* Re: High Fragmentation with XFS and NFS Sync
@ 2016-07-02 19:49 Richard Scobie
2016-07-02 20:41 ` Nick Fisk
0 siblings, 1 reply; 10+ messages in thread
From: Richard Scobie @ 2016-07-02 19:49 UTC (permalink / raw)
To: xfs
Nick Fisk wrote:
"So it looks like each parallel IO thread is being
allocated next to each other rather than at spaced out regions of the disk."
It's possible that the "filestreams" XFS mount option may help you out. See:
http://www.xfs.org/docs/xfsdocs-xml-dev/XFS_User_Guide/tmp/en-US/html/ch06s16.html
Regards,
Richard
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: High Fragmentation with XFS and NFS Sync
2016-07-02 19:49 High Fragmentation with XFS and NFS Sync Richard Scobie
@ 2016-07-02 20:41 ` Nick Fisk
0 siblings, 0 replies; 10+ messages in thread
From: Nick Fisk @ 2016-07-02 20:41 UTC (permalink / raw)
To: Richard Scobie; +Cc: xfs
On 2 July 2016 at 20:49, Richard Scobie <r.scobie@clear.net.nz> wrote:
>
> Nick Fisk wrote:
>
> "So it looks like each parallel IO thread is being
> allocated next to each other rather than at spaced out regions of the disk."
>
> It's possible that the "filestreams" XFS mount option may help you out. See:
>
>
> http://www.xfs.org/docs/xfsdocs-xml-dev/XFS_User_Guide/tmp/en-US/html/ch06s16.html
Thanks, I will see if this helps. Should I have any concern about
ongoing IO once the VM has been copied? That link seems to suggest
problems if the file is appended again at a later date. I believe the
holes shown below are due to it being a sparse file.
Would also using swalloc and a large swidth have any effect, or is
this like the allocsize, ignored once the file is closed?
I've been doing a bit more digging and I think the problem is a bit
more complex than I first thought. ESX seems to multithread the copy
with lots of parallel 64kb io's, I believe looking at xfs_bmap that
some these threads are also getting written out of order, which
properly isn't helping, although this is probably a minor problem in
comparison. This example file is 40GB and has nearly 50,000 extents.
0: [0..127]: 5742779480..5742779607 128 blocks
1: [128..2047]: hole 1920 blocks
2: [2048..2175]: 5742779736..5742779863 128 blocks
3: [2176..2303]: 5742779608..5742779735 128 blocks
4: [2304..2431]: 5742779864..5742779991 128 blocks
5: [2432..4351]: hole 1920 blocks
6: [4352..4607]: 5742779992..5742780247 256 blocks
7: [4608..12543]: hole 7936 blocks
8: [12544..12671]: 5742780248..5742780375 128 blocks
9: [12672..13695]: 5742798928..5742799951 1024 blocks
10: [13696..13823]: 5742813392..5742813519 128 blocks
11: [13824..13951]: 5742813648..5742813775 128 blocks
12: [13952..14079]: 5742813264..5742813391 128 blocks
13: [14080..14207]: 5742813520..5742813647 128 blocks <-- Next
to 10 on disk
14: [14208..14719]: 5742813776..5742814287 512 blocks <-- Next
to 11 on disk
15: [14720..15359]: 5742837840..5742838479 640 blocks
16: [15360..15487]: 5743255760..5743255887 128 blocks
17: [15488..15743]: 5742838608..5742838863 256 blocks
18: [15744..15871]: 5743133904..5743134031 128 blocks
19: [15872..15999]: 5743134288..5743134415 128 blocks
20: [16000..16127]: 5743255632..5743255759 128 blocks
21: [16128..16255]: 5743133776..5743133903 128 blocks
22: [16256..16383]: 5743134032..5743134159 128 blocks
23: [16384..16511]: 5742838480..5742838607 128 blocks
24: [16512..16895]: 5743255888..5743256271 384 blocks
25: [16896..17023]: 5743134416..5743134543 128 blocks
26: [17024..17151]: 5743134160..5743134287 128 blocks
Thanks,
Nick
>
> Regards,
>
> Richard
>
> _______________________________________________
> xfs mailing list
> xfs@oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* High Fragmentation with XFS and NFS Sync
@ 2016-07-02 8:52 Nick Fisk
2016-07-02 20:12 ` Darrick J. Wong
2016-07-03 12:34 ` Stefan Ring
0 siblings, 2 replies; 10+ messages in thread
From: Nick Fisk @ 2016-07-02 8:52 UTC (permalink / raw)
To: xfs
[-- Attachment #1.1: Type: text/plain, Size: 1540 bytes --]
Hi, hope someone can help me here.
I'm exporting some XFS fs's to ESX via NFS with the sync option enabled. I'm
seeing really heavy fragmentation when multiple VM's are copied onto the
share at the same time. I'm also seeing kmem_alloc failures, which is
probably the biggest problem as this effectively takes everything down.
Underlying storage is a Ceph RBD, the server the FS is running on, is
running kernel 4.5.7. Mount options are currently default. I'm seeing
Millions of extents, where the ideal is listed as a couple of thousand when
running xfs_db, there is only a couple of 100 files on the FS. It looks
like roughly the extent sizes roughly match the IO size that the VM's were
written to XFS with. So it looks like each parallel IO thread is being
allocated next to each other rather than at spaced out regions of the disk.
