os/bluestore: partial reshard support #13162
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
liewegas
changed the title
liewegas
force-pushed
the
wip-bluestore-reshard
branch
2 times, most recently
from
a3e3ffc
to
7230c6f
Compare
ifed01
reviewed
Jan 31, 2017
src/os/bluestore/BlueStore.cc
Outdated
|
|
||
| // unspan spanning blobs | ||
| if (e.blob->is_spanning()) { | ||
| spanning_blob_map.erase(e.blob->id); |
There was a problem hiding this comment.
Can we simply call spanning_blob_map.clear() after the loop?
There was a problem hiding this comment.
not anymore.. we only want to unspan the blobs within the given range
ifed01
reviewed
Jan 31, 2017
src/os/bluestore/BlueStore.cc
Outdated
| if (len > cct->_conf->bluestore_extent_map_shard_max_size) { | ||
| // we are big; reshard ourselves | ||
| request_reshard(p->offset, endoff); | ||
| return; |
There was a problem hiding this comment.
Is that correct to return here and below? Shouldn't we process all dirty shards as this can expand need_reshard range?
| b->id = bid++; | ||
| spanning_blob_map[b->id] = b; | ||
| dout(20) << __func__ << " adding spanning " << *b << dendl; | ||
| } else { |
There was a problem hiding this comment.
else -> else if(e->blob->is_spanning())
src/os/bluestore/BlueStore.cc
Outdated
| unsigned n = sv.size(); | ||
| si_end = si_begin + new_shard_info.size(); | ||
| for (unsigned i = si_begin; i < si_end; ++i) { | ||
| shards[i].offset = sv[i].offset; |
There was a problem hiding this comment.
Wondering if we need offset field in Shard at all, isn't one in shard_info enough?
src/os/bluestore/BlueStore.h
Outdated
| @@ -633,14 +662,31 @@ class BlueStore : public ObjectStore, | |||
| struct Shard { | |||
| bluestore_onode_t::shard_info *shard_info; | |||
| si_end < shards.size())) { | ||
| // we resharded a partial range; we must produce at least one output | ||
| // shard | ||
| new_shard_info.emplace_back(bluestore_onode_t::shard_info()); |
There was a problem hiding this comment.
You might want to implement shard_info(offset) ctor and apply it here and throughout?
Signed-off-by: Sage Weil <sage@redhat.com>
This gives us a single point of control for reshard, and tells us which range we care about at the same time. Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
…ently Split ourselves, or merge with our immediate predecessor or successor. Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
This ensure we mop up shards past EOF instead of encoding them empty and confusing future code (that, say, assumes no shards past EOF). Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
liewegas
force-pushed
the
wip-bluestore-reshard
branch
from
7230c6f
to
dd4f978
Compare
liewegas
changed the title
|
Updated, with several bugs fixed. And rebased now that use_tracker is merged. Please take a look! |
Signed-off-by: Sage Weil <sage@redhat.com>
It's much more helpful this way. Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Only reshard extents in the requested range. Note that the strategy for unspanning blobs changed; we now span or unspan specific blobs at the end based on our reshard result instead of unspanning all at the start. This keeps the spanning id stable, which is important because it may be referenced from another that we aren't even looking at. Also note that this requires a bit of restructuring: an encode_some may hit a spanning shard, *requiring* us to reshard, which means we should always conduct the initial pass through update to discover other reshard requirements, even if we already know some resharding will be needed. Signed-off-by: Sage Weil <sage@redhat.com>
We can do this with a single allocation with a simple vector<>. Signed-off-by: Sage Weil <sage@redhat.com>
This is mirrored in shard_info, which we have a pointer to. Note that shard_info is also a bit redundant as we can also look it up in the onode vector in the same position, but it makes it more awkward to use iterators. Something to consider later to save memory. Signed-off-by: Sage Weil <sage@redhat.com>
liewegas
force-pushed
the
wip-bluestore-reshard
branch
from
29a42ef
to
39bbac0
Compare
ifed01
approved these changes
Feb 3, 2017
src/os/bluestore/BlueStore.cc
Outdated
| // we are small; combine with a neighbor | ||
| if (p == shards.begin() && endoff == OBJECT_MAX_SIZE) { | ||
| // we are an only shard | ||
| request_reshard(0, OBJECT_MAX_SIZE); | ||
| return; |
There was a problem hiding this comment.
We'll exist the loop at this iteration and return the function anyway hence just a noice
src/os/bluestore/BlueStore.cc
Outdated
| } else if (p == shards.begin()) { | ||
| // combine with next shard | ||
| request_reshard(p->offset, endoff + 1); | ||
| } else if (endoff == OBJECT_MAX_SIZE) { | ||
| // combine with previous shard | ||
| request_reshard(prev_p->offset, endoff); | ||
| return; |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Reshard only a subset of the extent map. If a single extent gets big, we will reshard only
it. If one gets small, we will combine with the smaller of our two neighbors.