Asynchronous getNext errors out when going out of table

I am using one of the examples for GetNext operation for an SNMPWalk of the tree. I am using the asynchronous variant to collect the OIDs -

# GETNEXT Command Generator
from pysnmp.entity.rfc3413.oneliner import cmdgen
from pysnmp.proto import rfc1902

# ( ( authData, transportTarget, varNames ), ... )
targets = (
    # 1-st target (SNMPv1)
    ( cmdgen.CommunityData('test-agent-1', 'public'),
      cmdgen.UdpTransportTarget(('localhost', 161)),
      (rfc1902.ObjectName((1,3,6,1,2,1)), rfc1902.ObjectName((1,3,6,1,3,1)))),
    # 2-nd target (SNMPv2c)
    ( cmdgen.CommunityData('test-agent-2', 'public', 1),
      cmdgen.UdpTransportTarget(('localhost', 161)),
      (rfc1902.ObjectName((1,3,6,1,2,1,2)),) ),
    )

def cbFun(
    sendRequestHandle, errorIndication, errorStatus, errorIndex,
    varBindTable, (varBindHead, authData, transportTarget)
    ):
    if errorIndication:
        print 'SNMP engine error', errorIndication
        return 1
    if errorStatus:
        print 'SNMP error %s at %s' % (errorStatus, errorIndex)
        return 1
    varBindTableRow = varBindTable[-1]
    for idx in range(len(varBindTableRow)):
        name, val = varBindTableRow[idx]
        if val is not None and varBindHead[idx].isPrefixOf(name):
            # still in table
            break
    else:
        print 'went out of table at %s' % (name, )
        return

    for varBindRow in varBindTable:
        for oid, val in varBindRow:
            if val is None:
                print oid.prettyPrint()
            else:
                print '%s = %s' % (oid.prettyPrint(), val.prettyPrint())

    return 1 # continue table retrieval

cmdGen  = cmdgen.CommandGenerator()

for authData, transportTarget, varNames in targets:
    cmdGen.asyncNextCmd(
        authData, transportTarget, varNames,
        # User-space callback function and its context
        (cbFun, (varNames, authData, transportTarget))
        )

cmdGen.snmpEngine.transportDispatcher.runDispatcher()

I get the OIDs I need; however when it goes out the table and returns from cbFun, the Dispatcher in the last line throws an error, which I am not able to resolve, the output looks something like -

... (some 1.3.6.1.2.1.* stuff)
1.3.6.1.2.1.2.2.1.22.1 = 0.0
1.3.6.1.2.1.2.2.1.22.2 = 0.0
1.3.6.1.2.1.2.2.1.22.3 = 0.0
went out of table at (1, 3, 6, 1, 2, 1, 3, 1, 1, 1, 2, 1, 212, 201, 49, 186)
Traceback (most recent call last):
  File "asyncsnmpwalk.py", line 55, in <module>
    cmdGen.snmpEngine.transportDispatcher.runDispatcher()
  File "/var/lib/python-support/python2.5/pysnmp/v4/carrier/asynsock/dispatch.py", line 61, in runDispatcher
    self.handleTimerTick(time())
  File "/var/lib/python-support/python2.5/pysnmp/v4/carrier/base.py", line 81, in handleTimerTick
    self.__timerCbFun(timeNow)
  File "/var/lib/python-support/python2.5/pysnmp/v4/entity/engine.py", line 60, in __receiveTimerTickCbFun
    self.msgAndPduDsp.receiveTimerTick(self, timeNow)
  File "/var/lib/python-support/python2.5/pysnmp/v4/proto/rfc3412.py", line 491, in receiveTimerTick
    self.__cacheExpire(snmpEngine, self.__expireRequest)
  File "/var/lib/python-support/python2.5/pysnmp/v4/proto/rfc3412.py", line 56, in __cacheExpire
    if cbFun(snmpEngine, cachedParams):
  File "/var/lib/python-support/python2.5/pysnmp/v4/proto/rfc3412.py", line 486, in __expireRequest
    cbCtx
  File "/var/lib/python-support/python2.5/pysnmp/v4/entity/rfc3413/cmdgen.py", line 77, in processResponsePdu
    (self.processResponsePdu, (cbFun, cbCtx))
  File "/var/lib/python-support/python2.5/pysnmp/v4/entity/rfc3413/cmdgen.py", line 164, in _sendPdu
    (processResponsePdu, float(timeout)/100 + time.time(), cbCtx)
  File "/var/lib/python-support/python2.5/pysnmp/v4/proto/rfc3412.py", line 161, in sendPdu
    sendPduHandle
  File "/var/lib/python-support/python2.5/pysnmp/v4/proto/mpmod/rfc2576.py", line 89, in prepareOutgoingMessage
    scopedPDU
  File "/var/lib/python-support/python2.5/pysnmp/v4/proto/secmod/rfc2576.py", line 78, in genera开发者_运维技巧teRequestMsg
    errorIndication = 'unknownCommunityName'
pysnmp.proto.error.StatusInformation: {'errorIndication': 'unknownCommunityName'}

---

changing CommunityData string did the trick (I don't know why)

BEFORE

cmdgen.CommunityData('test-agent-1', 'public'), 

AFTER

cmdgen.CommunityData('test-agent', 'public'),

---

That seems to happen on request retry. So it may not be relevant to going out of table.

I'd suggest upgrading your pysnmp installation from CVS at SourceForge in hope that it's a already fixed bug. Otherwise, please report to pysnmp-users@lists.sourceforge.net