>From what I understand, this is because each NFS write opens and closes the
file which throws off any chance that XFS will be able to use its allocation
features to stop parallel write streams from interleaving with each other.
Is there anything I can tune to try and give each write to each file a
little bit of space, so that it at least gives readahead a chance when
reading, that it might hit at least a few MB of sequential data?
I have read that inode32 allocates more randomly compared to inode64, so I'm
not sure if it's worth trying this as there will likely be less than a 1000
files per FS.
Or am I best just to run fsr after everything has been copied on?
Thanks for any advice
Nick
[-- Attachment #1.2: Type: text/html, Size: 12208 bytes --]
[-- Attachment #2: Type: text/plain, Size: 121 bytes --]
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: High Fragmentation with XFS and NFS Sync
2016-07-02 8:52 Nick Fisk
@ 2016-07-02 20:12 ` Darrick J. Wong
2016-07-02 21:00 ` Nick Fisk
2016-07-03 12:34 ` Stefan Ring
1 sibling, 1 reply; 10+ messages in thread
From: Darrick J. Wong @ 2016-07-02 20:12 UTC (permalink / raw)
To: Nick Fisk; +Cc: xfs
On Sat, Jul 02, 2016 at 09:52:40AM +0100, Nick Fisk wrote:
> Hi, hope someone can help me here.
>
> I'm exporting some XFS fs's to ESX via NFS with the sync option enabled. I'm
> seeing really heavy fragmentation when multiple VM's are copied onto the
> share at the same time. I'm also seeing kmem_alloc failures, which is
> probably the biggest problem as this effectively takes everything down.
(Probably a result of loading the millions of bmbt extents into memory?)
> Underlying storage is a Ceph RBD, the server the FS is running on, is
> running kernel 4.5.7. Mount options are currently default. I'm seeing
> Millions of extents, where the ideal is listed as a couple of thousand when
> running xfs_db, there is only a couple of 100 files on the FS. It looks
> like roughly the extent sizes roughly match the IO size that the VM's were
> written to XFS with. So it looks like each parallel IO thread is being
> allocated next to each other rather than at spaced out regions of the disk.
>
> From what I understand, this is because each NFS write opens and closes the
> file which throws off any chance that XFS will be able to use its allocation
> features to stop parallel write streams from interleaving with each other.
>
> Is there anything I can tune to try and give each write to each file a
> little bit of space, so that it at least gives readahead a chance when
> reading, that it might hit at least a few MB of sequential data?
/me wonders if setting an extent size hint on the rootdir before copying
the files over would help here...
--D
>
> I have read that inode32 allocates more randomly compared to inode64, so I'm
> not sure if it's worth trying this as there will likely be less than a 1000
> files per FS.
>
> Or am I best just to run fsr after everything has been copied on?
>
> Thanks for any advice
> Nick
> _______________________________________________
> xfs mailing list
> xfs@oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: High Fragmentation with XFS and NFS Sync
2016-07-02 20:12 ` Darrick J. Wong
@ 2016-07-02 21:00 ` Nick Fisk
2016-07-02 21:30 ` Nick Fisk
0 siblings, 1 reply; 10+ messages in thread
From: Nick Fisk @ 2016-07-02 21:00 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: xfs
On 2 July 2016 at 21:12, Darrick J. Wong <darrick.wong@oracle.com> wrote:
> On Sat, Jul 02, 2016 at 09:52:40AM +0100, Nick Fisk wrote:
>> Hi, hope someone can help me here.
>>
>> I'm exporting some XFS fs's to ESX via NFS with the sync option enabled. I'm
>> seeing really heavy fragmentation when multiple VM's are copied onto the
>> share at the same time. I'm also seeing kmem_alloc failures, which is
>> probably the biggest problem as this effectively takes everything down.
>
> (Probably a result of loading the millions of bmbt extents into memory?)
Yes I thought that was the case.
>
>> Underlying storage is a Ceph RBD, the server the FS is running on, is
>> running kernel 4.5.7. Mount options are currently default. I'm seeing
>> Millions of extents, where the ideal is listed as a couple of thousand when
>> running xfs_db, there is only a couple of 100 files on the FS. It looks
>> like roughly the extent sizes roughly match the IO size that the VM's were
>> written to XFS with. So it looks like each parallel IO thread is being
>> allocated next to each other rather than at spaced out regions of the disk.
>>
>> From what I understand, this is because each NFS write opens and closes the
>> file which throws off any chance that XFS will be able to use its allocation
>> features to stop parallel write streams from interleaving with each other.
>>
>> Is there anything I can tune to try and give each write to each file a
>> little bit of space, so that it at least gives readahead a chance when
>> reading, that it might hit at least a few MB of sequential data?
>
> /me wonders if setting an extent size hint on the rootdir before copying
> the files over would help here...
I've set a 16M hint and will copy a new VM over, interested to see
what happens. Thanks for the suggestion.
>
> --D
>
>>
>> I have read that inode32 allocates more randomly compared to inode64, so I'm
>> not sure if it's worth trying this as there will likely be less than a 1000
>> files per FS.
>>
>> Or am I best just to run fsr after everything has been copied on?
