This is a discussion on Re: 1 second delay within the Bind Users forums, part of the DNS and Related Forums category; In article <d7acih$1pkq$1@sf1.isc.org>, Wayne Walker <wwalker@bybent.com> wrote: > I ...
|
|||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
|
|||
|
In article <d7acih$1pkq$1@sf1.isc.org>,
Wayne Walker <wwalker@bybent.com> wrote: > I have a number of servers that need fast lookups. I set up BIND named > on each one weeks ago, and changed /etc/resolv.conf to have 127.0.0.1 as > the first nameserver line. > > They were running fine with ~500 ms lookups the first time for any host > in our domain. Subsequent lookups for the same host were always in the > 7-15 ms range! :) > > Starting yesterday, at least 2 of the servers are now getting an > addition full second of delay talking to the local DNS servers. > > If I run "time host db", I get 0.020 when pointed to an external server, > I get 1.024 or 1.025 when pointed to 127.0.0.1. > > Any ideas what to do to resolve this? Sounds like you don't have enough memory in your servers. BIND is competing against the applications running on the servers for the available RAM. When too much of BIND's cache gets paged out, its performance goes to hell. -- Barry Margolin, barmar@alum.mit.edu Arlington, MA *** PLEASE post questions in newsgroups, not directly to me *** |
![]() |
| Thread Tools | |
| Display Modes | |
|
|