Eric found something like this and has a fix with in the email.
Sub: "I think "btrfs: fix leak of path in btrfs_find_item" broke stable
trees ..."
Anand
On 03/24/2015 06:40 PM, Rich Freeman wrote:
On Tue, Mar 24, 2015 at 2:31 AM, Anand Jain wrote:
Do you have this fix ..
[PATCH] Btrfs:
+bool debugfs_abort_transaction(struct btrfs_fs_info *fs_info)
+{
+ if (!btrfs_debugfs_label_trans_abort[0])
+ return false;
+ return strcmp(fs_info->super_copy->label,
+ btrfs_debugfs_label_trans_abort) == 0;
+}
+
label is not mandatory to be pr
Omar Sandoval writes:
> On Mon, Mar 30, 2015 at 02:30:34PM +0200, David Sterba wrote:
>> On Mon, Mar 30, 2015 at 02:02:17AM -0700, Omar Sandoval wrote:
>> > Before commit bafc9b754f75 ("vfs: More precise tests in d_invalidate"),
>> > d_invalidate() could return -EBUSY when a dentry for a director
Hi!
Use this at work. Releasing to list to prompt design of output that is
100% df format compatible for automated reporting and graphing.
This is so BTRFS space statistics acn bre reported back to existing
graphing tool chains that many of you would have in place. Quite useful
for munin, hint,
On Tue, Mar 31, 2015 at 7:45 PM, Zach Brown wrote:
We've known for eons that it's not great that readdir holds tree locks
while calling mkwrite by way of dir_emit(). I fiddled around and
found
a reliable if goofy deadlock reproducer. It typically takes a few
seconds on modest hardware here (
On Tue, Mar 31, 2015 at 5:54 PM, Martin Langhoff
wrote:
> On Tue, Mar 31, 2015 at 4:09 PM, Chris Murphy wrote:
>> There should be a reference to an rdsosreport.txt in /run/... so find
>> a way to get that posted somewhere.
>
> I'll try, but it truly says nothing of interest from a block device /
On Tue, Mar 31, 2015 at 4:11 PM, Chris Murphy wrote:
> While you're at it, try to mount the Btrfs volume in question normally
> and report kernel messages. If mount fails, try it with -o recovery
> mount option, and also report kernel messages and whether that fails.
Oh, I should have mentioned t
On Tue, Mar 31, 2015 at 4:09 PM, Chris Murphy wrote:
> There should be a reference to an rdsosreport.txt in /run/... so find
> a way to get that posted somewhere.
I'll try, but it truly says nothing of interest from a block device /
btrfs PoV. I have ample background debugging boot issues, disk
c
We've known for eons that it's not great that readdir holds tree locks
while calling mkwrite by way of dir_emit(). I fiddled around and found
a reliable if goofy deadlock reproducer. It typically takes a few
seconds on modest hardware here (single package, dual core/ht, single
spindle.)
I made a
Chris Murphy schrieb:
> On Tue, Mar 31, 2015 at 4:15 PM, Chris Murphy
> wrote:
>
>> The i915 regression right now is really annoying. With a Samsung 840
>> EVO I've had inexplicable and non-deterministic boot failures.
>
> Clarification: the boot failures happen following the i915 panic and
>
At the moment we can not reliably and deterministically test that the
transaction abortion code works as expected. For example in the past [1]
we had an issue where that code returned the pinned extents to the free
space caches allowing fstrim to perform a discard against the physical
locations of
On Tue, Mar 31, 2015 at 3:24 PM, Benjamin Hodgetts wrote:
> G. Richard Bellamy pteradigm.com> writes:
>
>>
>> When I upgrade to the 3.19.2 Kernel I get a deadlocked boot:
>> INFO: task mount:302 blocked for more than 120 seconds.
>> INFO: task btrfs-transacti:329 blocked for more than 120 seconds
On Tue, Mar 31, 2015 at 4:15 PM, Chris Murphy wrote:
> The i915 regression right now is really annoying. With a Samsung 840
> EVO I've had inexplicable and non-deterministic boot failures.
Clarification: the boot failures happen following the i915 panic and
subsequent forced power off.
