Caching not working

Hi All

I'm still trying to wrap my head around the Cayenne caching system,
(maybe Friday afternoon isn't doing me any good)

I have a test case as follows:

public void testBasicSelectQueryWithoutDaoWithCache() {
    initCache();
    assertEquals(0, getDataContext().getQueryCache().size());

    SelectQuery q = new SelectQuery(Admin.class);
    q.setCacheGroups("testGroup");
    q.setCacheStrategy(QueryCacheStrategy.SHARED_CACHE);

    List<Admin> result = getDataContext().performQuery(q);
    assertEquals(2, result.size());

    SelectQuery q2 = new SelectQuery(Admin.class);
    List<Admin> result2 = getDataContext().performQuery(q2);
    assertEquals(2, result2.size());

    assertEquals(1, getDataContext().getQueryCache().size());
}

private void initCache() {
    DataContext context = (DataContext) getDataContext();
    DataDomain domain = context.getParentDataDomain();
    domain.setQueryCacheFactory(new OSQueryCacheFactory());
    domain.setSharedCacheEnabled(true);
    this.osCache = true;
}

I would expect that the the second query q2 would not execute any SQL
however the QueryLogger logs the second select statement.

Also the assert on the query cache size fails the it appears the size
is zero rather than 1 as I would have expected.

If I change the second call to performQuery to query q (List<Admin>
result2 = getDataContext().performQuery(q);) then in the log I get
DEBUG [AbstractConcurrentReadCache.java:694] => get called (key=Admin)
but the cache size in the last Assert is still 0 ???

I've read through the docs a few times and I can not see what I am doing
wrong.

Thank you

Gary

I am trying to use the key cache, and I had some basic questions as to how
it works. There seem to be settings based on % and total count, but not on
total cache size like most caching mechanisms. Is the cache stored in the
jvm heap or somewhere else? We have 24gb RAM nodes with a 8gb jvm heap
size.
Is the key cache sharing the 8gb heap or is it separate? Can we end up
pushing out to swap if the key cache gets too big (if not in jvm)? We
would
prefer to set the jvm heap to 8gb and key cache to 8-10 gb, and then do
some
testing with cache hit rates to determine how many nodes we need to keep
most keys in cache. High key cache hit rates are vital to good read
performance, and we need some help to figure out the best way to
optimize/design our cluster.

What is the the key cache capacity in cfstats? How is this calculated? I
also see the stats get reset often and cache is cleared quite often, what
triggers the flushing of key cache? Is it compaction? Given that our keys
do
not change too much we would prefer them to stay within cache as long as
possible. We have a very wide data model so having all keys in cache is a
realistic possibility for us, we just need some help figuring out how to
design our cluster for this. We are currently testing with 4 nodes with
replication factor of 3 (24gb ram 8 core), and we plan to expand the node
count as required to fit most/all keys into memory.

Thanks in advance for any help you can provide.

Wayne

Hi,
I did try to read apache docs given here.
http://httpd.apache.org/docs/2.2/mod/....html#cacheenable
http://www.google.co.in/search?hl=en&...;oq=&gs_rfai=
actually I am a bit new to this so if some one can let me know.
I am having a reverse proxy based website.


The reverse proxy forwards any request to main website to an internal
server.Looks like this
I tried putting
CacheEnable disk http://site5.abc.com/
but after that apache2 failed to restart how should I do that can any
one let me know.

<VirtualHost *:80 >

        ServerName site5.abc.com
        ServerAdmin webma### @localhost
        ServerName site5.abc.com
ProxyRequests off
<Proxy *>
        Order deny,allow
        Allow from all
</Proxy>
        ProxyPass / http://192.168.1.5
        ProxyPassReverse / http://192.168.1.5
        ErrorLog /var/log/apache2/site5_error.log
        # Possible values include: debug, info, notice, warn, error, crit,
        # alert, emerg.
        LogLevel warn
        CustomLog /var/log/apache2/site5_access.log combined
</VirtualHost>

Hi Gurus,

I have a subversion server that authenticates with windows ad and i
want to cache my authentication for better performance.

here's my setting:

LoadModule authnz_ldap_module
/usr/lib/apache2-prefork/mod_authnz_ldap.so
LoadModule ldap_module                   
/usr/lib/apache2-prefork/mod_ldap.so

LDAPTrustedMode SSL
LDAPVerifyServerCert off
LDAPSharedCacheSize 500000
LDAPCacheEntries 1024
LDAPCacheTTL 43200
LDAPOpCacheEntries 1024
LDAPOpCacheTTL 43200
LDAPSharedCacheFile /srv/home/svn/ldap_cache
LDAPConnectionTimeout 10

