Hi,
The problem seems to be solved after upgrading jetty to version 9.2
Thanks,
Dhiraj Prajapati
On 24 Jun 2014 16:37, "Simone Bordet" wrote:
> Hi,
>
> On Tue, Jun 24, 2014 at 12:52 PM, dhiraj prajapati
> wrote:
> > Hi,
> > I used jstack to take the thread dumps.
> > PFA the thread dumps taken
Hi,
On Tue, Jun 24, 2014 at 12:52 PM, dhiraj prajapati wrote:
> Hi,
> I used jstack to take the thread dumps.
> PFA the thread dumps taken at intervals of 5 minutes.(order is td2014062301,
> then td2014062302 and then td2014062303 )
> JDK version is jdk1.7.0_45
> Jetty version is jetty-all-9.1.2.
Hi,
On Tue, Jun 24, 2014 at 11:29 AM, dhiraj prajapati wrote:
> I took 3 thread dumps after 5 minute gaps. The number of those threads was
> constant. And all were in RUNNABLE state. Please note that there was very
> very low traffic on the application.
You don't say what tool you used to take t
I took 3 thread dumps after 5 minute gaps. The number of those threads was
constant. And all were in RUNNABLE state. Please note that there was very
very low traffic on the application.
Regards,
Dhiraj
On Tue, Jun 24, 2014 at 2:53 PM, Jan Bartel wrote:
> Dhiraj,
>
> 500 threads in the pool lyi
Dhiraj,
500 threads in the pool lying idle is not necessarily a problem - you
should check your connector configuration. Also do several thread
dumps a minute or so apart to check on the thread activity. If there's
no traffic or very little traffic, then your threads may all indeed
mostly be idle,
Thanks Thomas.
But It is the native memory that is being eaten up and not the heap memory.
So I don't think Memory Analyzer will be of any help here. I am just
concerned about the sheer number of RUNNABLE threads. It was more than 500.
Can we be sure that they are just lying idle, because the threa
Hi Dhiraj,
then use jmap to create a heap dump and browse that dump by using a tool
like Eclipse Memory Analyzer, jhat or a Java Profiler that can read heap
dumps. There you'll find what's occupying your memory. The idle thread
you postet below is no harm as Jan already stated.
Cheers,
Thoma
Also, the thread dump shows many such idle threads in RUNNABLE state (more
than 500). Isn't that an issue?
Regards,
Dhiraj
On Tue, Jun 24, 2014 at 2:33 PM, dhiraj prajapati
wrote:
> Hi,
> But the application is using a lot of native memory and it ends up using a
> lot of swap space after which
Hi,
But the application is using a lot of native memory and it ends up using a
lot of swap space after which I am forced to restart the application.
Regards,
Dhiraj
On Tue, Jun 24, 2014 at 1:20 PM, Jan Bartel wrote:
> Dhiraj,
>
> That thread dump just shows an idle thread. In past versions of
Dhiraj,
That thread dump just shows an idle thread. In past versions of jetty,
there were issues with jdk epoll bugs, but they manifested themselves
as cpu spin. So if you're not seeing undue cpu activity then this is
not a problem.
Jan
On 23 June 2014 14:57, dhiraj prajapati wrote:
> I am usin
I am using Jetty 9.1.2 in my application. The application uses huge
native memory space. In the thread dump, I see a lot of instances
like:
"qtp2042324703-524-selector-ServerConnectorManager@5c95306c/7" prio=10
tid=0x7f4c1985a000 nid=0xb7bb runnable [0x7f4c0ef6e000]
java.lang.Thread.Sta
11 matches
Mail list logo