APR or Apache, virtual hosts and multi-core tuning...

Hi All,

I noticed using Firefox with a plugin yslow that it recommends using only
4 virtual hosts instead of the 16 I have for serving images. I am using a 6
core system and want to make sure I take advantage of it using Apache APR
so I setup more than 4 virtual hosts.

I am using multiple virtual hosts for serving images since I heard that
can improve user response time where they have browsers with 4 or more
concurrent connections supported. I have a screenshots page where I have 24
thumbnails being loaded into the browser and I am trying to tune APR for
the best user response time.

So any best practice for this you all would recommend?

Thanks,
-Tony



      

Post a question and let the community help.

Kevin
http://kevincastellow.workintel.com


On Fri, May 21, 2010 at 7:03 AM, Chris Johnson
<john### @nmr.mgh.harvard.edu>wrote:

     Hi.

     I really need to talk to someone offline who has multi domain
 certificates and
 named virtual hosts configured, up and running please.  I'm using
CentOS
 4.7 and 5.4 with Apache 2.x.  Help greatly appreciated.


 

Adam Hasselbalch Hansen wrote:
 We have a setup that uses an in-house module which works not entirely

 unlike mod_vhost_alias, in that it has a single virtual host
configured, 
 and then determines stuff like domain name, docroot, etc, from the
request.
 
 We'd love to be able to use SSL in this setup, but as far as I can
see, 
 the only way to do this would be to change (i.e. hack) mod_ssl to do
the 
 certificate loading sometime around request-time, since the apache 
 server and SSL have no clue what virtual hosts they will be serving
at 
 startup.
 
 How would such a hack, if at all possible, affect stuff like
certificate 
 caching and other things?
 
 I'd love any feedback!


Anyone?

to deploy 2 virtual hosts in Tomcat 5.5.29 (no Apache) what more than
the server.xml file below and a context.xml per each host in Catalina do
you need to set up?

Where do I set path, like for "/"? Alright in the context file (see
below)?

server.xml:
<?xml version="1.0" encoding="UTF-8"?>
<Server port="8005" shutdown="SHUTDOWN">
  <GlobalNamingResources>
    <!-- Used by Manager webapp -->
    <Resource name="UserDatabase" auth="Container"
              type="org.apache.catalina.UserDatabase"
       description="User database that can be updated and saved"
           factory="org.apache.catalina.users.MemoryUserDatabaseFactory"
          pathname="conf/tomcat-users.xml" />
  </GlobalNamingResources>

  <Service name="Catalina">
    <Connector port="80" />
 <Engine name="Catalina" defaultHost="kron">
      <Realm className="org.apache.catalina.realm.UserDatabaseRealm"
             resourceName="UserDatabase" />
      <Host name="medpsy" appBase="/home/webbplatser/medpsy.no" />
      <Host name="kron" appBase="/home/webbplatser/kron.no" />
    </Engine>
  </Service>
</Server>


Fragment context settings in /Catalina/kron/ROOT.xml (similar for the
other site: 

<?xml version="1.0" encoding="UTF-8"?>
<Context
    docBase="/home/webbplatser/kron.no/ROOT"
    path="/"
    reloadable="true"
    workDir="work/Catalina/kron/_">
  <Listener className="org.apache.catalina.startup.TldConfig"/>
</Context>:

This set up doesn't serve a pages physically put in any of the ROOT
folders of the sites. I get blank pages.

GUIDES?
Anyone that can point to a guide that focuses only on the steps
necessary to configure Tomcat 5.5.x for 2 or more virtual hosts?
Preferably without involvement with Apache.

Duane,

Thanks for responding to my post.  The error message I get from using the
NameVirgualHost command is:

[Sat Jun 05 15:38:49 2010] [warn] NameVirtualHost 192.168.0.93:80 has no
VirtualHosts

Here is a copy and paste right out of my httpd.conf

Listen 192.168.0.93:80
Listen 192.168.0.200:80

NameVirtualHost 192.168.0.93:80

I had in each virtual host configured with the IP address when I received
that error.  When I tried your example using the fqdn in the
<VirtualHost fqdn.com> container, I mulitple errors like this:

[Sat Jun 05 15:51:21 2010] [warn] VirtualHost fqdn1.com:80 overlaps with
VirtualHost fqdn2.com:80, the first has precedence, perhaps you need a
NameVirtualHost directive