--
Ch
On Tue, Mar 31, 2015 at 3:45 PM, Kai Krakow wrote:
> I had this happen, too, lately. It's quite often happening after an unclean
> shutdown (which currently quite often happend to me due to the xorg intel
> driver having GPU freezes). SysRq+W shows that the mount process is locked
> somewhere in
On Tue, Mar 31, 2015 at 2:39 PM, Matt Grant
wrote:
> Seen this before at home.
>
> You have to mount -o recovery off a 3.19 kernel to fix it...
>
> If you can get the SSD out, attach it to a desktop, as there will be no
> Install CDs using 3.19 yet.
Fedora 22 Workstation alpha has 4.0.0 (rc1 I t
G. Richard Bellamy pteradigm.com> writes:
>
> When I upgrade to the 3.19.2 Kernel I get a deadlocked boot:
> INFO: task mount:302 blocked for more than 120 seconds.
> INFO: task btrfs-transacti:329 blocked for more than 120 seconds.
>
> I have an LTS Kernel at 3.14.35 that also fails to boot wi
Kai Krakow schrieb:
> Chris Murphy schrieb:
>
>> On Tue, Mar 31, 2015 at 2:09 PM, Chris Murphy
>> wrote:
>>
>>> Well you have to rule that out before anyone on this list can really
>>> help. Try booting Fedora 21 install media, and using smartctl -x on
>>> the drive.
>>
>> While you're at it
Chris Murphy schrieb:
> On Tue, Mar 31, 2015 at 2:09 PM, Chris Murphy
> wrote:
>
>> Well you have to rule that out before anyone on this list can really
>> help. Try booting Fedora 21 install media, and using smartctl -x on
>> the drive.
>
> While you're at it, try to mount the Btrfs volume in
On Tue, Mar 31, 2015 at 2:09 PM, Chris Murphy wrote:
> Well you have to rule that out before anyone on this list can really
> help. Try booting Fedora 21 install media, and using smartctl -x on
> the drive.
While you're at it, try to mount the Btrfs volume in question normally
and report kernel
On Tue, Mar 31, 2015 at 12:55 PM, Martin Langhoff
wrote:
> Hi BTRFS folks,
>
> one of my dev boxes is a Thinkpad x220, with a single hybrid
> (HDD+Flash) disk, running F21 with BTRFS partitions for /home and / .
>
> After losing power (ran out of battery, possibly while trying to
> hibernate) -- t
At the moment we can not reliably and deterministically test that the
transaction abortion code works as expected. For example in the past [1]
we had an issue where that code returned the pinned extents to the free
space caches allowing fstrim to perform a discard against the physical
locations of
On 03/31/2015 08:12 PM, Chris Mason wrote:
On Mon, Mar 30, 2015 at 1:42 PM, Torbjørn wrote:
Hi,
Just a follow up on this report.
The file system in question is a raid1 across 2x320G old Western
Digital WD3200KS.
I janked them out of the server to run a fsck on another computer
(after a pr
Hi BTRFS folks,
one of my dev boxes is a Thinkpad x220, with a single hybrid
(HDD+Flash) disk, running F21 with BTRFS partitions for /home and / .
After losing power (ran out of battery, possibly while trying to
hibernate) -- the system will not boot. The initrd breaks out to a
shell where I find
On Mon, Mar 30, 2015 at 1:42 PM, Torbjørn wrote:
Hi,
Just a follow up on this report.
The file system in question is a raid1 across 2x320G old Western
Digital WD3200KS.
I janked them out of the server to run a fsck on another computer
(after a proper shutdown).
One of the disks did not g
On 31/03/15 18:37, David Sterba wrote:
> On Mon, Mar 30, 2015 at 09:59:05PM +0100, Martin wrote:
>> btrfs device delete /dev/sdf5 /mnt/data2
>>
>> ERROR: error removing the device '/dev/sdf5' - Inappropriate ioctl for
>> device
>
> Strange, that would mean that the ioctl number is not handled at a
On Mon, Mar 30, 2015 at 09:59:05PM +0100, Martin wrote:
> btrfs device delete /dev/sdf5 /mnt/data2
>
> ERROR: error removing the device '/dev/sdf5' - Inappropriate ioctl for
> device
Strange, that would mean that the ioctl number is not handled at all. I
haven't found any change in device delete
On Tue, Mar 31, 2015 at 12:04:58AM +0100, Filipe Manana wrote:
> Test cloning a file range with a length of zero into a destination offset
> greater than zero.