>>
>> Thanks for any advice
>> Nick
>
>> _______________________________________________
>> xfs mailing list
>> xfs@oss.sgi.com
>> http://oss.sgi.com/mailman/listinfo/xfs
>
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: High Fragmentation with XFS and NFS Sync
2016-07-02 21:00 ` Nick Fisk
@ 2016-07-02 21:30 ` Nick Fisk
2016-07-02 22:02 ` Darrick J. Wong
0 siblings, 1 reply; 10+ messages in thread
From: Nick Fisk @ 2016-07-02 21:30 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: xfs
On 2 July 2016 at 22:00, Nick Fisk <friskyfisk10@googlemail.com> wrote:
> On 2 July 2016 at 21:12, Darrick J. Wong <darrick.wong@oracle.com> wrote:
>> On Sat, Jul 02, 2016 at 09:52:40AM +0100, Nick Fisk wrote:
>>> Hi, hope someone can help me here.
>>>
>>> I'm exporting some XFS fs's to ESX via NFS with the sync option enabled. I'm
>>> seeing really heavy fragmentation when multiple VM's are copied onto the
>>> share at the same time. I'm also seeing kmem_alloc failures, which is
>>> probably the biggest problem as this effectively takes everything down.
>>
>> (Probably a result of loading the millions of bmbt extents into memory?)
>
> Yes I thought that was the case.
>
>>
>>> Underlying storage is a Ceph RBD, the server the FS is running on, is
>>> running kernel 4.5.7. Mount options are currently default. I'm seeing
>>> Millions of extents, where the ideal is listed as a couple of thousand when
>>> running xfs_db, there is only a couple of 100 files on the FS. It looks
>>> like roughly the extent sizes roughly match the IO size that the VM's were
>>> written to XFS with. So it looks like each parallel IO thread is being
>>> allocated next to each other rather than at spaced out regions of the disk.
>>>
>>> From what I understand, this is because each NFS write opens and closes the
>>> file which throws off any chance that XFS will be able to use its allocation
>>> features to stop parallel write streams from interleaving with each other.
>>>
>>> Is there anything I can tune to try and give each write to each file a
>>> little bit of space, so that it at least gives readahead a chance when
>>> reading, that it might hit at least a few MB of sequential data?
>>
>> /me wonders if setting an extent size hint on the rootdir before copying
>> the files over would help here...
>
> I've set a 16M hint and will copy a new VM over, interested to see
> what happens. Thanks for the suggestion.
Well, I set the 16M hint at the root of the FS and proceeded to copy
two VM's in parallel and got this after ~30-60s. But after rebooting,
it looks like the extents were being allocated in larger blocks, so I
guess you can call it progress. Any ideas?
Jul 2 22:11:56 Proxy3 kernel: [48777.591415] XFS (rbd8): Access to
block zero in inode 7054483473 start_block: 0 start_off: 0 blkcnt: 0
extent-state: 0 lastx: 30d18
Jul 2 22:11:56 Proxy3 kernel: [48777.602725] XFS (rbd8): Internal
error XFS_WANT_CORRUPTED_GOTO at line 1947 of file
/home/kernel/COD/linux/fs/xfs/libxfs/xfs_bmap.c. Caller
xfs_bmapi_write+0x749/0xa00 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608381] CPU: 3 PID: 1463 Comm:
nfsd Tainted: G OE 4.5.7-040507-generic #201606100436
Jul 2 22:11:56 Proxy3 kernel: [48777.608385] Hardware name: VMware,
Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS
6.00 09/17/2015
Jul 2 22:11:56 Proxy3 kernel: [48777.608389] 0000000000000286
000000008a69edda ffff8800b76fb548 ffffffff813e1173
Jul 2 22:11:56 Proxy3 kernel: [48777.608395] 00000000000000cc
ffff8800b76fb6e0 ffff8800b76fb560 ffffffffc07ad60c
Jul 2 22:11:56 Proxy3 kernel: [48777.608399] ffffffffc077c959
ffff8800b76fb658 ffffffffc0777a13 ffff8802116c9000
Jul 2 22:11:56 Proxy3 kernel: [48777.608403] Call Trace:
Jul 2 22:11:56 Proxy3 kernel: [48777.608471] [<ffffffff813e1173>]
dump_stack+0x63/0x90
Jul 2 22:11:56 Proxy3 kernel: [48777.608566] [<ffffffffc07ad60c>]
xfs_error_report+0x3c/0x40 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608620] [<ffffffffc077c959>] ?
xfs_bmapi_write+0x749/0xa00 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608669] [<ffffffffc0777a13>]
xfs_bmap_add_extent_delay_real+0x883/0x1ce0 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608721] [<ffffffffc077c959>]
xfs_bmapi_write+0x749/0xa00 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608777] [<ffffffffc07b8a8d>]
xfs_iomap_write_allocate+0x16d/0x380 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608822] [<ffffffffc07a25d3>]
xfs_map_blocks+0x173/0x240 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608871] [<ffffffffc07a33a8>]
xfs_vm_writepage+0x198/0x660 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608902] [<ffffffff811993a3>]
__writepage+0x13/0x30
Jul 2 22:11:56 Proxy3 kernel: [48777.608908] [<ffffffff81199efe>]
write_cache_pages+0x1fe/0x530
Jul 2 22:11:56 Proxy3 kernel: [48777.608912] [<ffffffff81199390>] ?
wb_position_ratio+0x1f0/0x1f0
Jul 2 22:11:56 Proxy3 kernel: [48777.608961] [<ffffffffc07bbbaa>] ?
