Re: Fwd: Current State of BTRFS
Hi Florian, On 10/02/13 00:35, Florian Hofmann wrote: > (Sadly) No SSD. I think Marc was asking as they can cause issues, so don't be sad. :-) > I just upgraded from kernel version 3.7.5 to 3.7.6 (running Arch > Linux) with no change in behavior. There are rarely any updates to btrfs in stable kernel releases, if you want to try the current state of the code you'll need to try 3.8-rc7. cheers, Chris -- Chris Samuel : http://www.csamuel.org/ : Melbourne, VIC -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
@Leonidas: I don't think so ... this is running on a i7 core and lzo
is pretty damn fast.
2013/2/11 Leonidas Spyropoulos :
> On Mon, Feb 11, 2013 at 4:05 PM, Josef Bacik wrote:
>> On Fri, Feb 08, 2013 at 04:12:17PM -0700, Florian Hofmann wrote:
>>> I ran it as root. The first times there was no output whatsoever. This
>>> time triggering gave the SysRq : Show Blocked State line. Strange
>>>
>>
>> Well that's weird, it looks like you don't have any blocked tasks, so either
>> sysrq+w is screwing up or you are getting stuck in something CPU intensive.
>> Try
>> doing sysrq+w next time it happens again and also run top and see if
>> something
>> is using up 100% of the CPU. If it's something chewing up CPU then I'll tell
>> you how to figure out what's going on. Thanks,
>>
> I noticed you have the FS mounted with compress flag (compress=lzo).
> Could it be that your CPU is bottle-necking the process?
>
>> Josef
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
>> the body of a message to majord...@vger.kernel.org
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
>
>
> --
> Caution: breathing may be hazardous to your health.
>
> #include
> int main(){printf("%s","\x4c\x65\x6f\x6e\x69\x64\x61\x73");}
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
On Mon, Feb 11, 2013 at 4:05 PM, Josef Bacik wrote:
> On Fri, Feb 08, 2013 at 04:12:17PM -0700, Florian Hofmann wrote:
>> I ran it as root. The first times there was no output whatsoever. This
>> time triggering gave the SysRq : Show Blocked State line. Strange
>>
>
> Well that's weird, it looks like you don't have any blocked tasks, so either
> sysrq+w is screwing up or you are getting stuck in something CPU intensive.
> Try
> doing sysrq+w next time it happens again and also run top and see if something
> is using up 100% of the CPU. If it's something chewing up CPU then I'll tell
> you how to figure out what's going on. Thanks,
>
I noticed you have the FS mounted with compress flag (compress=lzo).
Could it be that your CPU is bottle-necking the process?
> Josef
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
Caution: breathing may be hazardous to your health.
#include
int main(){printf("%s","\x4c\x65\x6f\x6e\x69\x64\x61\x73");}
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
On Fri, Feb 08, 2013 at 04:12:17PM -0700, Florian Hofmann wrote: > I ran it as root. The first times there was no output whatsoever. This > time triggering gave the SysRq : Show Blocked State line. Strange > Well that's weird, it looks like you don't have any blocked tasks, so either sysrq+w is screwing up or you are getting stuck in something CPU intensive. Try doing sysrq+w next time it happens again and also run top and see if something is using up 100% of the CPU. If it's something chewing up CPU then I'll tell you how to figure out what's going on. Thanks, Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
(Sadly) No SSD. I just upgraded from kernel version 3.7.5 to 3.7.6 (running Arch Linux) with no change in behavior. 2013/2/9 Marc MERLIN : > On Fri, Feb 08, 2013 at 10:30:04PM +0100, Florian Hofmann wrote: >> Hi everybody, >> >> I am using btrfs as my main fs for some time now, but I am >> experiencing severe performance drawbacks. I can't qualify the >> circumstances, but sometimes during disc access the whole system >> freezes for some time. >> >> Maybe somebody could suggest some general things I could try to search >> for the problem? > > Are you running on an SSD? > > If so, that is likely your problem. > > Depending on your SSD: > - it's going bad > - it's running out of blocks to reallocate, or being very slow > - other firmware issue > > Marc > -- > "A mouse is a device used to point at the xterm you want to type in" - A.S.R. > Microsoft is to operating systems > what McDonalds is to gourmet > cooking > Home page: http://marc.merlins.org/ 2013/2/9 Marc MERLIN : > On Fri, Feb 08, 2013 at 10:30:04PM +0100, Florian Hofmann wrote: >> Hi everybody, >> >> I am using btrfs as my main fs for some time now, but I am >> experiencing severe performance drawbacks. I can't qualify the >> circumstances, but sometimes during disc access the whole system >> freezes for some time. >> >> Maybe somebody could suggest some general things I could try to search >> for the problem? > > Are you running on an SSD? > > If so, that is likely your problem. > > Depending on your SSD: > - it's going bad > - it's running out of blocks to reallocate, or being very slow > - other firmware issue > > Marc > -- > "A mouse is a device used to point at the xterm you want to type in" - A.S.R. > Microsoft is to operating systems > what McDonalds is to gourmet > cooking > Home page: http://marc.merlins.org/ -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
On Fri, Feb 08, 2013 at 10:30:04PM +0100, Florian Hofmann wrote: > Hi everybody, > > I am using btrfs as my main fs for some time now, but I am > experiencing severe performance drawbacks. I can't qualify the > circumstances, but sometimes during disc access the whole system > freezes for some time. > > Maybe somebody could suggest some general things I could try to search > for the problem? Are you running on an SSD? If so, that is likely your problem. Depending on your SSD: - it's going bad - it's running out of blocks to reallocate, or being very slow - other firmware issue Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
I ran it as root. The first times there was no output whatsoever. This time triggering gave the SysRq : Show Blocked State line. Strange 2013/2/9 cwillu : > On Fri, Feb 8, 2013 at 4:56 PM, Florian Hofmann > wrote: >> Oh ... I should have mentioned that btrfs is running on top of LUKS. >> >> 2013/2/8 Florian Hofmann : >>> $ btrfs fi df / >>> Data: total=165.00GB, used=164.19GB >>> System, DUP: total=32.00MB, used=28.00KB >>> System: total=4.00MB, used=0.00 >>> Metadata, DUP: total=2.00GB, used=1.40GB >>> >>> $ btrfs fi show >>> failed to read /dev/sr0 >>> Label: none uuid: b4ec0b14-2a42-47e3-a0cd-1257e789ed25 >>> Total devices 1 FS bytes used 165.59GB >>> devid1 size 600.35GB used 169.07GB path /dev/dm-0 >>> >>> Btrfs Btrfs v0.19 >>> >>> --- >>> >>> I just noticed that I can force 'it' by transferring a large file from >>> my NAS. I did the sysrq-trigger thing, but there is no suspicious >>> output in dmesg (http://pastebin.com/swrCdC3U). >>> >>> Anything else? > > The pastebin didn't include any output from sysrq-w; even if there's > nothing to report there would still be a dozen lines or so per cpu; at > the absolute minimum there should be a line for each time you ran it: > > [4477369.680307] SysRq : Show Blocked State > > Note that you need to echo as root, or use the keyboard combo > alt-sysrq-w to trigger. -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
I just had another big freeze ... I could change to a terminal and sysrq-triggering gave this: http://pastebin.com/F1jQqtUQ 2013/2/8 Florian Hofmann : > Oh ... I should have mentioned that btrfs is running on top of LUKS. > > 2013/2/8 Florian Hofmann : >> $ btrfs fi df / >> Data: total=165.00GB, used=164.19GB >> System, DUP: total=32.00MB, used=28.00KB >> System: total=4.00MB, used=0.00 >> Metadata, DUP: total=2.00GB, used=1.40GB >> >> $ btrfs fi show >> failed to read /dev/sr0 >> Label: none uuid: b4ec0b14-2a42-47e3-a0cd-1257e789ed25 >> Total devices 1 FS bytes used 165.59GB >> devid1 size 600.35GB used 169.07GB path /dev/dm-0 >> >> Btrfs Btrfs v0.19 >> >> --- >> >> I just noticed that I can force 'it' by transferring a large file from >> my NAS. I did the sysrq-trigger thing, but there is no suspicious >> output in dmesg (http://pastebin.com/swrCdC3U). >> >> Anything else? >> >> 2013/2/8 Josef Bacik : >>> On Fri, Feb 08, 2013 at 02:30:04PM -0700, Florian Hofmann wrote: Hi everybody, I am using btrfs as my main fs for some time now, but I am experiencing severe performance drawbacks. I can't qualify the circumstances, but sometimes during disc access the whole system freezes for some time. Maybe somebody could suggest some general things I could try to search for the problem? >>> >>> Can I see >>> >>> btrfs fi df /mnt/point >>> btrfs fi show >>> >>> and then when you are having problems >>> >>> echo w > /proc/sysrq-trigger >>> >>> and capture dmesg. Thanks, >>> >>> Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