This is repeated for every name-based virtual host I have until I added
the fqdns to /etc/hosts.  However, I still get the error message about
NameVirtualHost 192.168.0.93 has no Virtual Hosts


Concerning your suggestion for the wild card certificate, I see there is a
multi-domain certificate offered that seems like just what I need.  I
wonder how I configure my SSLCertificateFile and SSLCertificateKeyFile
directives.  Do I simply point these directives to the same files within
the different ip-based virtual hosts?  Something like this:

<VirtualHost 192.168.0.200:443>
SSLCertificateFile "/files/server.crt"
 SSLCertificateKeyFile "/files/server.key"
</VirtualHost>

<VirtualHost 192.168.0.201:443>
SSLCertificateFile "/files/server.crt"
 SSLCertificateKeyFile "/files/server.key"
</VirtualHost>

<VirtualHost 192.168.0.202:443>
SSLCertificateFile "/files/server.crt"
 SSLCertificateKeyFile "/files/server.key"
</VirtualHost>


-----Original Message-----
From: "Duane Winner" [dua### @duanewinner.net]
Date: 06/04/2010 03:23 PM
To: use### @httpd.apache.org
Subject: Re: [us### @httpd] NameVirtualHost Directive - - - - Multiple SSL 
	virtual hosts



On Fri, Jun 4, 2010 at 6:10 PM, Jeff Shearer
<jef### @shearer-family.org> wrote:
I have an Apache 2.2 server sitting on a box with one physical address and
multiple virtual addresses.  The physical address servers my name-based
virtual hosts where as the virtual addresses server ip-based virtual hosts.
  I read about the NameVirtualHost directive and tried to implement it but
it seems it does nothing but cause error messages.  I tried to put into
the main server config near my listen directives and received an error that
there was no virtual host despite the fact that there are more than one
virtual host.

I also tried to put it just before each of the virtual host directives and
received errors.

Is this directive needed, everything seems to work fine with out it.

What are the error messages?

For your physical address serving your virtual hosts you should have:

Listen your-physical-ip:80
Listen your-physical-ip:443
NameVirtualHost your-physical-ip7:80
NameVirtualHost your physical-ip:443
...


Then for each virtual host served by physical IP, all you need is:

<VirtualHost virtual.host.fqdn:80>
  ServerName virtual.host.fqdn
   ...
</VirtualHost>


<VirtualHost virtual.host.fqdn:443>
  ServerName virtual.host.fqdn
   ...
</VirtualHost>


Also, for each virtual host, youll need an entry in /etc/hosts:

your-physical-ip     virtual.host.fqdn1
your-physical-ip     virtual.host.fqdn2
etc.


 
In a related matter.  I am having trouble implementing a second https
server one of virtual ip addresses.  Both domains have permanent redirects
to the https port.  However, the second ssl server does not even show up
when I conduct a port scan on its ip address.  Yet both have the same SSL
configuration elements.  I have compared the two configurations and I see
absolutely no difference except of course the SSL certificate and server
key targets.

When I disable the permanent redirect, I have not trouble getting to the
domain on port 80.   
Is it  possible to run 2 https servers using different virtual addresses
on the same physical server?

Yes -- its not recommended by apache, but its done.

The clients will either get warnings about the certificate being invalid,
or you can get a wildcard certificate if all your virtual ssl hosts use the
same domain name (box1.mydomain.com, box2.mydomain,com, box3.mydomain.com,
etc.). Apache still doesnt like this and will spit warnings, but it works.





 

Thanks that's great news.

I am still trying to resolve this error message I am getting.  Eric said I
need to be certain that my NameVirtualHost and VirtualHost directives
matched literally.  I reread the documentation on http://httpd.apache.org
and see what I believe is the same statement said differently.  That the
directives must match.  I checked each of the directives on my development
server and they match.  Here are the greps:

NameVirtualHost 192.168.0.93:80
<VirtualHost 192.168.0.93:80>
<VirtualHost 192.168.0.93:80>
<VirtualHost 192.168.0.93:80>
<VirtualHost 192.168.0.93:80>
<VirtualHost 192.168.0.93:80>

I also carefully verified that my /etc/hosts include a fqdn for each of
the VirtualHosts served from my development server.  Yet, I get the
following error:

[Sun Jun 06 14:02:11 2010] [warn] NameVirtualHost 192.168.0.93:80 has no
VirtualHosts

I am going nuts here with this.  What am I missing?

