SNMP Timeout/Error

[kribbas]

After tuning the poller settings, changing the php script server settings from 1 --> 5, I started getting SNMP Error on several hosts.
After reverting back to 1 in the php script server settings I'm still getting the error.

So some hosts are working fine, then some are getting SNMP Timeout causing the poller to exceed and exit.
When I try to snmpwalk to these nodes manually from the command line it works and sometimes it does not work,

Can I get cacti to try again on the nodes before stating they are down? since it seems to be working a bit on and off?

Finally how can I revert the changes, since it was working perfectly fine before with these hosts and nothing has changed accept the change in php script server settings from 1-->5 then back to 1.

Info below:

Cacti Version 0.8.8h
Cacti OS unix
SNMP Version NET-SNMP version: 5.5
RRDTool Version RRDTool 1.3.x
Hosts 691
Graphs 25761
Data Sources Script/Command: 1365
SNMP: 1893
SNMP Query: 17012
Script Query: 856
Script Query - Script Server: 7727
Total: 28853
Poller Information
Interval 300
Type SPINE 0.8.8h Copyright 2002-2016 by The Cacti Group
Items Action[0]: 29354
Action[1]: 2771
Action[2]: 5420
Total: 37545
Concurrent Processes 2
Max Threads 12
PHP Servers 1
Script Timeout 25
Max OID 15
Last Run Statistics Time:147.7859 Method:spine Processes:2 Threads:12 Hosts:564 HostsPerProcess:282 DataSources:37468 RRDsProcessed:20577