cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sotirios Delimanolis <>
Subject Re: What does Cassandra use (JNI?) that triggers GCLocker Initiated GCs?
Date Thu, 21 Apr 2016 18:08:43 GMT
We're running G1 at the moment, both young and mixed collections. 

    On Thursday, April 21, 2016 11:07 AM, Jake Luciani <> wrote:

 What kind of collection? if its par new I wouldn't worry.
On Thu, Apr 21, 2016 at 2:02 PM, Sotirios Delimanolis <> wrote:

Should this be of any concern? Are the corresponding threads spending too long in this JNI
critical region and delaying GC?
I don't get that impression at all from the GC log timings. They're very reasonable.
    On Thursday, April 21, 2016 10:57 AM, Jake Luciani <> wrote:

 It's only used by the Snappy and LZ4 Compressors
On Thu, Apr 21, 2016 at 1:54 PM, Sotirios Delimanolis <> wrote:

According to this Oracle document, GCLocker Initiated GC
    is triggered when a JNI critical region was released. GC is blocked when any thread
is in the JNI Critical region.    If GC was requested during that period, that GC is invoked
after all the threads come out of the JNI critical region.

What part of Cassandra's implementation does anything with JNI?
In our GC logs, this is by far the most common reason for GC pauses.




View raw message