btrfs: balance dirty metadata pages in btrfs_finish_ordered_io
PATCH 1/1 btrfs: balance dirty metadata pages in btrfs_finish_ordered_io
We should balance dirty metadata pages at the end of btrfs_finish_ordered_io, since a small, unmergeable random write can potentially produce dirty metadata which is multiple times larger than the data itself. For example, a small, unmergeable 4KiB write may produce:
16KiB dirty leaf (and possibly 16KiB dirty node) in subvolume tree 16KiB dirty leaf (and possibly 16KiB dirty node) in checksum tree 16KiB dirty leaf (and possibly 16KiB dirty node) in extent tree
Although we do call balance dirty pages in write side, but in the buffered write path, most metadata are dirtied only after we reach the dirty background limit (which by far onlys counts dirty data pages) and wakeup the flusher thread. If there are many small, unmergeable random writes spread in a large btree, we'll find a burst of dirty pages exceeds the dirty_bytes limit after we wakeup the flusher thread - which is not what we expect. In our machine, it caused out-of-memory problem since a page cannot be dropped if it is marked dirty.
Someone may worry about we may sleep in btrfs_btree_balance_dirty(), but since we do btrfs_finish_ordered_io in a separate worker, it will not stop the flusher consuming dirty pages. Also, we use different worker for metadata writeback endio, sleep in btrfs_finish_ordered_io help us throttle the size of dirty metadata pages.
--- fs/btrfs/inode.c | 2 ++ 1 file changed, 2 insertions(+)
fs/btrfs/inode.c +2 -0
@@ -3151,6 +3151,8 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)/* once for the tree */ btrfs_put_ordered_extent (ordered_extent); +btrfs_btree_balance_dirty (fs_info); +return ret; }
Signature · 1 line
2.17.0
Quoted 19 lines · Ethan Lien
We should balance dirty metadata pages at the end of
btrfs_finish_ordered_io, since a small, unmergeable random write can potentially produce dirty metadata which is multiple times larger than the data itself. For example, a small, unmergeable 4KiB write may produce:16KiB dirty leaf (and possibly 16KiB dirty node) in subvolume tree 16KiB dirty leaf (and possibly 16KiB dirty node) in checksum tree 16KiB dirty leaf (and possibly 16KiB dirty node) in extent tree
Although we do call balance dirty pages in write side, but in the buffered write path, most metadata are dirtied only after we reach the dirty background limit (which by far onlys counts dirty data pages) and wakeup the flusher thread. If there are many small, unmergeable random writes spread in a large btree, we'll find a burst of dirty pages exceeds the
dirty_byteslimit after we wakeup the flusher thread - which is not what we expect. In our machine, it caused out-of-memory problem since a page cannot be dropped if it is marked dirty.
Quoted 5 lines
Someone may worry about we may sleep in
btrfs_btree_balance_dirty(), but since we dobtrfs_finish_ordered_ioin a separate worker, it will not stop the flusher consuming dirty pages. Also, we use different worker for metadata writeback endio, sleep inbtrfs_finish_ordered_iohelp us throttle the size of dirty metadata pages.
The described scenario sounds interesting. Do you have some scripted steps to reproduce it?
btrfs_btree_balance_dirty detects congestion and can skip the balancing eventually, so the case when it actually does something and waits is the point where things can go bad. From the last paragraph, it is clear that you have considered that, that's good.
Have you also considered calling __btrfs_btree_balance_dirty with flush_delayed=0 ? This would avoid the waiting and I'm not sure if it's really required here to get out of the situation.
Anyway, I'll add the patch to for-next for more testing.
David Sterba 於 2018-05-23 00:28 寫到:
Quoted 20 lines · Ethan Lien
Quoted 19 lines
We should balance dirty metadata pages at the end of
btrfs_finish_ordered_io, since a small, unmergeable random write can potentially produce dirty metadata which is multiple times larger than the data itself. For example, a small, unmergeable 4KiB write may produce:16KiB dirty leaf (and possibly 16KiB dirty node) in subvolume tree 16KiB dirty leaf (and possibly 16KiB dirty node) in checksum tree 16KiB dirty leaf (and possibly 16KiB dirty node) in extent tree
Although we do call balance dirty pages in write side, but in the buffered write path, most metadata are dirtied only after we reach the dirty background limit (which by far onlys counts dirty data pages) and wakeup the flusher thread. If there are many small, unmergeable random writes spread in a large btree, we'll find a burst of dirty pages exceeds the
dirty_byteslimit after we wakeup the flusher thread - which is not what we expect. In our machine, it caused out-of-memory problem since a page cannot be dropped if it is marked dirty.
Quoted 8 lines
Quoted 5 lines
Someone may worry about we may sleep in
btrfs_btree_balance_dirty(), but since we dobtrfs_finish_ordered_ioin a separate worker, it will not stop the flusher consuming dirty pages. Also, we use different worker for metadata writeback endio, sleep inbtrfs_finish_ordered_iohelp us throttle the size of dirty metadata pages.The described scenario sounds interesting. Do you have some scripted steps to reproduce it?
It needs some time to reproduce the problem. In our case,
1. Create 4 subvolumes.
2. Run fio on each subvolume:
[global]
direct=0
rw=randwrite
ioengine=libaio
bs=4k
iodepth=16
numjobs=1group_reporting
size=128G
runtime=1800
norandommaptime_based
randrepeat=0
3. Take snapshot on each subvolume and repeat fio on existing files.
4. Repeat step 2&3 until we get really big btrees.
In our case, by observing btrfs_root_item->bytes_used, we have 2GiB of metadata in each subvolume tree and 12GiB of metadata in extent tree.
5. Stop all fio, take snapshot again, and wait until all delayed work is completed.
6. Start all fio. Few seconds later we hit OOM when the flusher starts to work.
It can be reproduced even when using nocow write.
Quoted 8 lines
btrfs_btree_balance_dirtydetects congestion and can skip the balancing eventually, so the case when it actually does something and waits is the point where things can go bad. From the last paragraph, it is clear that you have considered that, that's good.Have you also considered calling
__btrfs_btree_balance_dirtywith flush_delayed=0 ? This would avoid the waiting and I'm not sure if it's really required here to get out of the situation.
Yes, btrfs_btree_balance_dirty_nodelay seems to be a better choice, I'll add it to the v2 patch, thanks.
Quoted 5 lines
Anyway, I'll add the patch to for-next for more testing.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html