hcq/MaxScale
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

MXS-1203: Better handling of batch queries

Browse Source 
When batched queries are done through readwritesplit, it will now handle
them one by one. This allows batched queries to be used with
readwritesplit but it does impose a performance penalty when compared to
direct execution on the backend.
This commit is contained in:
Markus Mäkelä
2017-03-31 14:00:20 +03:00
parent a8b42d24b7
commit 66cf571412
9 changed files with 266 additions and 108 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
server/modules/routing/readwritesplit/rwsplit_session_cmd.c
View File

@ -321,16 +321,11 @@ GWBUF *sescmd_cursor_clone_querybuf(sescmd_cursor_t *scur)
bool execute_sescmd_history(backend_ref_t *bref)
{
bool succp = true;
sescmd_cursor_t *scur;
if (bref == NULL)
{
MXS_ERROR("[%s] Error: NULL parameter.", __FUNCTION__);
return false;
}
ss_dassert(bref);
CHK_BACKEND_REF(bref);
bool succp = true;
scur = &bref->bref_sescmd_cur;
sescmd_cursor_t *scur = &bref->bref_sescmd_cur;
CHK_SESCMD_CUR(scur);
if (!sescmd_cursor_history_empty(scur))