xfs_iunlock+0xea/0x120 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.608970] [<ffffffff8119a281>]
generic_writepages+0x51/0x80
Jul 2 22:11:56 Proxy3 kernel: [48777.609019] [<ffffffffc07a31c3>]
xfs_vm_writepages+0x53/0xa0 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.609028] [<ffffffff8119c73e>]
do_writepages+0x1e/0x30
Jul 2 22:11:56 Proxy3 kernel: [48777.609048] [<ffffffff8118f516>]
__filemap_fdatawrite_range+0xc6/0x100
Jul 2 22:11:56 Proxy3 kernel: [48777.609053] [<ffffffff8118f691>]
filemap_write_and_wait_range+0x41/0x90
Jul 2 22:11:56 Proxy3 kernel: [48777.609103] [<ffffffffc07af7d3>]
xfs_file_fsync+0x63/0x210 [xfs]
Jul 2 22:11:56 Proxy3 kernel: [48777.609130] [<ffffffff81249e9b>]
vfs_fsync_range+0x4b/0xb0
Jul 2 22:11:56 Proxy3 kernel: [48777.609156] [<ffffffffc04b856d>]
nfsd_vfs_write+0x14d/0x380 [nfsd]
Jul 2 22:11:56 Proxy3 kernel: [48777.609176] [<ffffffffc04babf0>]
nfsd_write+0x120/0x2f0 [nfsd]
Jul 2 22:11:56 Proxy3 kernel: [48777.609190] [<ffffffffc04c105c>]
nfsd3_proc_write+0xbc/0x150 [nfsd]
Jul 2 22:11:56 Proxy3 kernel: [48777.609207] [<ffffffffc04b3348>]
nfsd_dispatch+0xb8/0x200 [nfsd]
Jul 2 22:11:56 Proxy3 kernel: [48777.609243] [<ffffffffc03cd21c>]
svc_process_common+0x40c/0x650 [sunrpc]
Jul 2 22:11:56 Proxy3 kernel: [48777.609267] [<ffffffffc03ce5c3>]
svc_process+0x103/0x1b0 [sunrpc]
Jul 2 22:11:56 Proxy3 kernel: [48777.609286] [<ffffffffc04b2d8f>]
nfsd+0xef/0x160 [nfsd]
Jul 2 22:11:56 Proxy3 kernel: [48777.609297] [<ffffffffc04b2ca0>] ?
nfsd_destroy+0x60/0x60 [nfsd]
Jul 2 22:11:56 Proxy3 kernel: [48777.609322] [<ffffffff810a06a8>]
kthread+0xd8/0xf0
Jul 2 22:11:56 Proxy3 kernel: [48777.609329] [<ffffffff810a05d0>] ?
kthread_create_on_node+0x1a0/0x1a0
Jul 2 22:11:56 Proxy3 kernel: [48777.609369] [<ffffffff81825bcf>]
ret_from_fork+0x3f/0x70
Jul 2 22:11:56 Proxy3 kernel: [48777.609374] [<ffffffff810a05d0>] ?
kthread_create_on_node+0x1a0/0x1a0
Jul 2 22:11:56 Proxy3 kernel: [48777.609444] XFS (rbd8): Internal
error xfs_trans_cancel at line 990 of file
/home/kernel/COD/linux/fs/xfs/xfs_trans.c. Caller
xfs_iomap_write_allocate+0x270/0x380 [xfs]
>
>>
>> --D
>>
>>>
>>> I have read that inode32 allocates more randomly compared to inode64, so I'm
>>> not sure if it's worth trying this as there will likely be less than a 1000
>>> files per FS.
>>>
>>> Or am I best just to run fsr after everything has been copied on?
>>>
>>> Thanks for any advice
>>> Nick
>>
>>> _______________________________________________
>>> xfs mailing list
>>> xfs@oss.sgi.com
>>> http://oss.sgi.com/mailman/listinfo/xfs
>>
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: High Fragmentation with XFS and NFS Sync
2016-07-02 21:30 ` Nick Fisk
@ 2016-07-02 22:02 ` Darrick J. Wong
2016-07-02 22:10 ` Nick Fisk
0 siblings, 1 reply; 10+ messages in thread
From: Darrick J. Wong @ 2016-07-02 22:02 UTC (permalink / raw)
To: Nick Fisk; +Cc: xfs
On Sat, Jul 02, 2016 at 10:30:56PM +0100, Nick Fisk wrote:
> On 2 July 2016 at 22:00, Nick Fisk <friskyfisk10@googlemail.com> wrote:
> > On 2 July 2016 at 21:12, Darrick J. Wong <darrick.wong@oracle.com> wrote:
> >> On Sat, Jul 02, 2016 at 09:52:40AM +0100, Nick Fisk wrote:
> >>> Hi, hope someone can help me here.
> >>>
> >>> I'm exporting some XFS fs's to ESX via NFS with the sync option enabled. I'm
> >>> seeing really heavy fragmentation when multiple VM's are copied onto the
> >>> share at the same time. I'm also seeing kmem_alloc failures, which is
> >>> probably the biggest problem as this effectively takes everything down.
> >>
> >> (Probably a result of loading the millions of bmbt extents into memory?)
> >
> > Yes I thought that was the case.
> >
> >>
> >>> Underlying storage is a Ceph RBD, the server the FS is running on, is
> >>> running kernel 4.5.7. Mount options are currently default. I'm seeing
> >>> Millions of extents, where the ideal is listed as a couple of thousand when
> >>> running xfs_db, there is only a couple of 100 files on the FS. It looks
> >>> like roughly the extent sizes roughly match the IO size that the VM's were
> >>> written to XFS with. So it looks like each parallel IO thread is being
> >>> allocated next to each other rather than at spaced out regions of the disk.
