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

Add verbose logging for session command failures

Browse Source 
If the routing of a session command fails due to problems with the backend
connections, a more verbose error message is logged. The added status
information in the Backend class makes tracking the original cause of the
problem a lot easier due to knowing where, when and why the connection was
closed.
This commit is contained in:
Markus Mäkelä
2019-01-28 10:56:01 +02:00
parent df9335382d
commit 24c9b62a2f
5 changed files with 116 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

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

@ -951,6 +951,7 @@ void RWSplitSession::handleError(GWBUF* errmsgbuf,
}
backend->close();
backend->set_close_reason("Master connection failed: " + extract_error(errmsgbuf));
}
else
{
@ -964,6 +965,7 @@ void RWSplitSession::handleError(GWBUF* errmsgbuf,
// Try to replay the transaction on another node
can_continue = start_trx_replay();
backend->close();
backend->set_close_reason("Read-only trx failed: " + extract_error(errmsgbuf));
if (!can_continue)
{
@ -984,6 +986,7 @@ void RWSplitSession::handleError(GWBUF* errmsgbuf,
m_otrx_state = OTRX_INACTIVE;
can_continue = start_trx_replay();
backend->close();
backend->set_close_reason("Optimistic trx failed: " + extract_error(errmsgbuf));
}
else
{
@ -1073,6 +1076,7 @@ bool RWSplitSession::handle_error_new_connection(DCB* backend_dcb, GWBUF* errmsg
* is closed, it's possible that the routing logic will pick the failed
* server as the target. */
backend->close();
backend->set_close_reason("Slave connection failed: " + extract_error(errmsg));
if (route_stored)
{