This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Spike in High Response Time on Node

I am trying to investigate an issue on random high response times upwards to 400+ ms. The response time ranges in the spike so its not already that high but over 100 in the spikes. They are very random so no set timing on this. The path between node and solarwinds server appear to be clean with no errors. Has anyone experienced high spike in response times to a network switch and how to isolate where the problem lies?

  • Usually high response time is caused by high utilisation on the link to that device. Other times it could be related to high cpu spikes on the device itself.

    Does syslog show anything on the device? How's the link utilisation when response time is high?

  • I've seen backup jobs cause nodes to appear to be down or to have high response times.  The nodes would not really be down and there was not a network problem, but nodes were too busy to respond to ICMP traffic.  Packet captures from your poller, the polled nodes, and perhaps somewhere in between can be used to determine whether ping responses are leaving the polled nodes, whether pings even left the poller, if there were delays in responding, etc.

  • You're right,I forgot about that. As well as that the spike in latency can also be caused if the ICMP transverses a firewall or something like that. Netpath would help determine where the slowness occurs.