packages feed

ekg 0.4.0.3 → 0.4.0.4

raw patch · 4 files changed

+14/−7 lines, 4 filesPVP ok

version bump matches the API change (PVP)

API changes (from Hackage documentation)

Files

CHANGES.md view
@@ -1,3 +1,10 @@+## 0.4.0.4 (2014-10-20)++ * Fix bug where `forkServerWith` would register GC metrics, despite+   the docs saying that it doesn't.++ * Fix example code that broke in the package split.+ ## 0.4.0.3 (2014-09-30)   * Support text-1.2.
System/Remote/Monitoring.hs view
@@ -195,6 +195,7 @@            -> IO Server forkServer host port = do     store <- Metrics.newStore+    Metrics.registerGcMetrics store     forkServerWith store host port  -- | Start an HTTP server in a new thread.  The server replies to GET@@ -221,7 +222,6 @@                -> Int            -- ^ Port to listen on (e.g. 8000)                -> IO Server forkServerWith store host port = do-    Metrics.registerGcMetrics store     Metrics.registerCounter "ekg.server_timestamp_ms" getTimeMs store     tid <- forkIO $ startServer store host port     return $! Server tid store
ekg.cabal view
@@ -1,5 +1,5 @@ name:                ekg-version:             0.4.0.3+version:             0.4.0.4 synopsis:            Remote monitoring of processes description:   This library lets you remotely monitor a running process over HTTP.
examples/Basic.hs view
@@ -8,9 +8,9 @@ import Control.Exception import Data.List import Data.Time.Clock.POSIX (getPOSIXTime)-import qualified System.Remote.Event as Event-import qualified System.Remote.Counter as Counter-import qualified System.Remote.Label as Label+import qualified System.Metrics.Distribution as Distribution+import qualified System.Metrics.Counter as Counter+import qualified System.Metrics.Label as Label import System.Remote.Monitoring  -- 'sum' is using a non-strict lazy fold and will blow the stack.@@ -25,11 +25,11 @@     handle <- forkServer "localhost" 8000     counter <- getCounter "iterations" handle     label <- getLabel "args" handle-    event <- getEvent "runtime" handle+    event <- getDistribution "runtime" handle     Label.set label "some text string"     let loop n = do             t <- timed $ evaluate $ mean [1..n]-            Event.add event t+            Distribution.add event t             threadDelay 2000             Counter.inc counter             loop n