On Fri, Feb 8, 2013 at 4:56 PM, Florian Hofmann wrote: > Oh ... I should have mentioned that btrfs is running on top of LUKS. > > 2013/2/8 Florian Hofmann : >> $ btrfs fi df / >> Data: total=165.00GB, used=164.19GB >> System, DUP: total=32.00MB, used=28.00KB >> System: total=4.00MB, used=0.00 >> Metadata, DUP: total=2.00GB, used=1.40GB >> >> $ btrfs fi show >> failed to read /dev/sr0 >> Label: none uuid: b4ec0b14-2a42-47e3-a0cd-1257e789ed25 >> Total devices 1 FS bytes used 165.59GB >> devid1 size 600.35GB used 169.07GB path /dev/dm-0 >> >> Btrfs Btrfs v0.19 >> >> --- >> >> I just noticed that I can force 'it' by transferring a large file from >> my NAS. I did the sysrq-trigger thing, but there is no suspicious >> output in dmesg (http://pastebin.com/swrCdC3U). >> >> Anything else? The pastebin didn't include any output from sysrq-w; even if there's nothing to report there would still be a dozen lines or so per cpu; at the absolute minimum there should be a line for each time you ran it: [4477369.680307] SysRq : Show Blocked State Note that you need to echo as root, or use the keyboard combo alt-sysrq-w to trigger. -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
Oh ... I should have mentioned that btrfs is running on top of LUKS. 2013/2/8 Florian Hofmann : > $ btrfs fi df / > Data: total=165.00GB, used=164.19GB > System, DUP: total=32.00MB, used=28.00KB > System: total=4.00MB, used=0.00 > Metadata, DUP: total=2.00GB, used=1.40GB > > $ btrfs fi show > failed to read /dev/sr0 > Label: none uuid: b4ec0b14-2a42-47e3-a0cd-1257e789ed25 > Total devices 1 FS bytes used 165.59GB > devid1 size 600.35GB used 169.07GB path /dev/dm-0 > > Btrfs Btrfs v0.19 > > --- > > I just noticed that I can force 'it' by transferring a large file from > my NAS. I did the sysrq-trigger thing, but there is no suspicious > output in dmesg (http://pastebin.com/swrCdC3U). > > Anything else? > > 2013/2/8 Josef Bacik : >> On Fri, Feb 08, 2013 at 02:30:04PM -0700, Florian Hofmann wrote: >>> Hi everybody, >>> >>> I am using btrfs as my main fs for some time now, but I am >>> experiencing severe performance drawbacks. I can't qualify the >>> circumstances, but sometimes during disc access the whole system >>> freezes for some time. >>> >>> Maybe somebody could suggest some general things I could try to search >>> for the problem? >> >> Can I see >> >> btrfs fi df /mnt/point >> btrfs fi show >> >> and then when you are having problems >> >> echo w > /proc/sysrq-trigger >> >> and capture dmesg. Thanks, >> >> Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
$ btrfs fi df / Data: total=165.00GB, used=164.19GB System, DUP: total=32.00MB, used=28.00KB System: total=4.00MB, used=0.00 Metadata, DUP: total=2.00GB, used=1.40GB $ btrfs fi show failed to read /dev/sr0 Label: none uuid: b4ec0b14-2a42-47e3-a0cd-1257e789ed25 Total devices 1 FS bytes used 165.59GB devid1 size 600.35GB used 169.07GB path /dev/dm-0 Btrfs Btrfs v0.19 --- I just noticed that I can force 'it' by transferring a large file from my NAS. I did the sysrq-trigger thing, but there is no suspicious output in dmesg (http://pastebin.com/swrCdC3U). Anything else? 2013/2/8 Josef Bacik : > On Fri, Feb 08, 2013 at 02:30:04PM -0700, Florian Hofmann wrote: >> Hi everybody, >> >> I am using btrfs as my main fs for some time now, but I am >> experiencing severe performance drawbacks. I can't qualify the >> circumstances, but sometimes during disc access the whole system >> freezes for some time. >> >> Maybe somebody could suggest some general things I could try to search >> for the problem? > > Can I see > > btrfs fi df /mnt/point > btrfs fi show > > and then when you are having problems > > echo w > /proc/sysrq-trigger > > and capture dmesg. Thanks, > > Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Fwd: Current State of BTRFS
On Fri, Feb 08, 2013 at 02:30:04PM -0700, Florian Hofmann wrote: > Hi everybody, > > I am using btrfs as my main fs for some time now, but I am > experiencing severe performance drawbacks. I can't qualify the > circumstances, but sometimes during disc access the whole system > freezes for some time. > > Maybe somebody could suggest some general things I could try to search > for the problem? Can I see btrfs fi df /mnt/point btrfs fi show and then when you are having problems echo w > /proc/sysrq-trigger and capture dmesg. Thanks, Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
