nftw's nopenfd, reported by jheissler. aufs4.x-rcN branch in this release is for linux-v4.7-rc7. J. R. Okajima - aufs3-linux.git none - aufs3-standalone.git none - aufs4-linux.git none

Re: Merge a folder split over two drives into one...

Hello John, John Hendrikx: > When I mount aufs on /mnt/vb1 with br=/mnt/b1=rw,/mnt/b2=rw and do: Between two branches, :(colon) should be specified instead of comma(,) . You can confirm the current branch list via /sys/fs/aufs/si_*/brN. J. R. Okajima -

aufs3 and aufs4 GIT release

o news - linux-v4.7 is released and aufs4.7 branch too. o bugfix - the positive value from aufs_atomic_open(). - aufs-util.git, fix error_at_line() support for musl libc by Natanael Copa. o misc - a delayed-free approach is implemented to gain a performance. - minor optimization, reduce the sp

Re: Compiling aufs-util with debugging symbols - libau segfault with OpenSSH built-in SFTP server

Hello OmegaPhil, OmegaPhil: > Program received signal SIGSEGV, Segmentation fault. > 0x7fbd72a6988e in rdu_buf_lock () from /usr/lib/libau.so > (gdb) bt > #0 0x7fbd72a6988e in rdu_buf_lock () from /usr/lib/libau.so > #1 0x7fbd72a69e81 in closedir () from /usr/lib/libau.so ::

<    1   2   3   4   5   6   7   8   9   10   >