Mirrors/coredns
1
0
Fork 0
mirror of https://github.com/coredns/coredns.git synced 2025-10-27 08:14:18 -04:00
Code Issues Packages Projects Releases Wiki Activity

respond with REFUSED when max_concurrent is exceeded to avoid caching it (#4326)

Browse Source 
Signed-off-by: Chris O'Haver <cohaver@infoblox.com>
This commit is contained in:
Chris O'Haver
2020-12-15 08:02:15 -05:00
committed by GitHub
parent 45bf23f145
commit 9cb53487ec
3 changed files with 12 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
plugin/forward/README.md
View File

@@ -88,7 +88,7 @@ forward FROM TO... {
* `no_rec` - optional argument that sets the RecursionDesired-flag of the dns-query used in health checking to `false`.
The flag is default `true`.
* `max_concurrent` **MAX** will limit the number of concurrent queries to **MAX**. Any new query that would
raise the number of concurrent queries above the **MAX** will result in a SERVFAIL response. This
raise the number of concurrent queries above the **MAX** will result in a REFUSED response. This
response does not count as a health failure. When choosing a value for **MAX**, pick a number
at least greater than the expected *upstream query rate* * *latency* of the upstream servers.
As an upper bound for **MAX**, consider that each concurrent query will use about 2kb of memory.

2
plugin/forward/forward.go
View File

@@ -83,7 +83,7 @@ func (f *Forward) ServeDNS(ctx context.Context, w dns.ResponseWriter, r *dns.Msg
defer atomic.AddInt64(&(f.concurrent), -1)
if count > f.maxConcurrent {
MaxConcurrentRejectCount.Add(1)
return dns.RcodeServerFailure, f.ErrLimitExceeded
return dns.RcodeRefused, f.ErrLimitExceeded
}
}