> >>>
> >>> From what I understand, this is because each NFS write opens and closes the
> >>> file which throws off any chance that XFS will be able to use its allocation
> >>> features to stop parallel write streams from interleaving with each other.
> >>>
> >>> Is there anything I can tune to try and give each write to each file a
> >>> little bit of space, so that it at least gives readahead a chance when
> >>> reading, that it might hit at least a few MB of sequential data?
> >>
> >> /me wonders if setting an extent size hint on the rootdir before copying
> >> the files over would help here...
> >
> > I've set a 16M hint and will copy a new VM over, interested to see
> > what happens. Thanks for the suggestion.
>
> Well, I set the 16M hint at the root of the FS and proceeded to copy
> two VM's in parallel and got this after ~30-60s. But after rebooting,
> it looks like the extents were being allocated in larger blocks, so I
> guess you can call it progress. Any ideas?
Yikes. :(
Is that with or without filestreams?
--D
>
>
> Jul 2 22:11:56 Proxy3 kernel: [48777.591415] XFS (rbd8): Access to
> block zero in inode 7054483473 start_block: 0 start_off: 0 blkcnt: 0
> extent-state: 0 lastx: 30d18
> Jul 2 22:11:56 Proxy3 kernel: [48777.602725] XFS (rbd8): Internal
> error XFS_WANT_CORRUPTED_GOTO at line 1947 of file
> /home/kernel/COD/linux/fs/xfs/libxfs/xfs_bmap.c. Caller
> xfs_bmapi_write+0x749/0xa00 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608381] CPU: 3 PID: 1463 Comm:
> nfsd Tainted: G OE 4.5.7-040507-generic #201606100436
> Jul 2 22:11:56 Proxy3 kernel: [48777.608385] Hardware name: VMware,
> Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS
> 6.00 09/17/2015
> Jul 2 22:11:56 Proxy3 kernel: [48777.608389] 0000000000000286
> 000000008a69edda ffff8800b76fb548 ffffffff813e1173
> Jul 2 22:11:56 Proxy3 kernel: [48777.608395] 00000000000000cc
> ffff8800b76fb6e0 ffff8800b76fb560 ffffffffc07ad60c
> Jul 2 22:11:56 Proxy3 kernel: [48777.608399] ffffffffc077c959
> ffff8800b76fb658 ffffffffc0777a13 ffff8802116c9000
> Jul 2 22:11:56 Proxy3 kernel: [48777.608403] Call Trace:
> Jul 2 22:11:56 Proxy3 kernel: [48777.608471] [<ffffffff813e1173>]
> dump_stack+0x63/0x90
> Jul 2 22:11:56 Proxy3 kernel: [48777.608566] [<ffffffffc07ad60c>]
> xfs_error_report+0x3c/0x40 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608620] [<ffffffffc077c959>] ?
> xfs_bmapi_write+0x749/0xa00 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608669] [<ffffffffc0777a13>]
> xfs_bmap_add_extent_delay_real+0x883/0x1ce0 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608721] [<ffffffffc077c959>]
> xfs_bmapi_write+0x749/0xa00 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608777] [<ffffffffc07b8a8d>]
> xfs_iomap_write_allocate+0x16d/0x380 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608822] [<ffffffffc07a25d3>]
> xfs_map_blocks+0x173/0x240 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608871] [<ffffffffc07a33a8>]
> xfs_vm_writepage+0x198/0x660 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608902] [<ffffffff811993a3>]
> __writepage+0x13/0x30
> Jul 2 22:11:56 Proxy3 kernel: [48777.608908] [<ffffffff81199efe>]
> write_cache_pages+0x1fe/0x530
> Jul 2 22:11:56 Proxy3 kernel: [48777.608912] [<ffffffff81199390>] ?
> wb_position_ratio+0x1f0/0x1f0
> Jul 2 22:11:56 Proxy3 kernel: [48777.608961] [<ffffffffc07bbbaa>] ?
> xfs_iunlock+0xea/0x120 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.608970] [<ffffffff8119a281>]
> generic_writepages+0x51/0x80
> Jul 2 22:11:56 Proxy3 kernel: [48777.609019] [<ffffffffc07a31c3>]
> xfs_vm_writepages+0x53/0xa0 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609028] [<ffffffff8119c73e>]
> do_writepages+0x1e/0x30
> Jul 2 22:11:56 Proxy3 kernel: [48777.609048] [<ffffffff8118f516>]
> __filemap_fdatawrite_range+0xc6/0x100
> Jul 2 22:11:56 Proxy3 kernel: [48777.609053] [<ffffffff8118f691>]
> filemap_write_and_wait_range+0x41/0x90
> Jul 2 22:11:56 Proxy3 kernel: [48777.609103] [<ffffffffc07af7d3>]
> xfs_file_fsync+0x63/0x210 [xfs]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609130] [<ffffffff81249e9b>]
> vfs_fsync_range+0x4b/0xb0
> Jul 2 22:11:56 Proxy3 kernel: [48777.609156] [<ffffffffc04b856d>]
> nfsd_vfs_write+0x14d/0x380 [nfsd]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609176] [<ffffffffc04babf0>]
> nfsd_write+0x120/0x2f0 [nfsd]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609190] [<ffffffffc04c105c>]
> nfsd3_proc_write+0xbc/0x150 [nfsd]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609207] [<ffffffffc04b3348>]
> nfsd_dispatch+0xb8/0x200 [nfsd]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609243] [<ffffffffc03cd21c>]
> svc_process_common+0x40c/0x650 [sunrpc]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609267] [<ffffffffc03ce5c3>]
> svc_process+0x103/0x1b0 [sunrpc]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609286] [<ffffffffc04b2d8f>]
> nfsd+0xef/0x160 [nfsd]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609297] [<ffffffffc04b2ca0>] ?
