DNS response times high in MWG process but not OS causing impact to clients
We are seeing extremely high DNS response times on the MWG 184.108.40.206.0 and this is impacting users onsite. We are not seeing the same response time in nslookups on the CLI of the proxy, but any client connecting through the mwg is getting massive timeouts.
The only indicator that we are having a problem is the Average Processing Time in ms for DNS Lookup under the Performance graphs in the GUI vault immediately to the 20k and above. We are unable to see this activity in any log file in the system and cannot seem to find out why this is happening. The resolution for this is to do a service mwg restart, which brings response times back under 100ms.
We need to know what could be causing this in the mwg processes and how to resolve it. As a workaround, we are trying to figure out how to read that number and restart the service automatically after a set amount of failures. This is not going well.