Thanks.

I need to have a website running in IIS on port 80, and message queue on
port
80.  Is that possible?
They would have different DNS addresses, and can have multiple IPs for the
machine.
Or at least both of them accessible from a port 80 request.  Is there some
way to have a virtual host between IIS and ActiveMQ or anything?

The main help i need is simply which direction should i explore for this,
i
should be able to figure out the specific configuration myself after i
know
what direction should work.

From my basic knowledge in the area I would think it is not possible, but
i'm being told to see if i can find any solution.  My thought would be it
should be possible to run the message queue on say port 81, and have IIS
forward any traffic that goes to its DNS to that port, but i really have
no
clue how to configure IIS to do that either.  

This is on a Windows Server 2007 box.

The reason for this is that the producer for the message queue is behind a
strict firewall that we can't get anything but http on port 80 to work,
and
we seem to not have enough servers to dedicate one to a message queue on
port 80.

Thanks for any help you can give me.

Manoj,

I am using 2.2.11.  Hmmmmm.

Are you saying if I move to 2.2.12 I can server multiple name-based ssl
virtual hosts on the same IP and ports?  

Even so, I may choose to use a multi-domain SSL certificate.  Being able
to use this for up to 25 domains seems like a lower cost alternative to
buying 25 separate certificates..  Have you had any experience with this? 
If so, do you know how to configure the virtual hosts for this?  I am
guessing one just points the SSL directives in each VirutalHost to the same
server key and certificate.  But please, advise if you know for certain how
this works.

Thanks.



-----Original Message-----
From: "Manoj Samtani" [manoj.### @hotmail.com]
Date: 06/06/2010 10:20 AM
To: us### @httpd.apache.org
Subject: RE: [us### @httpd] NameVirtualHost Directive - - - - Multiple SSL
 	virtual hosts

  Which Apache version are you using?

Till now Multiple SSL VHOST's are not available for same port and IP. This
can be possible through Apache 2.2.12.
So, you do not need to take tension ;-)


Just change your Apache Version.


Regards,
Manoj

 Date: Sat, 5 Jun 2010 16:12:09 -0400
 From: cov### @gmail.com
 To: use### @httpd.apache.org
 Subject: Re: [us### @httpd] NameVirtualHost Directive - - - - Multiple
SSL 	virtual hosts
 
 On Sat, Jun 5, 2010 at 4:08 PM, Jeff Shearer
<jef### @shearer-family.org> wrote:
 > [Sat Jun 05 15:51:21 2010] [warn] VirtualHost fqdn1.com:80
overlaps with VirtualHost fqdn2.com:80, the first has precedence, perhaps
you need a NameVirtualHost directive
 >
 > This is repeated for every name-based virtual host I have until
I added the fqdns to /etc/hosts.  However, I still get the error message
about NameVirtualHost 192.168.0.93 has no Virtual Hosts
 >
 
 The arguments for your 1 NameVirtualHost and correspoinding N
 VirtualHosts have to be identical in the literal sense, not just
 logically the same.
 -- 
 Eric Covener
 cove### @gmail.com
 
 

I have an Apache 2.2 server sitting on a box with one physical address and
multiple virtual addresses.  The physical address servers my name-based
virtual hosts where as the virtual addresses server ip-based virtual hosts.
  I read about the NameVirtualHost directive and tried to implement it but
it seems it does nothing but cause error messages.  I tried to put into the
main server config near my listen directives and received an error that
there was no virtual host despite the fact that there are more than one
virtual host.

I also tried to put it just before each of the virtual host directives and
received errors.

Is this directive needed, everything seems to work fine with out it. 

In a related matter.  I am having trouble implementing a second https
server one of virtual ip addresses.  Both domains have permanent redirects
to the https port.  However, the second ssl server does not even show up
when I conduct a port scan on its ip address.  Yet both have the same SSL
configuration elements.  I have compared the two configurations and I see
absolutely no difference except of course the SSL certificate and server
key targets.  

When I disable the permanent redirect, I have not trouble getting to the
domain on port 80.  Is it  possible to run 2 https servers using different
virtual addresses on the same physical server? 

  Hi, First of all I would like to mention that this is my first post on 
this list, I hope that this question is sent in the right place.
I have tried finding a solution for my problem, but it either doesn't 
exist, or my limited brain can't manage to figure out how to do it (I 
strongly hope it is the latter).

