DeadlineExceededError连接到Google Cloud SQL

Question

We have a google app engine application connected to GCSQL, and we are having a lot of DeadlineExceededError problems while trying to connect the database. The error seems to be produced by a big amount of users opening the application in a short period of time. When we look up the opening connectios at the dashboard during this time, there aren't more than 45 opened connections at the same time and it fails on a big percentage of querys (reading, writing and connecting), taking too much time to do it. As a result of this spent time, the service shuts down and we receive the error.

An error example:

Traceback (most recent call last):
  File "/python27_runtime/python27_lib/versions/1/google/appengine/runtime/wsgi.py", line 223, in Handle
    result = handler(dict(self._environ), self._StartResponse)
  File "/python27_runtime/python27_lib/versions/third_party/webapp2-2.3/webapp2.py", line 1505, in __call__
    rv = self.router.dispatch(request, response)
  File "/python27_runtime/python27_lib/versions/third_party/webapp2-2.3/webapp2.py", line 1253, in default_dispatcher
    return route.handler_adapter(request, response)
  File "/python27_runtime/python27_lib/versions/third_party/webapp2-2.3/webapp2.py", line 1077, in __call__
    return handler.dispatch()
  File "/python27_runtime/python27_lib/versions/third_party/webapp2-2.3/webapp2.py", line 545, in dispatch
    return method(*args, **kwargs)
  File "/base/data/home/apps/s~ono-hat-prod/1.365115032772094991/onohatpreprod.py", line 5541, in post
    cursor.execute("SELECT %s FROM actuaciones WHERE numero = '%s' " % (rc,idHat))
  File "/python27_runtime/python27_lib/versions/1/google/storage/speckle/python/api/rdbms.py", line 499, in execute
    self._DoExec(request)
  File "/python27_runtime/python27_lib/versions/1/google/storage/speckle/python/api/rdbms.py", line 375, in _DoExec
    response = self._conn.MakeRequest('Exec', request)
  File "/python27_runtime/python27_lib/versions/1/google/storage/speckle/python/api/rdbms.py", line 873, in MakeRequest
    response = self._MakeRetriableRequest(stub_method, request)
  File "/python27_runtime/python27_lib/versions/1/google/storage/speckle/python/api/rdbms.py", line 897, in _MakeRetriableRequest
    response = self.MakeRequestImpl(stub_method, request)
  File "/python27_runtime/python27_lib/versions/1/google/storage/speckle/python/api/rdbms_apiproxy.py", line 67, in MakeRequestImpl
    except apiproxy_errors.ApplicationError, e:
 DeadlineExceededError

Piece of code until the error occurs:

class detalleActuacion(webapp2.RequestHandler):
    def post(self):
        idHat = self.request.get('id')
        logging.info(idHat)
        num_conexiones = 0
        num_cierres = 0
        conn = rdbms.connect(instance=_INSTANCE_NAME, database='Actuaciones')
        num_conexiones = num_conexiones + 1
        cursor = conn.cursor()

        dict_act = {}
        request_cad = 'solucionado,duracion,numero,fecha_inicio_ventana,fecha_fin_ventana,'
        request_cad = request_cad+'tipo,titulo,tipo_cliente,ubicacion,zona_regional,'
        request_cad = request_cad+'zona_local,ciudad_ticket,zona_regional_subticket,zona_local_subticket,'
        request_cad = request_cad+'zona_local_ticket,skill,num_recursos,num_actuaciones,ubi_emplaz,'
        request_cad = request_cad+'ubi_provincia,ubi_lat,ubi_long,ubi_via,ubi_calle,ubi_numero,ubi_adic,'
        request_cad = request_cad+'ubi_CP,miga_terminal,miga_tlf,miga_dir,severidad,prioridad,serv_CATV,'
        request_cad = request_cad+'serv_datos,serv_DTV,serv_empresa,serv_telefonia,instr,num_rep,'
        rc = request_cad+'fecha_creac,fecha_creac_tp,fecha_sol,fecha_sol_tp,fecha_cierre,fecha_cierre_tp,'
        rc = rc+'fecha_inicio,fecha_fin,descr_adic,descr_log,grupo_pos,accion,cola,estado,estado_ticket,'
        rc=rc+'doc_adj,detalle_equipos,tarea,subtarea,subzona,TIR,elem_red,descripcion,num_afectados,'
        rc=rc+'repeticion,fecha_repeticion,dias_repeticion,dias_semana,fin_repeticion,cl_afectados,'
        rc=rc+'plan_tp,marca_seguimiento,tipificacion,miga,zona,bandeja,alerta,descr_log1,descr_log2,descr_log3,descr_log4,descr_log5,descr_log6,descr_log7,descr_log8,descr_log9,'
        rc=rc+'elem_red_2,estado_CRM,tecnico,tecnico2,tecnico3,tecnico4,tecnico5,num_asignaciones,ultima_modificacion,tiempo_proceso,fecha_inicio_real,fecha_fin_real'
        rc=rc+',observaciones1,observaciones2,observaciones3,observaciones4,observaciones5,observaciones6,observaciones7,observaciones8,observaciones9,observaciones10'
        cursor = conn.cursor()
        cursor.execute("SELECT %s FROM actuaciones WHERE numero = '%s' " % (rc,idHat))

Thank you in advance for your assistance.

Answer 1

If there are 45 concurrent requests, there may be many more open connections. See What are the connection limits for Google Cloud SQL from App Engine, and how to best reuse DB connections?

You could also try increasing the instance size: more RAM may reduce the latency of each request and so reduce the number of connections.