<Location /ldap-status>
        SetHandler ldap-status
        Order allow,deny
        Allow from all
</Location>

    LDAPTrustedClientCert CERT_BASE64 /etc/apache2/ssl.crt/ad1.cer
    LDAPTrustedClientCert CERT_BASE64 /etc/apache2/ssl.crt/ad2.cer

    <Location /unix-repo>
        DAV svn
        SVNPath /srv/svn/unix-repo

        AuthName "Please use your ACTIVE DIRECTORY for Authentication"
        AuthType Basic
        AuthBasicProvider ldap
        AuthzLDAPAuthoritative off
        Include /etc/apache2/.ldapbinddn
        AuthLDAPURL "ldaps://172.23.12.137
172.23.12.136:636/DC=domain,DC=local?sAMAccountName?sub?(objectClass=user)"

        SSLRequireSSL
        AuthzSVNAccessFile /srv/config/svn_acl/unix-repo
        Require valid-user
        SVNPathAuthz off
    </Location>


Browsing https://localhost/ldap-status show an empty LDAP Cache
Information

Can anyone help me, what's missing?

Thanks,


West

Folks

I would like to start moving towards the freeze of 4.1 API. All new
features added in trunk since 4.0 seem reasonably okay. The only thing
that still blocks the 4.1-beta1 release is the the relatively untested
API of the recently added caching module. I personally find it difficult
to decide for myself whether the design is flexible enough unless there
are other implementations (for instance, a file system or a ehache
based) of the same API. 

I see three different approaches we could pursue:

(1) Take more time to stabilize HttpClientCache API at the price of
delaying 4.1 GA. 
(2) Freeze the API of HttpClient and HttpMime and clearly state
HttpClientCache is still experimental and its binary / API stability
cannot be guaranteed, despite being a part of a GA release
(3) Reduce the API to a bare minimum (make pretty much all but a few
must-have classes package private, make protected methods of public
classes private, etc) and gradually relax the restriction on class /
method visibility as the module matures.

What do you think? I am kind of leaning towards the last option.

Oleg

Hello

I noticed that the time needed by puppetd to write on disk its cached
catalog has increased a lot in Puppet 0.25.

I'm using 0.25.4 and, with a catalog with around:
~400 resources: i need 3 sec to write this on disk
~900 resources: i need 8 sec
~1900 resources: i need 21 sec.


I really have catalogs with 900 resources, in production., 1900 was only
to test the problem.
But I'm sure this value will increase.


I do not reproduce this issue with Puppet 0.24.

Jon et al

I started looking at the Caching API more closely primarily to find out
how difficult / feasible it should be to put together alternative cache
backends such as echache or file system based. I stumbled upon a few
minor issues I thought I should discuss with you. 

(1) HttpCacheUpdateCallback

I think I understand the rationale behind the interface but kind of
dislike the fact that its implementations often manipulate the cache
behind the scene in non-obvious ways. One call to #updateCacheEntry may
result in mutation of multiple entries and multiple invocations of
#putEntry methods. Do we need #updateCacheEntry at all? What we need is
ability to update the cache while ensuring the consistency of its
content. Why not just use an exclusive lock to execute multiple cache
operations as a unit of work? 

Please take a look at the patch attached and let me know if you can live
with the proposed changes?

(2) Do we really need HttpCacheEntrySerializer? It does not appear to be
used anywhere but in tests.

Oleg

Hello

In my route, I have 2 endpoints that shares the same real URL 
underneath. To be clear, the difference between the 2 endpoints is the 
dataFormat mode: one is PAYLOAD and the other is POJO.

The idea was to be able to performance test each of theses routes 
independently while sharing most of the route definition.

It seems this is not possible because the cache use the endpoint URL as 
a key (which is the same with my 2 Endpoints)

Is there any way to fix this behaviour? I have indeed seen that the 
ProducerCache has a doGetProducer method with a pooled parameter, but 
that parameter is always set to "true".

Cheers



Office: +33 4 76 29 76 19
GSM:    +33 66 43 00 555

Bull, Architect of an Open World TM
http://www.bull.com

Hi,

I have an installation of Apache running, so thinking of using it as a
personal caching server after being fed up of firefox crashes which
clears the cache.
I have set it up using the manual and some guides using Google, but
how do I know that its working ?
Here are my settings:

ProxyRequests On
ProxyVia Off
ProxyBadHeader Ignore
ProxyStatus On
<Proxy *>
Order deny,allow
Deny from All
Allow from 192.168.1.0 127.0.0.1 localhost
</Proxy>
AllowCONNECT 443
CacheDefaultExpire 3600
CacheMaxExpire 604800
CacheIgnoreCacheControl On
CacheStorePrivate On
CacheStoreNoStore On
CacheIgnoreNoLastMod On
CacheRoot /home/proxy_cache
CacheEnable disk /
CacheMaxFileSize 5242880
CacheIgnoreURLSessionIdentifiers jsessionid PHPSESSID sid

What happens if I remove the CacheMaxExpire directive ? Will Apache
check for the document's last-modified time ?
I would like latest stuff on websites I visit, so please guide me in
the right direction.

