What's the meaning of "setMaxNumColumnsPerPDU" in SNMP4J?

SNMP4J
#1

I know that the default MaxNumColumnsPerPDU is 10. Now I can not get the query result if I use the default config and I have query more that 5 oids. So I try to call the method setMaxNumColumnsPerPDU(5),now I can get the result but I find that result is wrong.So I try to call the method setMaxNumColumnsPerPDU(1) and I can get the correct result.
If I always set maxNumColumnsPerPDU with value 1, can I do that?Is there any side effect？I believe this will ensure the correctness of the data.

                 Target target = SnmpComunication.getSnmpTarget(snmp, snmpTarget);
		BER.setCheckSequenceLength(true);
		TableUtils tableUtils = new TableUtils(snmp, new DefaultPDUFactory(PDU.GETBULK));
		OID[] columns = new OID[oids.size()];
		for (int i = 0; i < oids.size(); i++) {
			columns[i] = new OID(oids.get(i));
		} 
		 tableUtils.setMaxNumColumnsPerPDU(1);
                List<TableEvent> events = tableUtils.getTable(target, columns, null, null);

image
Here is the correct result:
image

#2

If such a response is received, then the agent has a lexicographic ordering bug.

You can then safely reduce the max columns per PDU to 1 but the performance will likely to drop because the agent cannot optimize row data retrieval anymore.
But maybe the agent is that buggy that it doesn’t optimize row retrieval at all, then it doesn’t make any difference.

#3

Hi,what’s the meaning of "optimize row data retrieval "?Can you show me a example,please?