coredns/plugin/cache
Tobias Schmidt dd9fc8962c plugin/cache: Fix prefetching issues (#1363)
* Improve plugin/cache metrics

* Add coredns_cache_prefetch_total metric to track number of prefetches.
* Remove unnecessary Cache.get() call which would incorrectly increment
  cache counters.
* Initialize all counters and gauges at zero.

* Allow prefetching of a single request per ttl

The original implementation didn't allow prefetching queries which are
only requested once during the duration of a TTL. The minimum amount of
queries which had to be seen was therefore capped at 2.

This change also implements a real prefetch test. The existing test was
a noop and always passed regardless of any prefetch implementation.

* Fix prefetching for items with a short TTL

The default prefetch threshold (percentage) is 10% of the lifetime of a
cache item. With the previous implementation, this disabled prefetching
for all items with a TTL < 10s (the resulting percentage would be 0, at
which point a cached item is already discarded).

This change uses a time based threshold calculation and ensures that
a prefetch is triggered at a TTL of 1 at the latest.

* Fix wrong duration reporting of cached responses

The logging and metrics plugins (among others) included the duration of
a cache prefetch in the request latency of client request. This change
fixes this wrong reporting and executes the prefetch request in a
goroutine in the background.
2018-01-17 07:35:22 +00:00
..
freq Remove the word middleware (#1067) 2017-09-14 09:36:06 +01:00
cache.go plugin/cache: Fix prefetching issues (#1363) 2018-01-17 07:35:22 +00:00
cache_test.go plugin/cache: Fix prefetching issues (#1363) 2018-01-17 07:35:22 +00:00
fuzz.go plugin/cache: Fix prefetching issues (#1363) 2018-01-17 07:35:22 +00:00
handler.go plugin/cache: Fix prefetching issues (#1363) 2018-01-17 07:35:22 +00:00
item.go plugin/cache: Fix prefetching issues (#1363) 2018-01-17 07:35:22 +00:00
minttl_test.go plugin/cache: add minttl test (#1141) 2017-10-10 18:30:14 +02:00
prefech_test.go plugin/cache: Fix prefetching issues (#1363) 2018-01-17 07:35:22 +00:00
README.md plugin/cache: Fix prefetching issues (#1363) 2018-01-17 07:35:22 +00:00
setup.go plugin/cache: Fix prefetching issues (#1363) 2018-01-17 07:35:22 +00:00
setup_test.go Remove the word middleware (#1067) 2017-09-14 09:36:06 +01:00

cache

Name

cache - enables a frontend cache.

Description

With cache enabled, all records except zone transfers and metadata records will be cached for up to 3600s. Caching is mostly useful in a scenario when fetching data from the backend (upstream, database, etc.) is expensive.

Syntax

cache [TTL] [ZONES...]
  • TTL max TTL in seconds. If not specified, the maximum TTL will be used, which is 3600 for noerror responses and 1800 for denial of existence ones. Setting a TTL of 300: cache 300 would cache records up to 300 seconds.
  • ZONES zones it should cache for. If empty, the zones from the configuration block are used.

Each element in the cache is cached according to its TTL (with TTL as the max). For the negative cache, the SOA's MinTTL value is used. A cache can contain up to 10,000 items by default. A TTL of zero is not allowed.

If you want more control:

cache [TTL] [ZONES...] {
    success CAPACITY [TTL]
    denial CAPACITY [TTL]
    prefetch AMOUNT [[DURATION] [PERCENTAGE%]]
}
  • TTL and ZONES as above.
  • success, override the settings for caching successful responses. CAPACITY indicates the maximum number of packets we cache before we start evicting (randomly). TTL overrides the cache maximum TTL.
  • denial, override the settings for caching denial of existence responses. CAPACITY indicates the maximum number of packets we cache before we start evicting (LRU). TTL overrides the cache maximum TTL. There is a third category (error) but those responses are never cached.
  • prefetch will prefetch popular items when they are about to be expunged from the cache. Popular means AMOUNT queries have been seen with no gaps of DURATION or more between them. DURATION defaults to 1m. Prefetching will happen when the TTL drops below PERCENTAGE, which defaults to 10%, or latest 1 second before TTL expiration. Values should be in the range [10%, 90%]. Note the percent sign is mandatory. PERCENTAGE is treated as an int.

Metrics

If monitoring is enabled (via the prometheus directive) then the following metrics are exported:

  • coredns_cache_size{type} - Total elements in the cache by cache type.
  • coredns_cache_capacity{type} - Total capacity of the cache by cache type.
  • coredns_cache_hits_total{type} - Counter of cache hits by cache type.
  • coredns_cache_misses_total{} - Counter of cache misses.

Cache types are either "denial" or "success".

Examples

Enable caching for all zones, but cap everything to a TTL of 10 seconds:

. {
    cache 10
    whoami
}

Proxy to Google Public DNS and only cache responses for example.org (or below).

. {
    proxy . 8.8.8.8:53
    cache example.org
}