> nfsd_destroy+0x60/0x60 [nfsd]
> Jul 2 22:11:56 Proxy3 kernel: [48777.609322] [<ffffffff810a06a8>]
> kthread+0xd8/0xf0
> Jul 2 22:11:56 Proxy3 kernel: [48777.609329] [<ffffffff810a05d0>] ?
> kthread_create_on_node+0x1a0/0x1a0
> Jul 2 22:11:56 Proxy3 kernel: [48777.609369] [<ffffffff81825bcf>]
> ret_from_fork+0x3f/0x70
> Jul 2 22:11:56 Proxy3 kernel: [48777.609374] [<ffffffff810a05d0>] ?
> kthread_create_on_node+0x1a0/0x1a0
> Jul 2 22:11:56 Proxy3 kernel: [48777.609444] XFS (rbd8): Internal
> error xfs_trans_cancel at line 990 of file
> /home/kernel/COD/linux/fs/xfs/xfs_trans.c. Caller
> xfs_iomap_write_allocate+0x270/0x380 [xfs]
>
> >
> >>
> >> --D
> >>
> >>>
> >>> I have read that inode32 allocates more randomly compared to inode64, so I'm
> >>> not sure if it's worth trying this as there will likely be less than a 1000
> >>> files per FS.
> >>>
> >>> Or am I best just to run fsr after everything has been copied on?
> >>>
> >>> Thanks for any advice
> >>> Nick
> >>
> >>> _______________________________________________
> >>> xfs mailing list
> >>> xfs@oss.sgi.com
> >>> http://oss.sgi.com/mailman/listinfo/xfs
> >>
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: High Fragmentation with XFS and NFS Sync
2016-07-02 22:02 ` Darrick J. Wong
@ 2016-07-02 22:10 ` Nick Fisk
0 siblings, 0 replies; 10+ messages in thread
From: Nick Fisk @ 2016-07-02 22:10 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: xfs
On 2 July 2016 at 23:02, Darrick J. Wong <darrick.wong@oracle.com> wrote:
> On Sat, Jul 02, 2016 at 10:30:56PM +0100, Nick Fisk wrote:
>> On 2 July 2016 at 22:00, Nick Fisk <friskyfisk10@googlemail.com> wrote:
>> > On 2 July 2016 at 21:12, Darrick J. Wong <darrick.wong@oracle.com> wrote:
>> >> On Sat, Jul 02, 2016 at 09:52:40AM +0100, Nick Fisk wrote:
>> >>> Hi, hope someone can help me here.
>> >>>
>> >>> I'm exporting some XFS fs's to ESX via NFS with the sync option enabled. I'm
>> >>> seeing really heavy fragmentation when multiple VM's are copied onto the
>> >>> share at the same time. I'm also seeing kmem_alloc failures, which is
>> >>> probably the biggest problem as this effectively takes everything down.
>> >>
>> >> (Probably a result of loading the millions of bmbt extents into memory?)
>> >
>> > Yes I thought that was the case.
>> >
>> >>
>> >>> Underlying storage is a Ceph RBD, the server the FS is running on, is
>> >>> running kernel 4.5.7. Mount options are currently default. I'm seeing
>> >>> Millions of extents, where the ideal is listed as a couple of thousand when
>> >>> running xfs_db, there is only a couple of 100 files on the FS. It looks
>> >>> like roughly the extent sizes roughly match the IO size that the VM's were
>> >>> written to XFS with. So it looks like each parallel IO thread is being
>> >>> allocated next to each other rather than at spaced out regions of the disk.
>> >>>
>> >>> From what I understand, this is because each NFS write opens and closes the
>> >>> file which throws off any chance that XFS will be able to use its allocation
>> >>> features to stop parallel write streams from interleaving with each other.
>> >>>
>> >>> Is there anything I can tune to try and give each write to each file a
>> >>> little bit of space, so that it at least gives readahead a chance when
>> >>> reading, that it might hit at least a few MB of sequential data?
>> >>
>> >> /me wonders if setting an extent size hint on the rootdir before copying
>> >> the files over would help here...
>> >
>> > I've set a 16M hint and will copy a new VM over, interested to see
>> > what happens. Thanks for the suggestion.
>>
>> Well, I set the 16M hint at the root of the FS and proceeded to copy
>> two VM's in parallel and got this after ~30-60s. But after rebooting,
>> it looks like the extents were being allocated in larger blocks, so I
>> guess you can call it progress. Any ideas?
>
> Yikes. :(
>
> Is that with or without filestreams?
No filestreams at the moment. That was just after running xfs_io on
the root of the FS with "-D -R 16M". I realise now after re-reading
the man page I didn't need to specify it recursively to take effect on
new files/directories.
xfs_repair has just finished and it looks like the inode that had the
problem was for another file, not the two that i was copying. Output
from xfs_repair:
correcting bt key (was 1068456, now 1068420) in inode 7054483473
data fork, btree block 882994806
bad fwd (right) sibling pointer (saw 886329853 parent block says 896589269)
in inode 7054483473 (data fork) bmap btree block 896588736
bad data fork in inode 7054483473
cleared inode 7054483473
entry "CLI-*****-FP-000001-sesparse.vmdk" at block 0 offset 1168 in
directory inode 7054483457 references free inode 7054483473
clearing inode number in entry at offset 1168...