My issue is with the mod_jk plugin. I want to redirect my subdomain 
requests jira.example.com to a tomcat 6 virtual host called jiraapp. If 
I call the subdomain jira.example.com Im always getting the default 
tomcat 6 host called localhost.

I tested every part:
1.) mod_jk: if i change the workers name in either workers.properties or 
und er my enabled-site <VirtualHost> or if i change the mapping to
my 
virtual host (/etc/hosts: 127.0.0.1 localhost jiraapp) apache doesnt 
start. So I think this should work and is configured right.
2.) if I do a lynx jiraapp:8080 on the machine it works and shows the 
right webapp.

But if I access jira.example.com I always getting the default 
root-webapp of tomcat 6. So theres no difference in calling 
jira.example.com or example.com:8080. In general my setup looks like this:
1.) browse to jira.example.com
2.) apache <VirtualHost> hands over to mod_jk
3.) mod_jk redirects to worker.jiraworker.host:worker.jiraworker.port 
(jiraa pp:8009) via ajp
4.) jiraapp resolves to 127.0.0.1 trough /etc/hosts
5.) tomcat relays to the jiraapp name based virtual host

Do I miss something here? My guess would be that iam doing something 
wrong in step 4 because i can call lynx jiraapp on the local bash (5) 
and i can access the default virtual host from remote (1-3).

Help would be appreciated.

Versions:
Apache Tomcat/6.0.29
Sun-Java 1.6.0_20-b02
Apache/2.2.9 (Debian) mod_jk/1.2.26 proxy_html/3.0.0 Server
libapache2-mod-jk 1:1.2.26-2+lenny1

== jk.load ==

LoadModule jk_module /usr/lib/apache2/modules/mod_jk.so
JkWorkersFile /etc/apache2/workers.properties
JkLogFile /var/log/apache2/mod_jk.log
JkLogLevel info
# JkOptions +ForwardKeySize +ForwardURICompat -ForwardDirectories 
-ForwardLocalAddress
JkRequestLogFormat "%w %V %T"


== workers.properties ==

workers.tomcat_home=/opt/tomcat
workers.java_home=/usr/lib/jvm/java-6-sun
worker.list=jiraworker
worker.jiraworker.port=8009
worker.jiraworker.host=jiraapp
worker.jiraworker.type=ajp13


== /etc/hosts ==

::1 localhost ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts
127.0.0.1 localhost.localdomain localhost jiraapp


== server.xml ===

<Engine name="Catalina" defaultHost="localhost">
[...]
<Host name="localhost" appBase="webapps"
unpackWARs="true" autoDeploy="true"
xmlValidation="false" xmlNamespaceAware="false">
[...]
</Host>
<Host name="jiraapp" appBase="jiraapp" docBase="/opt/tomcat/jiraapp"
unpackWARs="true" autoDeploy="true"
xmlValidation="false" xmlNamespaceAware="false" >

== apache example.com virtual hosts configuration file ==
<VirtualHost *:80>
ServerName www.jira.example.com
ServerAlias jira.example.com
JkMount /* jiraworker
</VirtualHost>

Hi Moritz,

Try adding:

<Alias>jira.example.com</Alias>
<Alias>www.jira.example.com</Alias>

To the server.xml inside the <Host name="jirapp">

Best Regards,

Brett


-----Original Message-----
        From: Moritz Winter [mailto:moritz.a### @googlemail.com] 
        Sent: Tuesday, July 27, 2010 7:11 PM
        To: us### @tomcat.apache.org
        Subject: mod_jk and tomcat virtual hosts not connecting
        
          Hi, First of all I would like to mention that this is my first
        post on 
        this list, I hope that this question is sent in the right place.
        I have tried finding a solution for my problem, but it either
        doesn't 
        exist, or my limited brain can't manage to figure out how to do
        it (I 
        strongly hope it is the latter).
        
        My issue is with the mod_jk plugin. I want to redirect my
        subdomain 
        requests jira.example.com to a tomcat 6 virtual host called
        jiraapp. If 
        I call the subdomain jira.example.com Im always getting the
        default 
        tomcat 6 host called localhost.
        