>
> This made btrfs create an extent state record with a start offset greater than
> the end offset, resulting in chaos such as an infinit
On Mon, Mar 30, 2015 at 05:09:52PM +0200, Marc Cousin wrote:
> > So it would be good to sample the active threads and see where it's
> > spending the time. It could be the somewhere in the rb-tree representing
> > extents, but that's a guess.
> >
> I just need to be told how to do that :)
>
> Som
Quoting André-Sebastian Liebe :
Hi Dave,
you should forward your question directly to kernel.org webmaster
. They are using mediawiki in a multisite
environment for all kernel.org related wikis.
regards,
André
Thanks!
Dave
On 03/31/2015 04:00 AM, Dave Stevens wrote:
I've been reading th
Test btrfs file range cloning with the same file as a source and
destination.
This tests a specific scenario where the extent layout of the file
confused the clone ioctl implementation making it return -EEXIST to
userspace. This issue was fixed by the following linux kernel patch:
Btrfs: fix r
While searching for extents to clone we might find one where we only use
a part of it coming from its tail. If our destination inode is the same
the source inode, we end up removing the tail part of the extent item and
insert after a new one that point to the same extent with an adjusted
key file o
On Mon, Mar 30, 2015 at 10:21:11PM +0200, Fabian Frederick wrote:
> cow_file_range_inline() was called with 0 instead of
> actual definition.
Please also fix the initialization in btrfs_finish_ordered_io .
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a m
Am 31.03.2015 14:31 schrieb Tom Van Braeckel:
Err, upon further inspection, I think that this was a false positive.
Btrfs relies on the initial value of the private_data member of a file
being NULL in the regular ioctl operation handler for
BTRFS_IOC_TRANS_START but it does not use the miscdevic
On Tue, Mar 31, 2015 at 11:35 AM, Robbie Ko wrote:
> Hi,
>
> At first, I use fsstress with following script to do random move.
> Errors could happen after several rounds of incremental send.
> Directory structure generated by fsstress is hard to find simple
> reproducible steps, so I started to ma
Err, upon further inspection, I think that this was a false positive.
Btrfs relies on the initial value of the private_data member of a file
being NULL in the regular ioctl operation handler for
BTRFS_IOC_TRANS_START but it does not use the miscdevice framework for
those files.
It *does* use the
Hi,
At first, I use fsstress with following script to do random move.
Errors could happen after several rounds of incremental send.
Directory structure generated by fsstress is hard to find simple
reproducible steps, so I started to make a small directory structure
which has several files/director
On Tue, Mar 31, 2015 at 10:39 AM, Robbie Ko wrote:
> HI,
>
> After taking a look at design of send/receive, I feel that severe
> directory structure movement combined with name collision might be a
> trouble maker for send to calculate the snapshot difference.
Well the problem is not calculating
HI,
After taking a look at design of send/receive, I feel that severe
directory structure movement combined with name collision might be a
trouble maker for send to calculate the snapshot difference.
Therefore, I started using fsstress + random directory movement to
test. Since the diff commands m
Hi Dave,
you should forward your question directly to kernel.org webmaster
. They are using mediawiki in a multisite
environment for all kernel.org related wikis.
regards,
André
On 03/31/2015 04:00 AM, Dave Stevens wrote:
> I've been reading the wiki to help sort out an issue and learn more
> ab
On Tue, Mar 31, 2015 at 4:06 AM, Robbie Ko wrote:
> Hi,
>
> I have testing btrfs send/receive recently.
>
> I got an error "send ioctl failed with -12: Cannot allocate memory" on
> send side.
> WARN_ON happened on "len > PATH_MAX" in fs_path_ensure_buf.
>
> I got an error "utime failed: No such f
On 2015/03/27 19:21, Zhaolei wrote:
> From: Zhao Lei
>
> This is another fix of no_space case.
>
> All patchs for fix no_space bug are available at fix_no_space
> branch on:
>git://github.com/zhaoleidd/btrfs
I tested in the environment that applied the following patch to
'fix_no_space' bran
On 30/03/2015 22:21, Chris Mason wrote:
On Mon, Mar 30, 2015 at 10:05 AM, Sophie Dexter
wrote:
On 24/03/15 17:34, Chris Mason wrote:
You have great timing, there are two reports of a very similar abort
with 4.0-rc5, but your report makes it clear these are not a regression
from 4.0-rc4.
Are
42 matches
Mail list logo
