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

Fix response counter discrepancy on retried read

Browse Source 
When a read was successfully retried, the original expected response was
not decremented from the reponse counter. This caused one extra response
to be expected for successfully retried reads which caused a hang after a
retried read.
This commit is contained in:
Markus Mäkelä
2017-08-17 08:14:29 +03:00
parent fb9b027842
commit d3a77b64c0
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
server/modules/routing/readwritesplit/readwritesplit.cc
View File

@ -396,6 +396,9 @@ static bool handle_error_new_connection(RWSplit *inst,
if (backend->is_waiting_result()) if (backend->is_waiting_result())
{ {
ss_dassert(myrses->expected_responses > 0);
myrses->expected_responses--;
/** /**
* A query was sent through the backend and it is waiting for a reply. * A query was sent through the backend and it is waiting for a reply.
* Try to reroute the statement to a working server or send an error * Try to reroute the statement to a working server or send an error
@ -413,7 +416,6 @@ static bool handle_error_new_connection(RWSplit *inst,
* and decrement the expected response count. * and decrement the expected response count.
*/ */
gwbuf_free(stored); gwbuf_free(stored);
myrses->expected_responses--;
if (backend->session_command_count() == 0) if (backend->session_command_count() == 0)
{ {