        I tested every part:
        1.) mod_jk: if i change the workers name in either
        workers.properties or 
        und er my enabled-site <VirtualHost> or if i change the
mapping
        to my 
        virtual host (/etc/hosts: 127.0.0.1 localhost jiraapp) apache
        doesnt 
        start. So I think this should work and is configured right.
        2.) if I do a lynx jiraapp:8080 on the machine it works and
        shows the 
        right webapp.
        
        But if I access jira.example.com I always getting the default 
        root-webapp of tomcat 6. So theres no difference in calling 
        jira.example.com or example.com:8080. In general my setup looks
        like this:
        1.) browse to jira.example.com
        2.) apache <VirtualHost> hands over to mod_jk
        3.) mod_jk redirects to
        worker.jiraworker.host:worker.jiraworker.port 
        (jiraa pp:8009) via ajp
        4.) jiraapp resolves to 127.0.0.1 trough /etc/hosts
        5.) tomcat relays to the jiraapp name based virtual host
        
        Do I miss something here? My guess would be that iam doing
        something 
        wrong in step 4 because i can call lynx jiraapp on the local
        bash (5) 
        and i can access the default virtual host from remote (1-3).
        
        Help would be appreciated.
        
        Versions:
        Apache Tomcat/6.0.29
        Sun-Java 1.6.0_20-b02
        Apache/2.2.9 (Debian) mod_jk/1.2.26 proxy_html/3.0.0 Server
        libapache2-mod-jk 1:1.2.26-2+lenny1
        
        == jk.load ==
        
        LoadModule jk_module /usr/lib/apache2/modules/mod_jk.so
        JkWorkersFile /etc/apache2/workers.properties
        JkLogFile /var/log/apache2/mod_jk.log
        JkLogLevel info
        # JkOptions +ForwardKeySize +ForwardURICompat
        -ForwardDirectories 
        -ForwardLocalAddress
        JkRequestLogFormat "%w %V %T"
        
        
        == workers.properties ==
        
        workers.tomcat_home=/opt/tomcat
        workers.java_home=/usr/lib/jvm/java-6-sun
        worker.list=jiraworker
        worker.jiraworker.port=8009
        worker.jiraworker.host=jiraapp
        worker.jiraworker.type=ajp13
        
        
        == /etc/hosts ==
        
        ::1 localhost ip6-localhost ip6-loopback
        fe00::0 ip6-localnet
        ff00::0 ip6-mcastprefix
        ff02::1 ip6-allnodes
        ff02::2 ip6-allrouters
        ff02::3 ip6-allhosts
        127.0.0.1 localhost.localdomain localhost jiraapp
        
        
        == server.xml ===
        
        <Engine name="Catalina" defaultHost="localhost">
        [...]
        <Host name="localhost" appBase="webapps"
        unpackWARs="true" autoDeploy="true"
        xmlValidation="false" xmlNamespaceAware="false">
        [...]
        </Host>
        <Host name="jiraapp" appBase="jiraapp"
        docBase="/opt/tomcat/jiraapp"
        unpackWARs="true" autoDeploy="true"
        xmlValidation="false" xmlNamespaceAware="false" >
        
        == apache example.com virtual hosts configuration file ==
        <VirtualHost *:80>
        ServerName www.jira.example.com
        ServerAlias jira.example.com
        JkMount /* jiraworker
        </VirtualHost>
        



Brett Delle Grazie

NameVirtualHost your-physical-ip7:80
NameVirtualHost your physical-ip:443
...

<VirtualHost virtual.host.fqdn:80>

The name after VirtualHost should match the name after NameVirtualHost, so
this example is wrong, and should be
<VirtualHost your-physical-ip7:80>

Joost

Hi Gurus,

Any advise on initial tuning values for apache MinSpareServers,
MaxSpareServers, and StartServers, tcp tunings, ulimits, sysctl....

I'm running svn 1.6 over apache2 pre-fork. System load goes high as
much as 10 during heavy usage.

What would be your recommended MaxRequestsPerChild for heavy svn usage?

Thanks,


West

Weijunli,

I also have an environment that has similar very large datasets with
strict
latency.  Can you please elaborate on the custom changes you added to
cassandra to meet these sla, either code or configuration.  i am very
interested in learning more about the internal workings of cassandra and
performance.

