Re: [thredds] "Too Many Open Files" Error. Dataset too big?

• To: "thredds@xxxxxxxxxxxxxxxx" <thredds@xxxxxxxxxxxxxxxx>
• Subject: Re: [thredds] "Too Many Open Files" Error. Dataset too big?
• From: Kevin Manross <manross@xxxxxxxx>
• Date: Thu, 24 Oct 2013 15:15:19 -0600

-kevin.


On 10/24/13 3:09 PM, Michael McDonald wrote:

Kevin,
We needed to add a script to "/etc/profile.d" that increases the stock
CentOS/RHEL 6.x number of open files. Change $USER == "tomcat" to
whatever you run tomcat/apache as (see below). Also assumes that the
user is running the sh/bash shell.

# cat /etc/profile.d/thredds.sh
export TOMCAT_HOME="/usr/share/tomcat6"
if [ $USER == "tomcat" ]; then ulimit -n 1000000; fi


Our THREDDS servers providing only OPENDAP requests are averaging 40k
open files. Our THREDDS servers providing only NCSS requests are
averaging 15k open files. We have a lot of files in some of our hycom
datasets.

/mike

On Thu, Oct 24, 2013 at 1:50 PM, Kevin Manross <manross@xxxxxxxx> wrote:

Hi there,

I am trying to aggregate netcdf files for some rather large datasets with
tomcat6.0.36/TDS4.19 - 20131008.1358 and am running into a "Too many open
files" error.

My catalog scan is:

   <dataset name="ASR 30 km 2d surface analysis"
            ID="ds631.0_anl_2d-AGG"
            urlPath="ds631.0/9/best"
            harvest="true">
      <serviceName>all</serviceName>
      <netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2";>
        <aggregation dimName="time" type="Union">
          <scan location="/glade/p/rda/data/ds631.0/asr30km.anl.2d/"
suffix=".nc" subdirs="false"/>
        </aggregation>
      </netcdf>
    </dataset>

which contains about 4020 files and totals 470G.

The message that the server sends when trying to access the OPeNDAP service
is:


Error {
     code = 404;
     message =
"/glade/p/rda/data/ds631.0/asr30km.anl.2d/asr30km.anl.2d.20010619.nc (Too
many open files)";
};



When trying the CdmRemote service catalina.out repeatedly reads:

Oct 24, 2013 11:20:39 AM org.apache.jk.common.ChannelSocket
acceptConnections
WARNING: Exception executing accept
java.net.SocketException: Too many open files
     at java.net.PlainSocketImpl.socketAccept(Native Method)
     at
java.net.AbstractPlainSocketImpl.accept(AbstractPlainSocketImpl.java:375)
     at java.net.ServerSocket.implAccept(ServerSocket.java:478)
     at java.net.ServerSocket.accept(ServerSocket.java:446)
     at org.apache.jk.common.ChannelSocket.accept(ChannelSocket.java:311)
     at
org.apache.jk.common.ChannelSocket.acceptConnections(ChannelSocket.java:668)
     at
org.apache.jk.common.ChannelSocket$SocketAcceptor.runIt(ChannelSocket.java:879)
     at
org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:690)
     at java.lang.Thread.run(Thread.java:679)


The cache settings in my threddsConfig.xml are the default settings.

Have I overlooked something?

-kevin.


--
Kevin Manross
NCAR/CISL/Data Support Section
Phone: (303)-497-1218
Email:manross@xxxxxxxx
Web:http://rda.ucar.edu

_______________________________________________
thredds mailing list
thredds@xxxxxxxxxxxxxxxx
For list information or to unsubscribe,  visit:
http://www.unidata.ucar.edu/mailing_lists/

--
Kevin Manross
NCAR/CISL/Data Support Section
Phone: (303)-497-1218
Email:manross@xxxxxxxx <mailto:manross@xxxxxxxx>
Web:http://rda.ucar.edu

• Follow-Ups:
  • Re: [thredds] "Too Many Open Files" Error. Dataset too big?
    • From: John Caron

• References:
  • [thredds] "Too Many Open Files" Error. Dataset too big?
    • From: Kevin Manross
  • Re: [thredds] "Too Many Open Files" Error. Dataset too big?
    • From: Michael McDonald