Troubleshooting High Latency
From Ubiquity Server Wiki
High latency / pings can occur for a wide variety of reasons.
- High load on the client's computer
- Latency in reaching the ISP's network
- Latency at the end user's ISP's upstream/peering point
- Latency in that end user's NSP before reaching us
- Latency at that NSP and our NSP's peering points
- Latency within our NSP's network
- Latency within our network
- High load on your server causing high latency
Fun huh? As you can imagine, the first thing we need to do is figure out at what place high latency is occurring, and what can be done. To do this, we want to run a traceroute inbound, and outbound, from your server.
After you've run a traceroute, you're looking at what point a big jump in latency occurs over the route, as well as whether it occurs going both directions, and why. Sometimes this is very simple in presenting itself (ie. the end user's ping to their own router is 500ms). Sometimes the only fix is for someone to call their home ISP and ask them to sort things out. Sometimes it's more complex, such as a peering point jumping in latency in one direction, but not the other.
In the event that the traceroute does not point definitively to something locally, server-related, or ISP-related - you will want to open a support request with our staff including both the inbound and outbound traceroute, and we will look into what can be done, either directly or by contacting the support department of our upstream providers if necessary.