Thanx,
Artie

On Thu, May 6, 2010 at 10:06 AM, Weijun Li <weiju### @gmail.com>
wrote:

 Our use case is a little different: our server is a typical high
volume
 transaction server that processes more than half billion requests per
day.
 The write/read ratio is close to 1, and the cluster needs to serve
10k
 write+read with strict latency (<20ms) otherwise the client will
treat it as
 failure. Plus we have hundreds of millions of keys so the generated
sstable
 files are much bigger that the ram size. In this case using mmap will
cause
 Cassandra to use sometimes > 100G virtual memory which is much
more than the
 physical ram, since we are using random partitioner the OS will be
busy
 doing swap.

 I have finally customized cassandra to meet the above requirement by
using
 cheap hardware (32G ram + SATA drives): one thing I learned is that
you have
 to carefully avoid swapping, especially when you need to cache most
of the
 keys in memory, swap can easily damage the performance of your in
memory
 cache. I also made some optimizations to reduce memory/disk
consumption and
 to make it easier for us to diagnose issues. In one word: cassandra
is very
 well written but there's still huge potential for you to improve it
to meet
 your special requirements.

 -Weijun


 On Wed, May 5, 2010 at 9:43 AM, Jordan Pittier
<jordan.p### @gmail.com>wrote:

> I disagree. Swapping could be avoided. I don't know Cassandra
internals
> mechanisms but what I am expecting is that whenever I want to
read rows that
> are not in RAM, Cassandra load them from hard drive to RAM if
space is
> available, and, if RAM is full to reply my query without saving
rows in RAM.
> No need for swapping.
>
> I have no try yet to change DiskAccessMode to standard, I hope it
will
> help me.
>
> Another thing : please dont post your benchmark figures without
any
> explanation on the work load generator or your cluster settings.
It really
> doesn't make any sense...
>
>
> On Wed, May 5, 2010 at 6:16 PM, Weijun Li
<weij### @gmail.com> wrote:
>
>> When you have much more data than you can hold in memory, it
will be
>> difficult for you to get around of swap which will most
likely ruin your
>> performance. Also in this case mmap doesn't seem to make much
sense if you
>> use random partitioner which will end up with crazy swap too.
However we
>> found a way to get around read/write performance issue by
integrating
>> memcached into Cassandra: in this case you need to ask
memcached to disable
>> disk swap so you can achieve move than 10k read+write with
milli-second
>> level of latency. Actually this is the only way that we
figured out that can
>> gracefully solve the performance and memory issue.
>>
>> -Weijun
>>
>>
>> On Wed, May 5, 2010 at 8:19 AM, Ran Tavory
<rant### @gmail.com> wrote:
>>
>>> I'm still trying to figure out where my slowness is
coming from...
>>> By now I'm pretty sure it's the reads are slow, but not
sure how to
>>> improve them.
>>>
>>> I'm looking at cfstats. Can you say if there are better
configuration
>>> options? So far I've used all default settings, except
for:
>>>
>>>     <Keyspace Name="outbrain_kvdb">
>>>       <ColumnFamily CompareWith="BytesType"
Name="KvImpressions"
>>> KeysCached="50%"/>
>>>
>>> 
<ReplicaPlacementStrategy>org.apache.cassandra.locator.RackAwareStrategy</ReplicaPlacementStrategy>
>>>      
<ReplicationFactor>2</ReplicationFactor>
>>>
>>> 
<EndPointSnitch>org.apache.cassandra.locator.EndPointSnitch</EndPointSnitch>
>>>     </Keyspace>
>>>
>>>
>>> What does a good read latency look like? I was expecting
10ms, however
>>> so far it seems that my KvImpressions read latency is
30ms and in the system
>>> keyspace I have 800ms :(
>>> I thought adding KeysCached="50%" would improve my
situation but
>>> unfortunately looks like the hitrate is about 0. I
realize that's
>>> application specific, but maybe there are other magic
bullets...
>>>
>>> Is there something like adding cache to the system
keyspace? 800 ms is
>>> pretty bad, isn't it?
>>>
>>> See stats below and thanks.
>>>
>>>
>>> Keyspace: outbrain_kvdb
>>>         Read Count: 651668
>>>         Read Latency: 34.18622328547666 ms.
>>>         Write Count: 655542
>>>         Write Latency: 0.041145092152752985 ms.
>>>         Pending Tasks: 0
>>>                 Column Family: KvImpressions
>>>                 SSTable count: 13
>>>                 Space used (live): 23304548897
>>>                 Space used (total): 23304548897
>>>                 Memtable Columns Count: 895
>>>                 Memtable Data Size: 2108990
>>>                 Memtable Switch Count: 8
>>>                 Read Count: 468083
>>>                 Read Latency: 151.603 ms.
>>>                 Write Count: 552566
>>>                 Write Latency: 0.023 ms.
>>>                 Pending Tasks: 0
>>>                 Key cache capacity: 17398656
>>>                 Key cache size: 567967
>>>                 Key cache hit rate: 0.0
>>>                 Row cache: disabled
>>>                 Compacted row minimum size: 269
>>>                 Compacted row maximum size: 54501
>>>                 Compacted row mean size: 933
>>> ...
>>> 