I'm not 100% sure if the subject correctly describes the problem I've
been having, but it's the closest I can get with my troubleshooting.
My setup looks like this:

* 2 puppetmasters running 0.25.4 on Ubuntu, running under passenger
* backend content (etc and var) shared over NFS
* haproxy load balancing across the 2 puppetmasters
* mysql for stored configs

I just upgraded from 0.24.8 to 0.25.4 a couple of weeks ago.  The
setup we've been using above has worked fine since we implemented it
months ago, so I don't believe that there is any problem with NFS or
the load balancer.  I have a handful of custom functions, and after
updating to 0.25.4, puppetmaster started complaining about one of
them, a simple function called nagios_name.  This function takes an
FQDN and turns it into a name we use in Nagios and mcollective
(turning "support.arces.net" into "arces.support" for example).  The
function is basic ruby and is available for you to look at here:
http://monachus.pastebin.com/yLF1syqU.  The function works fine.

The error that puppetmaster reports is:

Unknown function nagios_name at /var/www/localhost/puppet/etc/
manifests/outsidein_nodes.pp:16 on node some.node.com.

It doesn't report this all of the time - instead it reports it about
40% of the time, while other nodes before and after it do not report
the error.  It seems that a node with a problem will always have the
problem, and a node where it works will always work.  This reinforces
the fact that the function is fine - it works and has worked for
months.

My thought is that it's some sort of caching issue, and I even thought
it might be a race condition with the backend storage being NFS - one
puppetmaster loading a cached yaml file before the other was done
writing it or something.  I've done all of the following, all with no
success:

* turn off one puppetmaster so traffic isn't split across them
* move yaml files for node/facts to local storage instead of NFS
* enable IP-based persistence in haproxy so that traffic from a client
always goes to the same puppetmaster
* --ignorecache in config.ru for puppetmaster

What I've discovered, however, is more interesting.  It appears that
if I go into the actual nagios_name.rb file and change it in any way
(add a single character of whitespace) and restart Apache, the error
goes away.  The file is detected as different and loaded for delivery
to the clients, and everything works fine after that.  I discovered
this by adding debug() statements to the function 2 weeks ago, only to
find that it worked fine from then on.  The problem resurfaced today
when I turned the 2nd puppetmaster back on, and I decided to try it
with whitespace - same thing.  Clears it right up.  This tells me that
there is some sort of caching wonkiness happening somewhere, but I'm
not able to figure out where.

Perhaps one of the variables the function is looking for (fqdn?) isn't
available at the time it's requested, resulting in a compile error
that isn't always visible?

I'm pleased to have a workaround, but to go from "Unknown function" to
"everything is cool" by adding a space to the file and saving it isn't
really much of a long-term solution.

I'm sending this to the list rather than filing a bug report to see if
anyone has experienced anything like this or has any thoughts.  If
there's any further information I can give to help narrow down the
source of the problem, I'm happy to do so.

Adrian

Hi,

  I just wanted to confirm that lazy-loading and caching in myBatis is
incompatible because of serialization issues with the enhanced
versions of objects loaded in a result-map.

  I have tried to use a different caching implementation (EHCache),
but the same problem arises, as myBatis appears to always make a call
to serialize(...) in the SerializedCache class before sending the
value to the custom caching solution.

  Is there anyway to bypass this serialization feature for a pure in-
memory cache (other than with a custom build)?

Thanks in advance,

Mike

Hi guys,

I'm stuck with a JNDI resource object factory problem. Hopefully you can
help me out with this issue.

My idea was to provide multiple web applications running inside a Tomcat
instance with configuration properties from config files located outside
the WAR file of the applications. I'm currently trying to provide
these configuration properties with a global naming resource (defined
inside the server.xml) which is consumed by the said applications to
keep the configuration container independent and to avoid direct file
access from a web application.

So far almost everything already works perfectly! Except that Tomcat
obviously caches resources which are looked up via JNDI. This way I have
to restart Tomcat every time the external config files changes because
the web applications never access my ObjectFactory again after Tomcat
has cached the result of the ObjectFactory.

