From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hans Reiser Subject: Re: reiser4 sending process to D state Date: Tue, 16 Aug 2005 04:50:09 -0700 Message-ID: <4301D2F1.1030907@namesys.com> References: <43012902.1010600@webalive.biz> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: list-help: list-unsubscribe: list-post: Errors-To: flx@namesys.com In-Reply-To: <43012902.1010600@webalive.biz> List-Id: Content-Type: text/plain; charset="us-ascii" To: mohammad.rahman@webalive.biz Cc: reiserfs-list@namesys.com, Nate Diller Nate also reported this, yes?.... Hans Mohammad A. Rahman wrote: > Hello, > > I am having an issue with reiser4. I am wondering if it is a known > issue for reiser4 or I have missed something along the path. > > #cat /etc/redhat-release > Red Hat Enterprise Linux ES release 3 (Taroon Update 5) > > Downloaded 2.6.12 vanila kernel, applied mm2 patch (2.6.12-mm2) and > then aapplied reiser4 patch downloaded from namesys for 2.6.12-mm2 > (ftp://ftp.namesys.com/pub/reiser4-for-2.6/2.6.12-mm2/reiser4-vfs-3.patch). > Kernel compiles and boots ok. > > Compiled libaal-1.0.5 and reiser4progs-1.0.5 with ./configure, make > and make install. > > Created reiser4 fs on my disk (/usr/local/sbin/mkfs.reiser4 /dev/sda5) > > Mounted the partition (/dev/sda5 /test > reiser4 defaults 1 2) > > "dmesg | grep reiser" shows: loading reiser4 bitmap......done (181 > jiffies) > > But when I try to write big amount of data on that partition the > process hangs and goes to uninterruptible sleep (D) state. > > the PS output shows: > > root 2375 2.7 0.0 3828 692 pts/1 D 23:21 0:15 cp -i > -r /usr . > > I can easily simulate it with bonnie++ > > bonnie++ -u user:user -d /test -s 4096m -m hudson -n 32:100000:16:256 > -b -q > > which also jumps into uninterruptible sleep (D) state. > > After having this issue, I tried using > libaal-1.0.4/reiser4progs-1.0.4. Tried 2.6.11 kernel, 2.6.13-rc4 - but > all with no joy. > > For all these trials, I used vanilla kernel with appropriate mm patch > and the appropriate patch downloaded from namesys (on top of that). > > Once the process goes to state D, the log shows the following output: > > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > reiser4[pdflush(99)]: writeout (fs/reiser4/vfs_ops.c:586)[nikita-31001]: > code: -12 at fs/reiser4/wander.c:1268 > WARNING: Flush failed: -12 > > Thanks and will look for some feedback. > > MR > > > P.S: If the partition is mounted as ext3, or xfs - it is working as > expected. > > >