summaryrefslogtreecommitdiffstats
path: root/health/health.d/bcache.conf
diff options
context:
space:
mode:
Diffstat (limited to 'health/health.d/bcache.conf')
-rw-r--r--health/health.d/bcache.conf22
1 files changed, 22 insertions, 0 deletions
diff --git a/health/health.d/bcache.conf b/health/health.d/bcache.conf
new file mode 100644
index 000000000..f0da9ac5e
--- /dev/null
+++ b/health/health.d/bcache.conf
@@ -0,0 +1,22 @@
+
+template: bcache_cache_errors
+ on: disk.bcache_cache_read_races
+ lookup: sum -10m unaligned absolute
+ units: errors
+ every: 1m
+ warn: $this > 0
+ crit: $this > ( ($status >= $CRITICAL) ? (0) : (10) )
+ delay: down 1h multiplier 1.5 max 2h
+ info: the number of times bcache had issues using the cache, during the last 10 mins (this usually means your SSD cache is failing)
+ to: sysadmin
+
+template: bcache_cache_dirty
+ on: disk.bcache_cache_alloc
+ calc: $dirty + $metadata + $undefined
+ units: %
+ every: 1m
+ warn: $this > ( ($status >= $WARNING ) ? ( 70 ) : ( 90 ) )
+ crit: $this > ( ($status >= $CRITICAL) ? ( 90 ) : ( 95 ) )
+ delay: up 1m down 1h multiplier 1.5 max 2h
+ info: the percentage of cache space used for dirty and metadata (this usually means your SSD cache is too small)
+ to: sysadmin