bahman/scylla
1
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

database: add a comment explaining the choice of function in CF stop

Browse Source 
We have recently commited a fix to a broken streaming bug that involved
reverting column_family::stop() back to calling the custom seal functions
explicitly for both memtables and streaming memtables.

We here add a comment to explain why that had to be done.

Signed-off-by: Glauber Costa <glauber@scylladb.com>
Message-Id: <fe94b5883e9c29adc7fc9ee9f498894c057e7b64.1464293167.git.glauber@scylladb.com>
This commit is contained in:
Glauber Costa
2016-05-26 16:06:10 -04:00
committed by Avi Kivity
parent 8e124b31aa
commit 30d54cef38
1 changed files with 6 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
database.cc
View File

@@ -771,6 +771,12 @@ column_family::start() {
future<>
column_family::stop() {
// Please note that in here, we shouldn't use the implicit seal function in each memtable's
// list. The reason is that for the streaming memtables, the memtable_list's seal function does
// not guarantee anything to be immediately flushed, and will set a timer instead (so we can
// coalesce writes). During stop, we need to force flushing behavior so we call
// seal_active_streaming_memtable() instead. That problem does not exist for memtables and we
// could call _memtables->seal_active_memtable() here. We don't, for consistency with streaming.
seal_active_memtable();
seal_active_streaming_memtable();
return _compaction_manager.remove(this).then([this] {