So I don't know if:
1. The changing of the hint had something to do with it
2. There was a xfs_fsr still running on the FS, so this could have
been the cause, but timing seems a bit coincidental.
I've kicked off another copy and it seems to be working better now,
certainly much bigger extents, so don't know if I was just unlucky.
0: [0..32767]: 268460040..268492807 32768 blocks
1: [32768..65535]: 268541784..268574551 32768 blocks
2: [65536..131071]: 268591112..268656647 65536 blocks
3: [131072..163839]: 268664960..268697727 32768 blocks
4: [163840..229375]: 268706904..268772439 65536 blocks
5: [229376..262143]: 268786776..268819543 32768 blocks
6: [262144..327679]: hole 65536 blocks
7: [327680..458751]: 268885080..269016151 131072 blocks
8: [458752..655359]: 269025408..269222015 196608 blocks
9: [655360..884735]: 269246080..269475455 229376 blocks
10: [884736..950271]: 269501016..269566551 65536 blocks
11: [950272..983039]: 269737960..269770727 32768 blocks
12: [983040..1048575]: 269794920..269860455 65536 blocks
13: [1048576..1114111]: 268819544..268885079 65536 blocks
14: [1114112..1212415]: 274521984..274620287 98304 blocks
15: [1212416..1238143]: 275996672..276022399 25728 blocks
16: [1238144..1310719]: 276022400..276094975 72576 blocks
17: [1310720..1376255]: 276119424..276184959 65536 blocks
18: [1376256..1441791]: 276188416..276253951 65536 blocks
19: [1441792..1671167]: 276282496..276511871 229376 blocks
20: [1671168..1802239]: 277119992..277251063 131072 blocks
>
> --D
>
>>
>>
>> Jul 2 22:11:56 Proxy3 kernel: [48777.591415] XFS (rbd8): Access to
>> block zero in inode 7054483473 start_block: 0 start_off: 0 blkcnt: 0
>> extent-state: 0 lastx: 30d18
>> Jul 2 22:11:56 Proxy3 kernel: [48777.602725] XFS (rbd8): Internal
>> error XFS_WANT_CORRUPTED_GOTO at line 1947 of file
>> /home/kernel/COD/linux/fs/xfs/libxfs/xfs_bmap.c. Caller
>> xfs_bmapi_write+0x749/0xa00 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608381] CPU: 3 PID: 1463 Comm:
>> nfsd Tainted: G OE 4.5.7-040507-generic #201606100436
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608385] Hardware name: VMware,
>> Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS
>> 6.00 09/17/2015
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608389] 0000000000000286
>> 000000008a69edda ffff8800b76fb548 ffffffff813e1173
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608395] 00000000000000cc
>> ffff8800b76fb6e0 ffff8800b76fb560 ffffffffc07ad60c
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608399] ffffffffc077c959
>> ffff8800b76fb658 ffffffffc0777a13 ffff8802116c9000
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608403] Call Trace:
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608471] [<ffffffff813e1173>]
>> dump_stack+0x63/0x90
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608566] [<ffffffffc07ad60c>]
>> xfs_error_report+0x3c/0x40 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608620] [<ffffffffc077c959>] ?
>> xfs_bmapi_write+0x749/0xa00 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608669] [<ffffffffc0777a13>]
>> xfs_bmap_add_extent_delay_real+0x883/0x1ce0 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608721] [<ffffffffc077c959>]
>> xfs_bmapi_write+0x749/0xa00 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608777] [<ffffffffc07b8a8d>]
>> xfs_iomap_write_allocate+0x16d/0x380 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608822] [<ffffffffc07a25d3>]
>> xfs_map_blocks+0x173/0x240 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608871] [<ffffffffc07a33a8>]
>> xfs_vm_writepage+0x198/0x660 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608902] [<ffffffff811993a3>]
>> __writepage+0x13/0x30
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608908] [<ffffffff81199efe>]
>> write_cache_pages+0x1fe/0x530
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608912] [<ffffffff81199390>] ?
>> wb_position_ratio+0x1f0/0x1f0
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608961] [<ffffffffc07bbbaa>] ?
>> xfs_iunlock+0xea/0x120 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.608970] [<ffffffff8119a281>]
>> generic_writepages+0x51/0x80
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609019] [<ffffffffc07a31c3>]
>> xfs_vm_writepages+0x53/0xa0 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609028] [<ffffffff8119c73e>]
>> do_writepages+0x1e/0x30
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609048] [<ffffffff8118f516>]
>> __filemap_fdatawrite_range+0xc6/0x100
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609053] [<ffffffff8118f691>]
>> filemap_write_and_wait_range+0x41/0x90
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609103] [<ffffffffc07af7d3>]
>> xfs_file_fsync+0x63/0x210 [xfs]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609130] [<ffffffff81249e9b>]
>> vfs_fsync_range+0x4b/0xb0
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609156] [<ffffffffc04b856d>]
>> nfsd_vfs_write+0x14d/0x380 [nfsd]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609176] [<ffffffffc04babf0>]
>> nfsd_write+0x120/0x2f0 [nfsd]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609190] [<ffffffffc04c105c>]
>> nfsd3_proc_write+0xbc/0x150 [nfsd]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609207] [<ffffffffc04b3348>]
>> nfsd_dispatch+0xb8/0x200 [nfsd]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609243] [<ffffffffc03cd21c>]
>> svc_process_common+0x40c/0x650 [sunrpc]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609267] [<ffffffffc03ce5c3>]
>> svc_process+0x103/0x1b0 [sunrpc]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609286] [<ffffffffc04b2d8f>]
>> nfsd+0xef/0x160 [nfsd]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609297] [<ffffffffc04b2ca0>] ?