Accordingly to the documentation an implementation of ObjectFactory
should be accessed each time a web application looks it up with its JNDI
key. But that's definitely not the case here.

Any ideas or thoughts on this issue? Is there any way to disable this
behavior of Tomcat's JNDI resource management?
Or any comments on this plan to externalize the application
configuration in general?

Thanks in advance!

Marco

Am 17.07.2010 12:56, schrieb Mark Thomas:
 On 17/07/2010 09:50, Marco Ehrentreich wrote:
> Accordingly to the documentation an implementation of
ObjectFactory
> should be accessed each time a web application looks it up with
its JNDI
> key. But that's definitely not the case here.
 
 What documentation?
 
 Mark
 
Sorry, I'll try to find the said documentation again. I think it was
something from the original Tomcat documentation but I can't find it at
the moment.

Maybe I missunderstood it but at least it didn't mention that the result
of an ObjectFactory is only produced once and then cached by Tomcat.

Can you tell me some more details about this issue? Or would you
recommend a completely different approach?

Marco

caching module should use HttpParams-style configuration

client cache may be a shared cache but is caching responses to requests
with Authorization headers

Hi, looking in the  http://activemq.apache.org/c-integration.html C
cross-client  implementations  
http://svn.apache.org/repos/asf/activemq/sandbox/openwire-c/ OpenWire  and

http://svn.stomp.codehaus.org/browse/stomp/trunk/c/ STOMP  C
implementations, I see only prototype code, which project would be most
easy
to get running though?

Does anyone have a working example or experience building one of the
prototype C implementations (possibly with brief build/run
instructions)???

It would be GREATLY appreciated! I need to get a JMS consumer/producer
going
using C, like yesterday!!

Hi All,

I am working on upgrading tomcat from 5.5.28 to 5.5.29 for one of the
applications. I see that the website renders and works fine in 5.5.29 on
port 8080 (non SSL) but with  SSL (port 8443) the website doesnot run at
all. When I try to see what's going on in Fiddle, I see 502 error. Also
nothing is written to the log flies. It is as if tomcat is not even
running
in port 8443.

Under tomcat 5.5.28, the site renders fine with SSL and non SSL.

Is there something I could be missing?

Regards,
Kareem

I'm trying to ban certain IPs from visiting my site, so that they 
instead see a message saying "Your IP has been banned, email me if 
you think this is an error."  I've *almost* got it working -- when 
people visit URLs like
http://209.160.28.154/index.html
or
http://209.160.28.154/foo-does-not-exist
they see the "banned IP" message.  However, the problem is that if 
you try to access the front page:
http://209.160.28.154/
from a banned IP address, you see the "Apache Test Page for CentOS" 
page, instead of seeing the "banned IP" message.  Anybody recognize 
this problem or have an idea of what could be causing it?

In my httpd.conf file, I changed "AllowOverride None" to 
"AllowOverride All" in both the default <Directory /> tag and inside

the <Directory "/var/www/html"> tag -- I placed a modified copy of 
httpd.conf at:
http://209.160.28.154/httpd.conf
and in /var/www/html I placed a .htaccess file containing these lines:
 >>>
ErrorDocument 403 /banned_ip.php
order deny,allow
deny from 71.112.32.149
 >>>
and restarted the server.  (The page 
http://209.160.28.154/banned_ip.php shows the message you're supposed 
to see when connecting from a banned IP.  71.112.32.149 is my home 
machine IP which I've "banned" for testing purposes.)

So like I said, that almost works, where 
http://209.160.28.154/index.html gives the right error message, but 
http://209.160.28.154/ does not.  Any idea how to change is to that 
all URLs under http://209.160.28.154/ will give the "banned IP" 
message if connecting from a banned IP?

	-Bennett

On 24/05/2010, Mark Thomas <mar### @apache.org> wrote:
 On 24/05/2010 22:34, sebb wrote:
  > Is anyone working on - or wanting to work on - BCEL at present?


 Try the Jakarta dev list. BCEL isn't a commons component.

Duh -sorry!

  Mark


  >
  > I know that Findbugs are keen to get away from using their
patched version.
  >
  > I can do some tidying up of test cases and basic warnings if
that would help.
  >
  > There seem to have been a lot of fixes since 5.2, so it would
be good
  > to get a new release out.
  >

 > 

Hi guys,

I'm using the Apache Commons FileUpload lib and I'm getting this strange
behaviour in just one of my servers. All others work perfectly.

servletfileupload.parseRequest(request) always returns an empty list.

But if I check for:

ServletFileUpload.isMultipartContent(request)

This returns true. So, the list should not be empty.

I'm really stuck here. Any ideas?

Thanks,
Fernando