This looks good but may generate quite a load of threads.
Did you try without the AJP connector?
Do you know the "ulimit -n" of the tomcat process?
Did you see any error in the artifactory.log file?
Fred Simon
www.jfrog.org
On Jul 7, 2010 11:05 PM, "Clarence" <[email protected]> wrote:
> Here is my server.xml file.
>
>
> <Service name="Catalina">
> <Connector port="8585" maxHttpHeaderSize="8192"
> maxThreads="400" minSpareThreads="125" maxSpareThreads="200"
> enableLookups="false" redirectPort="8443" acceptCount="200"
> connectionTimeout="0" maxKeepAliveRequests="1"
> disableUploadTimeout="true" />
>
> <!-- This is here for compatibility only, not required -->
> <Connector port="8509" protocol="AJP/1.3" enableLookups="false"
> redirectPort="8443" maxThreads="400" minSpareThreads="125"
> maxSpareThreads="200" connectionTimeout="60000"/>
>
> <Engine name="Catalina" defaultHost="localhost">
> <Realm className="org.apache.catalina.realm.UserDatabaseRealm"
> resourceName="UserDatabase" />
> <Host name="localhost" appBase="webapps"
> unpackWARs="true" autoDeploy="true"
> xmlValidation="false" xmlNamespaceAware="false">
> </Host>
> </Engine>
>
> </Service>
> </Server>
>
> Clarence
>
>
> On Wed, Jul 7, 2010 at 12:31 PM, Frederic Simon <[email protected]> wrote:
>
>> Hi,
>>
>> From memory, a 503 using a direct connection to tomcat HTTP connector
means
>> that Tomcat server connector cannot accept any more connections.
>> I think you should look at the server.xml file in tomcat and configure a
>> higher number of threads and may be accept. You can copy your current
>> server.xml here if you want.
>>
>> Your memory and machine looks OK to me.
>>
>> Hope this help,
>> Fred.
>>
>> On Wed, Jul 7, 2010 at 6:31 PM, Clarence <[email protected]> wrote:
>>
>>> Hello everyone. I am having no luck trying to resolve an issue with my
>>> Artifactory installation. First of all Artifactory is installed on a box
>>> with 12 GB of RAM it has Intel Xeon Dual Core Processes 2.27GHz and its
>>> running CentOS 5.4. The artifactory installation is using as its
>>> jcr:repo/filesystem-mysql. Its deployed via tomcat 5.5.28 using JDK
1.5.0_21
>>> as its JAVA_HOME.
>>>
>>> When the nightly builds are kicked off from Hudson it generates failures
>>> because its reporting that when attempting to upload to Artifactory its
>>> receiving 503 errors. When the calls are made from hudson its using the
URI
>>> of http://hostname:8585/artifactory/<repo-location>. I am not using ajp
>>> proxy in this direct request to the tomcat. I would like some help in
trying
>>> to find a root cause as to why this would happen (My opinion is that
Hudson
>>> process is timing out awaiting for the upload to complete and once
>>> artifactory returns it generates a 503 as the client has already closed
the
>>> connection too early for artifactory to respond.)
>>>
>>>
>>>
>>> I have read the mail thread countless times and google 503 and have run
>>> into a few comments regarding Poorly configured VM and possible tuning
>>> issues with the tomcat or JAVA. Could this be the issue in my case. The
>>> Tomcat is deployed out of the box with no major configuration changes.
>>> Here's what I set in my setenv.sh file:
>>>
>>>
>>> export JAVA_OPTS="-server -Xms1536m -Xmx1536m -XX:MaxPermSize=512m
>>> -XX:ParallelGCThreads=20 -XX:+UseConcMarkSweepGC -XX:+UseParNewGC
>>> -XX:SurvivorRatio=8 -XX:TargetSurvivorRatio=75
-XX:MaxTenuringThreshold=0
>>> -XX:+AggressiveOpts -XX:+PrintGCDetails -XX:+PrintGCTimeStamps
>>> -XX:-TraceClassUnloading -Dsun.io.useCanonCaches=false
>>> -Dsun.net.client.defaultConnectTimeout=60000
>>> -Dsun.net.client.defaultReadTimeout=60000"
>>> export JAVA_HOME=/usr/java/jdk1.5.0_21
>>>
>>>
>>>
>>> Would upgrading to the latest artifactory help me in resolving this
issue?
>>>
>>> Also I notice when I perform a top on the box I see that most to all of
my
>>> memory is consumed by the tomcat-artifactory process is from a bad
setting
>>> or I have setup tomcat incorrectly for artifactory?
>>>
>>> Here's the top from the machine in question.
>>>
>>> top - 11:26:36 up 136 days, 15:08, 1 user, load average: 1.17, 1.24,
>>> 1.07
>>> Tasks: 287 total, 1 running, 286 sleeping, 0 stopped, 0 zombie
>>> Cpu(s): 3.9%us, 2.5%sy, 0.0%ni, 93.7%id, 0.0%wa, 0.0%hi, 0.0%si,
>>> 0.0%st
>>> Mem: 12292916k total, 11265908k used, 1027008k free, 605780k buffers
>>> Swap: 16777208k total, 240k used, 16776968k free, 5460072k cached
>>>
>>> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+
>>> COMMAND
>>> 8682 tomcat-a 17 0 2446m 1.7g 16m S 101.1 14.6 299:38.82 java
>>>
>>> Any help would be appreciated thank you in advance
>>>
>>> Clarence
>>>
>>>
>>>
------------------------------------------------------------------------------
>>> This SF.net email is sponsored by Sprint
>>> What will you do first with EVO, the first 4G phone?
>>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>>> _______________________________________________
>>> Artifactory-users mailing list
>>> [email protected]
>>> https://lists.sourceforge.net/lists/listinfo/artifactory-users
>>>
>>>
>>
>>
>> --
>> Co. Founder and Chief Architect
>> JFrog Ltd
>> http://www.jfrog.org/
>> http://twitter.com/freddy33
>>
>>
>>
------------------------------------------------------------------------------
>> This SF.net email is sponsored by Sprint
>> What will you do first with EVO, the first 4G phone?
>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>> _______________________________________________
>> Artifactory-users mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/artifactory-users
>>
>>
------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Artifactory-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/artifactory-users
