`btrfs receive` almost coming to a halt

`btrfs receive` almost coming to a halt

[Remco Hosman - Yerf IT]

kernel: 3.9.0
btrfs-progs: pulled from git this morning

Trying to receive a 5gig send file. the first bit is fast, doing 10 - 50MB/sec.
then it slows down. cpu usage is 50% (dual core machine).
when i do a strace, it looks like this, repeating over an over, about 1 piece each second:
--
read(3, "q\0\0\0\20\0008\352\327o", 10) = 10
read(3, "\22\0\10\0\0\0$~\0\0\0\0\30\0\10\0\0\0\2\0\0\0\0\0\17\0\24\0DB2/"..., 113) = 113
open("/media/snaps/yerf-2013-05-02-03:15:01/DB2/DB2-flat.vmdk", O_RDONLY|O_NOATIME) = 6
ioctl(5, 0x4020940d, 0x7fffc6d41c60)    = 0
close(6)                                = 0
read(3, "q\0\0\0\20\0\242>\357\263", 10) = 10
read(3, "\22\0\10\0\0\0&~\0\0\0\0\30\0\10\0\0\0\2\0\0\0\0\0\17\0\24\0DB2/"..., 113) = 113
open("/media/snaps/yerf-2013-05-02-03:15:01/DB2/DB2-flat.vmdk", O_RDONLY|O_NOATIME) = 6
ioctl(5, 0x4020940d, 0x7fffc6d41c60)    = 0
close(6)                                = 0
--

it pauses for a second after "ioctl(5, 0x4020940d"
it has been running like that for 3 hours now.
the file its working is large (80gig) and filefrag reports 648862 extends.
filesystem is mounted with rw,relatime,compress-force=lzo,space_cache

anything i can do to see what the problem is?

Remco

Re: `btrfs receive` almost coming to a halt

[Arne Jansen]

On 09.05.2013 17:14, Remco Hosman - Yerf IT wrote:
> kernel: 3.9.0
> btrfs-progs: pulled from git this morning
> 
> Trying to receive a 5gig send file. the first bit is fast, doing 10 - 50MB/sec.
> then it slows down. cpu usage is 50% (dual core machine).
> when i do a strace, it looks like this, repeating over an over, about 1 piece each second:
> --
> read(3, "q\0\0\0\20\0008\352\327o", 10) = 10
> read(3, "\22\0\10\0\0\0$~\0\0\0\0\30\0\10\0\0\0\2\0\0\0\0\0\17\0\24\0DB2/"..., 113) = 113
> open("/media/snaps/yerf-2013-05-02-03:15:01/DB2/DB2-flat.vmdk", O_RDONLY|O_NOATIME) = 6
> ioctl(5, 0x4020940d, 0x7fffc6d41c60)    = 0
> close(6)                                = 0
> read(3, "q\0\0\0\20\0\242>\357\263", 10) = 10
> read(3, "\22\0\10\0\0\0&~\0\0\0\0\30\0\10\0\0\0\2\0\0\0\0\0\17\0\24\0DB2/"..., 113) = 113
> open("/media/snaps/yerf-2013-05-02-03:15:01/DB2/DB2-flat.vmdk", O_RDONLY|O_NOATIME) = 6
> ioctl(5, 0x4020940d, 0x7fffc6d41c60)    = 0
> close(6)                                = 0
> --
> 

Is this the receive side?
Where does the data come from, a local file or via network?

-Arne

> it pauses for a second after "ioctl(5, 0x4020940d"
> it has been running like that for 3 hours now.
> the file its working is large (80gig) and filefrag reports 648862 extends.
> filesystem is mounted with rw,relatime,compress-force=lzo,space_cache
> 
> anything i can do to see what the problem is?
> 
> Remco--
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: `btrfs receive` almost coming to a halt

[Remco Hosman - Yerf IT]

On May 10, 2013, at 9:27 AM, Arne Jansen <sensille@gmx.net> wrote:

> On 09.05.2013 17:14, Remco Hosman - Yerf IT wrote:
>> kernel: 3.9.0
>> btrfs-progs: pulled from git this morning
>> 
>> Trying to receive a 5gig send file. the first bit is fast, doing 10 - 50MB/sec.
>> then it slows down. cpu usage is 50% (dual core machine).
>> when i do a strace, it looks like this, repeating over an over, about 1 piece each second:
>> --
>> read(3, "q\0\0\0\20\0008\352\327o", 10) = 10
>> read(3, "\22\0\10\0\0\0$~\0\0\0\0\30\0\10\0\0\0\2\0\0\0\0\0\17\0\24\0DB2/"..., 113) = 113
>> open("/media/snaps/yerf-2013-05-02-03:15:01/DB2/DB2-flat.vmdk", O_RDONLY|O_NOATIME) = 6
>> ioctl(5, 0x4020940d, 0x7fffc6d41c60)    = 0
>> close(6)                                = 0
>> read(3, "q\0\0\0\20\0\242>\357\263", 10) = 10
>> read(3, "\22\0\10\0\0\0&~\0\0\0\0\30\0\10\0\0\0\2\0\0\0\0\0\17\0\24\0DB2/"..., 113) = 113
>> open("/media/snaps/yerf-2013-05-02-03:15:01/DB2/DB2-flat.vmdk", O_RDONLY|O_NOATIME) = 6
>> ioctl(5, 0x4020940d, 0x7fffc6d41c60)    = 0
>> close(6)                                = 0
>> --
>> 
> 
> Is this the receive side?
> Where does the data come from, a local file or via network?
> 

Yes, this is the receiving side. data comes from a local file.

sometimes it does hit a 'good' portion, then i get a strafe like this:
read(3, "(\300\0\0\17\0N0\346\307", 10) = 10
read(3, "\17\0\24\0DB2/DB2-flat.vmdk.ok\22\0\10\0\0@\25\325"..., 49192) = 49192
pwrite(5,"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,49152,3574939648) = 49152


Managed to find what ioctl 0x4020940d is in the meantime: BTRFS_IOC_CLONE_RANGE, with 32 bytes parameters (4x int64). have not managed to get the parameters yet. i have no idea how to work gbd.

but i guess they are coming from the 113 bytes it is reading.

Remco

> -Arne
> 
>> it pauses for a second after "ioctl(5, 0x4020940d"
>> it has been running like that for 3 hours now.
>> the file its working is large (80gig) and filefrag reports 648862 extends.
>> filesystem is mounted with rw,relatime,compress-force=lzo,space_cache
>> 
>> anything i can do to see what the problem is?
>> 
>> Remco--
>> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>