Hi,

I'm using Tomcat 6.0.26, Java 1.6 and wondering what tools/strategies you
use to tune your garbage collection parameters?

Further, does anyone know how to read entries in the garbage collection
log? 
Entries in my log look like

Desired survivor size 10944512 bytes, new threshold 1 (max 15)
 [PSYoungGen: 129311K->3232K(136512K)] 558882K->434085K(585920K),
0.0090900
secs]

Thanks, - Dave

This is a multi-part message in MIME format.Quite some time ago, one of my
colleagues started the following thread:

http://marc.info/?l=apache-httpd-dev&...942300453&w=2
<http://marc.info/?l=apache-httpd-dev&...942300453&w=2>

There was some discussion around the Windows utf-8 codepage and it's
relationship to the local codepage.  Here's Bill Rowe's comment on that:
http://marc.info/?l=apache-httpd-dev&...046012720&w=2
<http://marc.info/?l=apache-httpd-dev&...046012720&w=2>

I think I finally understand it, and at the same time I don't understand
it :).
I've been playing with this recently trying to get this to work, and I
found that if I set the system locale to the same language as the UTF-8
encoded character in LDAP is from it works.  Which makes sense.  The
code page for the system language contains a mapping for the UTF-8
character.

I don't even pretend to understand Windows and/or encoding enough to
truly get this limitation, but is it just some microsoft-ism that limits
UTF-8 character mapping to the default code page?

Does this mean that if you needed to support DNs with multiple languages
that are not all included on a single Windows codepage (that's
associated to a specific language locale) that this would not be
possible on Windows?

Am I even making sense?

Thanks,
Andy

https://issues.apache.org/bugzilla/show_bug.cgi?id=49568

           Summary: Apache mod_authnz_ldap built with the MS LDAPSDK does
                    not handle multi-byte DNs
           Product: Apache httpd-2
           Version: 2.2.15
          Platform: PC
        OS/Version: Windows Server 2003
            Status: NEW
          Severity: major
          Priority: P2
         Component: mod_authz_ldap
        AssignedTo: bu### @httpd.apache.org
        ReportedBy: dop### @moonteeth.com


Created an attachment (id=25729)
 --> (https://issues.apache.org/bugzilla/attachment.cgi?id=25729)
ldif file containing user that doesn't work.  Password for the user is the
string: two

If a user's DN contains a multi-byte character, auth_ldap will not
authenticate
the user because the multi-byte characters returned by the searchRequest
will
be converted to ?'s when attempting to bind.

For example a user with the following dn:
cn=t[UTF-8: 0xE6 0x88 0x91], ou=people

UTF-8: 0xE6 0x88 0x91 is the utf-8 character for the chinese word for me.

Will not work because the bind request will be for
cn=t?,ou=people

See this mailing list thread with a discussion on this:
http://marc.info/?l=apache-httpd-dev&...942300453&w=2

Attached is an ldif file containing the dn I used.

Author: kkolinko
Date: Sun Jun  6 19:17:58 2010
New Revision: 951930

URL: http://svn.apache.org/viewvc?rev=951930&view=rev
Log:
Fix https://issues.apache.org/bugzilla/show_bug.cgi?id=48971
Make stopping of leaking Timer threads optional and disabled by default.
It is merge of r950164, r950614 from trunk.