>> nfsd_destroy+0x60/0x60 [nfsd]
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609322] [<ffffffff810a06a8>]
>> kthread+0xd8/0xf0
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609329] [<ffffffff810a05d0>] ?
>> kthread_create_on_node+0x1a0/0x1a0
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609369] [<ffffffff81825bcf>]
>> ret_from_fork+0x3f/0x70
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609374] [<ffffffff810a05d0>] ?
>> kthread_create_on_node+0x1a0/0x1a0
>> Jul 2 22:11:56 Proxy3 kernel: [48777.609444] XFS (rbd8): Internal
>> error xfs_trans_cancel at line 990 of file
>> /home/kernel/COD/linux/fs/xfs/xfs_trans.c. Caller
>> xfs_iomap_write_allocate+0x270/0x380 [xfs]
>>
>> >
>> >>
>> >> --D
>> >>
>> >>>
>> >>> I have read that inode32 allocates more randomly compared to inode64, so I'm
>> >>> not sure if it's worth trying this as there will likely be less than a 1000
>> >>> files per FS.
>> >>>
>> >>> Or am I best just to run fsr after everything has been copied on?
>> >>>
>> >>> Thanks for any advice
>> >>> Nick
>> >>
>> >>> _______________________________________________
>> >>> xfs mailing list
>> >>> xfs@oss.sgi.com
>> >>> http://oss.sgi.com/mailman/listinfo/xfs
>> >>
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: High Fragmentation with XFS and NFS Sync
2016-07-02 8:52 Nick Fisk
2016-07-02 20:12 ` Darrick J. Wong
@ 2016-07-03 12:34 ` Stefan Ring
2016-07-03 12:40 ` Nick Fisk
1 sibling, 1 reply; 10+ messages in thread
From: Stefan Ring @ 2016-07-03 12:34 UTC (permalink / raw)
To: Nick Fisk; +Cc: Linux fs XFS
On Sat, Jul 2, 2016 at 10:52 AM, Nick Fisk <friskyfisk10@googlemail.com> wrote:
> Hi, hope someone can help me here.
>
> I'm exporting some XFS fs's to ESX via NFS with the sync option enabled. I'm
> seeing really heavy fragmentation when multiple VM's are copied onto the
> share at the same time. I'm also seeing kmem_alloc failures, which is
> probably the biggest problem as this effectively takes everything down.
>
The same thing happens with MySQL's InnoDB files if O_DIRECT is in use.
The kmem_alloc failure taking everything down is most likely a result
of you calling xfs_bmap on these files and does not happen during
normal use. I've had the same experience some time ago.
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: High Fragmentation with XFS and NFS Sync
2016-07-03 12:34 ` Stefan Ring
@ 2016-07-03 12:40 ` Nick Fisk
0 siblings, 0 replies; 10+ messages in thread
From: Nick Fisk @ 2016-07-03 12:40 UTC (permalink / raw)
To: Stefan Ring; +Cc: Linux fs XFS
[-- Attachment #1.1: Type: text/plain, Size: 1002 bytes --]
On 3 Jul 2016 1:34 p.m., "Stefan Ring" <stefanrin@gmail.com> wrote:
>
> On Sat, Jul 2, 2016 at 10:52 AM, Nick Fisk <friskyfisk10@googlemail.com>
wrote:
> > Hi, hope someone can help me here.
> >
> > I'm exporting some XFS fs's to ESX via NFS with the sync option
enabled. I'm
> > seeing really heavy fragmentation when multiple VM's are copied onto the
> > share at the same time. I'm also seeing kmem_alloc failures, which is
> > probably the biggest problem as this effectively takes everything down.
> >
>
> The same thing happens with MySQL's InnoDB files if O_DIRECT is in use.
>
> The kmem_alloc failure taking everything down is most likely a result
> of you calling xfs_bmap on these files and does not happen during
> normal use. I've had the same experience some time ago.
Unfortunately this did happen during normal use. It occurred over night
after kicking a bunch copies off. Something like 5 million extents for
10-20 files. I suspect as suggested, it's to do with memory fragmentation.
[-- Attachment #1.2: Type: text/html, Size: 1315 bytes --]
[-- Attachment #2: Type: text/plain, Size: 121 bytes --]
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2016-07-03 12:40 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-07-02 19:49 High Fragmentation with XFS and NFS Sync Richard Scobie
2016-07-02 20:41 ` Nick Fisk
-- strict thread matches above, loose matches on Subject: below --
2016-07-02 8:52 Nick Fisk
2016-07-02 20:12 ` Darrick J. Wong
2016-07-02 21:00 ` Nick Fisk
2016-07-02 21:30 ` Nick Fisk
2016-07-02 22:02 ` Darrick J. Wong
2016-07-02 22:10 ` Nick Fisk
2016-07-03 12:34 ` Stefan Ring
2016-07-03 12:40 ` Nick Fisk
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox