Quantcast
Channel: VMware Communities : Popular Discussions - Hyperic User Forums
Viewing all 52618 articles
Browse latest View live
↧

no sigar-x86-winnt.dll in java.library.path, using Sigar with Webstart

March 25, 2009, 9:30 am
$
0
0
I am facing problem where sigar dlls are not being picked up when using via Webstart. We need the dlls to be downloaded on the client machine so that Machine specific Sigar calls can be done.

I am able to use sigar API/dll's, when I am using it in a Java application by configuring "java.library.path" to point the location of dlls. (I cannot keep the dll's and jars in the same location.)

For making Sigar.dll work with webstart, I have done the following
1) Made a Jar out of the dll's, named it sigarNative.jar.
2) I have updated the jnlp file on the server as follows.
       <resources os="Windows">
          <property name="sun.java2d.noddraw" value="true" />
          <property name="JINTEGRA_NATIVE_MODE" value="" />
          <nativelib href="<path>/sigarNative.jar"/>
     </resources>

This is how other custom native libraries are configured in our application.

Even after doing the above steps I get the below errors.

[LOGGED STACK TRACE BEGIN]
org.hyperic.sigar.SigarException: no sigar-x86-winnt.dll in java.library.path
     at org.hyperic.sigar.Sigar.load(Sigar.java:154)
     at com.portware.util.health.SigarHealthRetreiver.initialize(SigarHealthRetreiver.java:193)

Can anybody point me to the work around for making Sigar dll's work with Webstart.

Thanks,

VK
↧
Search

Sigar Python binding

January 12, 2010, 5:47 am
$
0
0
Hi all!

How should I build Sigar so I can use it with Python?

I've just downloaded hyperic-sigar-1.6.3 from SourceForge, unpacked it and, from bindings/python, ran:
$ python setup.py install

I tried running this under Fedora 12  and MacOS X 10.5, but the same error occured (attached error log). I also checked out the source from the svn repository, and the same error occured.

Thanks in advance!

Tatiana

-----
running install
running build
running build_py
creating build
creating build/lib.macosx-10.3-fat-2.6
copying sigar.py -> build/lib.macosx-10.3-fat-2.6
running build_ext
building '_sigar' extension
creating build/temp.macosx-10.3-fat-2.6
gcc -arch ppc -arch i386 -isysroot /Developer/SDKs/MacOSX10.4u.sdk -fno-strict-aliasing -fno-common -dynamic -DNDEBUG -g -O3 -I../java/sigar-bin/include -I/Library/Frameworks/Python.framework/Versions/2.6/include/python2.6 -c _sigar.c -o build/temp.macosx-10.3-fat-2.6/_sigar.o -Wall
_sigar.c:20:19: error: sigar.h: No such file or directory
_sigar.c:21:28: error: sigar_fileinfo.h: No such file or directory
_sigar.c:22:26: error: sigar_format.h: No such file or directory
(...)
error: command 'gcc' failed with exit status 1
↧
↧

hq Server and agent configuration

December 31, 2010, 9:05 pm
$
0
0
After installing the hq-server (on default ports ) and the agent on same machine, I installed same agent on another machine, I am not able to see the agents on other machine.
server port 7080
server sec port 7443

agent listening on different machine same subnet 192.168.1.xxx 2144

communication between two machines are fine i used telnet from agent machine to server on port 7080 works fine , tried telnet from server to agent machine on port 2144 works fine. brought down all internal and external firewalls still the agent says it cannot send auto discovery message to server , here is the exact error message.
AutoInventoryCommansServer unable to send autoinventory platform data to server  ... provider not setup .

After reading forums i understood its the networking error , but after telnet i found there is no network problem , I used traffic monitor to see if any messages going out , i dont see any messages going out of 2144 port to server .

Is there any way that JVM itself would be blocking the traffic,
is there any thing i am missing , please help.

thanks
↧

Apache 2.0 vhost and mod_rt trouble.

January 21, 2011, 8:38 am
$
0
0
Hello,
I'm new to Hyperic so my question can be obvius but I can't find documentation or other forum about that.
I'have an apache with 4 virtual host (in separate config files included by httpd.conf) on a centOs 5.x system.
I have configured snmp e mod_rt as documentationa and in hyperchq the service is discovered and all virtualhost seen.


The first problem was about rt_log files that in httpd.conf don't contains any data, so i add in any vhost definition the lines:

CustomLog logs/rt_log  rt_log
CustomLog logs/enduser enduser

And now the logs is write at any http call but  the traffic data is not send in virtualhost metric but only in the general metric. I look in the log format and no domainname is found so I suppose that the rt_log is for general and I must write a rt_log for domain. In internet I found that on redhat mod_rt installation the name on virtualhost must be <domain><port>rt_log but don't work.

What is the right configuration to allow metric data on vhost?

My httpd.config part:

# Hyperic SMNP module
LoadModule snmpcommon_module modules/libsnmpcommon.so
LoadModule snmpagt_module modules/libsnmpmonagt.so

SNMPConf conf
SNMPVar var

LoadModule rt_module modules/mod_rt.so

LogFormat "%S" rt_log
LogFormat "%R" enduser
CustomLog logs/rt_log  rt_log
CustomLog logs/enduser enduser


An example of my vhost.conf

<VirtualHost *:80>
  ServerName test.localhost:80

   DocumentRoot "/home/www/combo/"

  <Directory /home/www/combo>
    Options Includes FollowSymLinks
    AllowOverride AuthConfig Limit FileInfo
    DirectoryIndex home.html
    order allow,deny
    allow from all
  </Directory>

        CustomLog logs/rt_log  rt_log
        CustomLog logs/test.localhost80rt_log rt_log
        CustomLog logs/enduser enduser

</VirtualHost>



An example of my logs/rt_log

/core/js/A_scriptLayout.js 1295520443284 2726 200 93.148.242.199
/core/js/cuwa/xtcore.js 1295520443288 1601 200 79.48.123.101
/core/js/B_scriptForms.js 1295520443301 653 200 93.148.242.199
/core/media/img/bg_spriteCD.png 1295520443307 367 200 79.48.123.101
/promozioni-sconti/offerte/e-coupon/101213-wb.html 1295520443341 5483 200 82.49.203.189
/core/media/img/bg_spriteBoxSpec1.png 1295520443347 286 2



Any halp is welcome.

Regards
Lucio
↧

agent communication

February 8, 2011, 3:56 pm
$
0
0
I am new to Hyperic. I have installed the Server with agent on one machine and two agents on other machines. The agents install correctly and contact the hyperic server, but they do not show up on the dashboard and in the HQ health it only shows the agent for the server. Any insights would be appreciated.
↧
↧

Not able to use Extended File system Configuration property.

February 22, 2011, 10:08 pm
$
0
0
It was specified that servers list will not be displayed in case of Host Device while using Extended  File system Configuration property in New Auto scan Discovery .

I am unable to submit the form by providing the directories in the Extended  File system Configuration section. Can some one help me in resolving this issue ...please.
↧

Noob - Custom plugin for log file tracking / alerting

March 9, 2011, 11:01 am
$
0
0
Well, I'm new, so I'll start from the beginning. Please point out anything I'm doing wrong, because I can't get my plugin to work :)

I have a Hyperic Agent 4.5.1 installed and running at /monitoring/hyperic.

I have a java process running, which looks like 'ps -ef | grep log-alert-test.jar' (and returns as 'user PID PPID C STIME TTY TIME java -jar log-alert-test.jar' ), on the same machine.  It spits out 'severe' logs to /var/log/log-alert-test.log.  I want to *eventually* send e-mail alerts based on when these severe messages get logged.

I 'wrote' an xml plugin called 'log-alert-test.xml' which doesn't work (as I'm not really sure how to specify the process name or process id), and installed it to /monitoring/hyperic/hq-plugins.  I know the plugin isn't configured right, but I try to run it like:

sudo java -jar /monitoring/hyperic/bundles/agent-4.5.1/pdk/lib/hq-product.jar -Dplugins.include=log-alert-test -m discover -a metric

The message I receive upon running:
Failed to load Main-Class manifest attribute from
/monitoring/hyperic/bundles/agent-4.5.1/pdk/lib/hq-product.jar

The plugins looks like:

<?xml version="1.0"?>

<!DOCTYPE plugin [
  <!ENTITY process-metrics SYSTEM "/pdk/plugins/log-alert-test.xml">
]>

<plugin name="log-alert-test">

  <server name="The Tagged Name of the Linux Box in HQ Server?">
     
     <property name="HAS_BUILTIN_SERVICES"
          value="true"/>

        <property name="DEFAULT_LOG_FILE"
         value="/var/log/log-alert-test.log"/>
         
     <property name="DEFAULT_LOG_TRACK_ENABLE"
          value="true"/>
          
     <property name="DEFAULT_LOG_LEVEL"
          value="Info"/>
             
     <service name="log alert test">
        <plugin type="log_track"
             class="org.hyperic.hq.product.LogFileTailPlugin"/>
        <config>
              <option name="process.query"
                 description="Process Query"
                   default="State.Name.eq=java -jar log-alert-test.jar,CredName.User.eq=root"/>
        </config>
    </service>
   
  </server>

</plugin>



One of the many tutorials I'm using:
http://support.hyperic.com/display/DOC/Log+File+Tracking+Plugin
↧

SNMP Trap problem on Hyperic 4.5

March 11, 2011, 2:19 am
$
0
0
I have a problem and i call it like:
1. [AgentDaemon] Agent startup failed:
java.lang.NullPointerException
     at org.hyperic.hq.plugin.netdevice.SNMPTrapReceiver.shutdown(SNMPTrapReceiver.java:96)

2. 2011-03-11 03:07:28,531 ERROR [Thread-0] [ProductPluginManager] registerPlugin=Network Device
org.hyperic.hq.product.PluginException: Address already in use: Cannot bind
     at org.hyperic.hq.plugin.netdevice.SNMPTrapReceiverPlugin.init(SNMPTrapReceiverPlugin.java:68)

Can somebody explain why these( http://support.hyperic.com/display/EVO/Configuring+HQ+as+an+SNMP+Trap+Receiver and http://support.hyperic.com/display/EVO/snmpTrapReceiver.listenAddress) doesn't work?
↧
Search

Agent not able to communicate with Server

March 16, 2011, 10:03 am
$
0
0
Hi,
I've set up Hyperic HQ , server and agent on the same Windows 2003 Server machine and everytime the agent starts up and tries to communicate with the server I get the following exception:

2011-03-16 16:48:19,199 DEBUG [Thread-1] [WindowsRegistryScan] Windows Registry Scan completed, took: 0.02
2011-03-16 16:48:19,199 DEBUG [Thread-1] [AutoinventoryCommandsServer] Sending autoinventory report to server: [ScanState platform={agentToken=null queueStatus=0 customProperties=[B@1ad0839 productConfig=null controlConfig=null measurementConfig=[B@f6f1b6 diff=0 ignored=false platformTypeName=Win32 lastApproved=null certdn=CAM-AGENT-1300294071707-5765534761088932561-6738951194760866720 fqdn=SSOPP1.LCH_MEMBER.COM name=null location=null description=Microsoft Windows 2003 cpuCount=16 id=null mTime=1300294080232 cTime=1300294080232} ips={queueStatus=0 diff=0 ignored=false address=127.0.0.1 mACAddress=00:00:00:00:00:00 netmask=255.0.0.0 id=null mTime=1300294080232 cTime=1300294080232},{queueStatus=0 diff=0 ignored=false address=10.20.5.56 mACAddress=D8:D3:85:F9:7E:EC netmask=255.255.255.0 id=null mTime=1300294080232 cTime=1300294080232},{queueStatus=0 diff=0 ignored=false address=10.20.81.126 mACAddress=D8:D3:85:F9:7E:EE netmask=255.255.255.0 id=null mTime=1300294080232 cTime=1300294080232} serversIncluded=true]
2011-03-16 16:48:19,199 DEBUG [Thread-1] [AutoinventoryCommandsServer] Unable to send autoinventory platform data to server, sleeping for 15 secs before retrying.  Error: Unable to communicate with server -- provider not yet setup
org.hyperic.hq.bizapp.client.AgentCallbackClientException: Unable to communicate with server -- provider not yet setup
     at org.hyperic.hq.bizapp.client.AgentCallbackClient.getProvider(AgentCallbackClient.java:87)
     at org.hyperic.hq.bizapp.client.AutoinventoryCallbackClient.aiSendReport(AutoinventoryCallbackClient.java:49)
     at org.hyperic.hq.autoinventory.agent.server.AutoinventoryCommandsServer.scanComplete(AutoinventoryCommandsServer.java:331)
     at org.hyperic.hq.autoinventory.ScanManager.scanComplete(ScanManager.java:308)
     at org.hyperic.hq.autoinventory.Scanner.notifyScanComplete(Scanner.java:278)
     at org.hyperic.hq.autoinventory.Scanner.start(Scanner.java:213)
     at org.hyperic.hq.autoinventory.ScanManager.mainRunLoop(ScanManager.java:141)
     at org.hyperic.hq.autoinventory.ScanManager.access$000(ScanManager.java:41)
     at org.hyperic.hq.autoinventory.ScanManager$1.run(ScanManager.java:107)
2011-03-16 16:48:20,010 DEBUG [ScheduleThread] [ScheduleThread] Platform schedule is null
2011-03-16 16:48:20,010 DEBUG [ScheduleThread] [ScheduleThread] Waiting 1000 ms until 16/03/11 16:48:21

This "Unable to communicate with server -- provider not yet setup" we've seen time and time again on other test platforms.

Any help to resolve this issue would be appreciated.
↧
↧

False Alerts

May 5, 2011, 12:44 am
$
0
0
Is anyone experiencing false alerts with Availability. (i.e - The service which is being monitoring for Availability is up and running but Hyperic is sending alerts "Service is not available")
↧

Communicate with HQServer via agent proxy

May 6, 2011, 8:35 am
$
0
0
Hi,

I'm trying to use the proxy functionality of the agents but I'm having issues.

My setup is effectively:


Server <----> Agent A <----> Agent B

Agent A is what I want as the proxy for Agent B so I have put the entries into Agent B's agent.properties file for agent.proxyHost=Agent A and agent.proxyPort=2144.

Agent B will not communicate with either the Server or Agent A. Latest version of server and agent (4.5.1.2) are being used (currently evaluation version with timie limited license)

Has anyone got proxies working? Any advice would be appreciated.

Regards,

Mark.
↧

Bug in Sigar CpuInfo.getModel() on solaris x86

May 9, 2011, 12:31 am
$
0
0
his bug is specific to the native library of solaris x86. when calling from java CpuInfo.getModel(), the return value seems to always be 'x86' (the architecture) whereas i tried it on several machines whose model should return something like 'opteron' or 'Xeon'
↧

Network discovery help (Hyperic 4.5)

May 12, 2011, 9:28 am
$
0
0
I am very new to Hyperic. Just installed,configured Server and agent sucessfully on RHEl 5.6. I want to discover my ip network. Where is the network discovery piece?
how do I discover my ip network with wildcard using snmp community settings. Also, where is the global configuration settings for snmp strings?
Please advice
Gurmit
↧
↧

Monitor Zimbra Log Files set to Error shows ALL and INFO entries

May 25, 2011, 2:06 pm
$
0
0
I have an agent running on a zimbra host for which ive imported into the inventory. Everything is running great except for some reason I am getting a TON of log noise when I am only monitoring for errors and only one log file.

platform.log_track.files
/opt/zimbra/log/audit.log

platform.log_track.level
ERROR

Am I missing something here I have a DOT for log entry showing up on every index. Here is a sample..


    * [05/25/2011 04:52 PM] Log Message (/opt/zimbra/log/audit.log): /opt/zimbra/log/audit.log: 2011-05-25 16:51:29,772 INFO [Pop3SSLServer-6799] [ip=10.10.10.10;] security - cmd=Auth; account=info@mydomain.com; protocol=pop3;
    * [05/25/2011 04:54 PM] Log Message (/opt/zimbra/log/audit.log): /opt/zimbra/log/audit.log: 2011-05-25 16:53:31,862 INFO [Pop3SSLServer-6800] [ip=66.66.66.66;] security - cmd=Auth; account=request@mydomain.com; protocol=pop3;


I have error selected why is above showing INFO type entries?
↧

Distinguishing JMX servers

June 27, 2011, 1:05 am
$
0
0
Hi,

I'm a bit of Hyperic newbie so please be gentle :)

Following JMX plugin tutorial available at Hyperic 4.5 Documentation pages I wrote simple plugin for monitoring our Netty based server (attached). And it works great.

The problem I'm facing now is that I have been given the task to add similar custom MBean monitoring capabilities to our applications hosted on WebLogic and Tomat servers. What I did was to create almost identical plugin changing name of the plugin, name of the server, metrics, process query and jmx properties and to put a file created this way amongst other plugins. Unfortunately, what I achieved was that WebLogic and Tomcat were both recognized as Input Handler server and as such they started throwing exceptions about metrics unavailable.

I thought that process query itself is enough to distinguish Java servers with JMX capabilities but that doesn't seem to be true. What is it then? What can I do to make Hyperic recognize my applications properly?

Thanks,
Adam
↧
Search

Server multiple listen ports

June 28, 2011, 4:45 am
$
0
0
Is it possible for the HQ server to listen on more than one port?

The reason I ask is that some of the remotely monitored sites have multiple agents, so to monitor the second agent I would create a firewall rule with a different port other than 2144 or 7443.

Any help would be greatly appreciated!
↧

Hyperic 4.4 and Active MQ 5.5

July 1, 2011, 3:50 am
$
0
0
Hi everyone!!

I´m trying to configure Hyperic 4.4 because I want to monitoring Active MQ 5.5. Is it possible?? How can I do??

Thanks.

Regards
↧
↧

Monitor a Windows service

July 15, 2011, 2:02 pm
$
0
0
What is the best way to throw an alert if a windows service crashes? Availability seems to be average over time and takes a while if you use it as an alert metric. I would like to send an alert immediately if the service stops

Thanks you
↧

Hyperic 4.5.2 fails to complete autoscan of RabbitMQ

July 21, 2011, 7:51 am
$
0
0
I'm trying to get Hyperic to detect RabbitMQ .

I've upped the log level for the RabbitMQ plugin in hyperic and I can see that it detects rabbitmq.

but fails to complete the autoscan seems to be when it requests a list of the virtual hosts.

I'm using RabbitMQ 2.5.1-1
on RHEL 6 x86_64
Rabbit MQ is working fine.

Hyperic HQ Agent, version 4.5.1.3  & 4.5.2
Hyperic Server 4.5.2

It find the install ok and reads the erlang cookie fine. It then says it's closing the erlang connection
Next thing we get is a bad rpc exception.

Please help.

I've change the erlang cookie value to protect the innocent.


2011-07-21 15:33:36,642 DEBUG [Thread-3] [RabbitServerDetector] [getServerResources] platformConfig=platform.type=Linux,platform.name=nogbad.shef.ac.uk,platf
orm.fqdn=nogbad.shef.ac.uk,
2011-07-21 15:33:36,654 DEBUG [Thread-3] [RabbitServerDetector] [getServerResources] pids.length=1
2011-07-21 15:33:36,655 DEBUG [Thread-3] [RabbitServerDetector] mnesia -mnesia dir "/var/lib/rabbitmq/mnesia/rabbit@nogbad"
2011-07-21 15:33:36,655 DEBUG [Thread-3] [RabbitServerDetector] doCreateServerResource
2011-07-21 15:33:36,659 DEBUG [Thread-3] [RabbitServerDetector] [getProcessHome] home=/var/lib/rabbitmq
2011-07-21 15:33:36,661 DEBUG [Thread-3] [ErlangCookieHandler] Attempting to read file /var/lib/rabbitmq/.erlang.cookie
2011-07-21 15:33:36,662 DEBUG [Thread-3] [ErlangCookieHandler] Successfully read NREHPIHJCVYWWOYYGLFD from /var/lib/rabbitmq/.erlang.cookie
2011-07-21 15:33:36,662 DEBUG [Thread-3] [RabbitServerDetector] ProductConfig[node=rabbit@nogbad,authentication=NREHPIHJCVYWWOYYGLFD,]
2011-07-21 15:33:36,663 DEBUG [Thread-3] [RabbitServerDetector] Discovered nogbad.shef.ac.uk RabbitMQ Node rabbit@nogbad productConfig=node=rabbit@nogbad,aut
hentication=NREHPIHJCVYWWOYYGLFD,process.query=State.Name.re=[beam|erl],Args.*.eq=-sname,Args.*.eq=%node%, customProps=rabbit.boot=/var/lib/rabbitmq/mnesia/r
abbit@nogbad-plugins-expand/rabbit,erlang.process=/usr/lib64/erlang/erts-5.8.1/bin/beam.smp,node.name=rabbit@nogbad,node.path=/var/lib/rabbitmq/mnesia/rabbit
@nogbad,node.pid=25379,
2011-07-21 15:33:36,668 DEBUG [Thread-3] [HypericRabbitAdmin] [HypericRabbitAdmin] init(rabbit@nogbad,NREHPIHJCVYWWOYYGLFD)
2011-07-21 15:33:36,825 INFO  [Thread-3] [SingleConnectionFactory] Established shared Rabbit Connection: org.springframework.erlang.connection.DefaultConnection@79ec41c0
2011-07-21 15:33:36,907 DEBUG [Thread-3] [HypericErlangTemplate] