Files
vpp-snmp-agent/vppapi.py
Pim van Pelt 95d96d5e61 bugfix: add a control_ping() before each update
If VPP were to disconnect either the Stats Segment or the API endpoint,
for example if it crashes and restarts, vpp-snmp-agent will not detect
this. In such a situation, it will hold on to the stale stats and no
longer receive interface updates.

Before each run, send a control_ping() API request, and if that were to
fail (for example with Broken Pipe, or Connection Refused), disconnect
both API and Stats (in the vpp.disconnect() call, also invalidate the interface
and LCP cache), and then fail the update. The Agent runner will then retry
once per second until the connection (and control_ping()) succeeds.

TESTED:
- Start vpp-snmp-agent, it connects and starts up per normal.
- Exit / Kill vpp
- Upon the next update(), the control_ping() call will fail, causing the
  agent to disconnect
- The agent will now loop:
[ERROR   ]      agentx.agent - update         : VPP API: [Errno 1] Sendall error: BrokenPipeError(32, 'Broken pipe'), retrying
[WARNING ]      agentx.agent - run            : Update failed, last successful update was 1673345631.7658572
[INFO    ]     agentx.vppapi - connect        : Connecting to VPP
[ERROR   ]      agentx.agent - update         : VPP API: Not connected, api definitions not available, retrying

- Start VPP again, when its API endpoint is ready:
[INFO    ]     agentx.vppapi - connect        : Connecting to VPP
[INFO    ]     agentx.vppapi - connect        : VPP version is 23.02-rc0~199-gcfaf44020
[INFO    ]     agentx.vppapi - connect        : Enabling VPP API interface events
[DEBUG   ]      agentx.agent - update         : VPP API: control_ping_reply(_0=24, context=12, retval=0, client_index=0, vpe_pid=705326)
[INFO    ]     agentx.vppapi - get_ifaces     : Requesting interfaces from VPP API
[INFO    ]     agentx.vppapi - get_lcp        : Requesting LCPs from VPP API

- The agent resumes where it left off
2023-01-10 11:24:44 +01:00

145 lines
4.3 KiB
Python

"""
The functions in this file interact with the VPP API to retrieve certain
interface metadata.
"""
from vpp_papi import VPPApiClient, VPPApiJSONFiles
import os
import fnmatch
import logging
import socket
class NullHandler(logging.Handler):
def emit(self, record):
pass
logger = logging.getLogger("agentx.vppapi")
logger.addHandler(NullHandler())
class VPPApi:
def __init__(self, address="/run/vpp/api.sock", clientname="vppapi-client"):
self.address = address
self.connected = False
self.clientname = clientname
self.vpp = None
self.iface_dict = None
self.lcp_dict = None
def _sw_interface_event(self, event):
# NOTE(pim): this callback runs in a background thread, so we just clear the
# cached interfaces and LCPs here, subsequent call to get_ifaces() or get_lcp()
# will refresh them in the main thread.
logger.info(f"Clearing iface and LCP cache due to interface event")
self.iface_dict = None
self.lcp_dict = None
def _event_callback(self, msg_type_name, msg_type):
logger.debug(f"Received callback: {msg_type_name} => {msg_type}")
if msg_type_name == "sw_interface_event":
self._sw_interface_event(msg_type)
else:
logger.warning(f"Ignoring unkonwn event: {msg_type_name} => {msg_type}")
def connect(self):
if self.connected:
return True
vpp_json_dir = VPPApiJSONFiles.find_api_dir([])
vpp_jsonfiles = VPPApiJSONFiles.find_api_files(api_dir=vpp_json_dir)
if not vpp_jsonfiles:
logger.error("no json api files found")
return False
self.vpp = VPPApiClient(apifiles=vpp_jsonfiles, server_address=self.address)
self.vpp.register_event_callback(self._event_callback)
try:
logger.info("Connecting to VPP")
self.vpp.connect(self.clientname)
except:
return False
v = self.vpp.api.show_version()
logger.info("VPP version is %s" % v.version)
logger.info("Enabling VPP API interface events")
r = self.vpp.api.want_interface_events(enable_disable=True)
if r.retval != 0:
logger.error("Could not enable VPP API interface events, disconnecting")
self.disconnect()
return False
self.connected = True
return True
def disconnect(self):
if not self.connected:
return True
self.vpp.disconnect()
self.iface_dict = None
self.lcp_dict = None
self.connected = False
return True
def get_ifaces(self):
ret = {}
if not self.connected and not self.connect():
logger.warning("Can't connect to VPP API")
return ret
if type(self.iface_dict) is dict:
logger.debug("Returning cached interfaces")
return self.iface_dict
ret = {}
try:
logger.info("Requesting interfaces from VPP API")
iface_list = self.vpp.api.sw_interface_dump()
except Exception as e:
logger.error("VPP API communication error, disconnecting", e)
self.disconnect()
return ret
if not iface_list:
logger.error("Can't get interface list, disconnecting")
self.disconnect()
return ret
for iface in iface_list:
ret[iface.interface_name] = iface
self.iface_dict = ret
logger.debug(f"Caching interfaces: {ret}")
return self.iface_dict
def get_lcp(self):
ret = {}
if not self.connected and not self.connect():
logger.warning("Can't connect to VPP API")
return ret
if type(self.lcp_dict) is dict:
logger.debug("Returning cached LCPs")
return self.lcp_dict
try:
logger.info("Requesting LCPs from VPP API")
lcp_list = self.vpp.api.lcp_itf_pair_get()
except Exception as e:
logger.error("VPP communication error, disconnecting", e)
self.disconnect()
return ret
if not lcp_list:
logger.error("Can't get LCP list")
return ret
for lcp in lcp_list[1]:
ret[lcp.host_if_name] = lcp
self.lcp_dict = ret
logger.debug(f"Caching LCPs: {ret}")
return self.lcp_dict