Modified:
    tomcat/tc6.0.x/trunk/   (props changed)
    tomcat/tc6.0.x/trunk/STATUS.txt
   
tomcat/tc6.0.x/trunk/java/org/apache/catalina/core/StandardContext.java
   
tomcat/tc6.0.x/trunk/java/org/apache/catalina/loader/WebappClassLoader.java
    tomcat/tc6.0.x/trunk/java/org/apache/catalina/loader/WebappLoader.java
    tomcat/tc6.0.x/trunk/webapps/docs/changelog.xml
    tomcat/tc6.0.x/trunk/webapps/docs/config/context.xml

Propchange: tomcat/tc6.0.x/trunk/

Hi guys, sorry for the bother. I do have a question related to how a
certain feature was implemented in Tomcat (6.0, probably 
all of them), that's why I address the question to the dev-list and the
user-list. 
I do have a fairly complicated case, in which I have a filter for jsp's,
and I use a custom request wrapper in the respective filter. 
My request wrapper overrides setAttribute and getAttribute (does some
internal management, not relying on the "super"), 
and I had the surprise to see that the setAttribute from my wrapper
doesn't get called from the 
org.apache.catalina.core.ApplicationHttpRequest setAttribute, at least not
for the "special attributes" (and I need the 
javax.servlet.include attributes, which Tomcat considers "special"). 

The code that I find is problematic is copied underneath (the
if(!setSpecial) part), and the question is "why doesn't the 
getRequest.setAttribute get called ALL THE TIME". The problem manifests
when I do a jsp:include, and when the
filter on the included jsp checks my wrapper, can't find the
javax.servlet.include.* attributes. This is somewhat bothering, 
considering that I don't have this issue with Weblogic and Websphere, both
do call my wrapper's set/getAttribute... 

p.s. the code I believe is at fault is:

org.apache.catalina.core.ApplicationHttpRequest
...
public  void setAttribute(String name,
    Object value) 
{
        if (name.equals(Globals.DISPATCHER_TYPE_ATTR)) {
            dispatcherType = value;
            return;
        } else if (name.equals(Globals.DISPATCHER_REQUEST_PATH_ATTR)) {
            requestDispatcherPath = value;
            return;
        }
        if (!setSpecial(name, value)) {
            getRequest().setAttribute(name, value);
        }
}So, the questions are: 

1. do you think is a big issue if I remove the "if" and I just use instead
to instructions - 
- setSpecial(name, attribute);
- getRequest().setAttribute...
In my tests it seemed to work fine, but maybe I don't grasp all the
implications. 

2. any particular reason for calling the getRequest().setAttribute only
for some (non-special) attributes, but not for others
(javax.servlet.include.*)? 

thanks for any suggestion/
ranjix 



 		 	   		  
Hotmail has tools for the New Busy. Search, chat and e-mail from your
inbox. Learn more. 		 	   		  

Hi All,

It seems that o.a.Catalina.Core MBeans descriptor has got operations
containing POJO objects instead of primitive types which make them
inaccessible via JConsole.
I suggest the following to avoid this situation. Any comments and
suggestions are appreciated.


*Class: StandardContext *
'
Method         : addValve (org.apache.catalina.Valve)
Suggestion : Modify the signature to addValve(String className) where
className is that of the Valve to be instantiated. This is acceptable
since
all the Valve classes have no arg constructors AFA I noticed.

Method         : removeValve(org.apache.catalina.Valve)
Suggestion :   Modify the signature to removeValve(String oname)

Methods       :
add/removeLifecycleListener(org.apache.catalina.LifecycleListener)
Suggestion  :  Remove the methods since the method can be provided in a
generic way with only involving primitives. Any ideas on this?


These methods can be included in a seperate MBean class
(e.g:StandardContextMBean) so that original class interface is left
intact.

*Class: StandardEngine*

Method           :    addChild(org.apache.catalina.Container)
Suggestion   :    addHost(String name, String appBase, boolean
autoDeploy,boolean deployOnStartup,boolean deployXML, boolean unpackWARs)

This method can be included in a seperate MBean class so that original
class
interface is left intact. Since only a host can be added as a child mbean
method name can be named as addHost to remove any ambiguity.

*Class: StandardHost*

Method            :  addChild(org.apache.catalina.Container)
Suggestion    :  addContext(String path, String docBase, boolean
xmlValidation, boolean xmlNamespaceAware, boolean tldValidation, boolean
tldNamespaceAware)


*Class: StandardService*

Method            :  addConnector(org.apache.catalina.connector.Connector)
Suggestion    :   addConnector(String address, int port, boolean isAjp,
boolean isSSL)


*Class: StandardWrapper*

Method             :
add/removeLifeCycleListener(org.apache.catalina.LifecycleListener)
Suggestion     : Remove the methods since the method can be provided in a
generic way with only involving primitives.


Regards,
Chamith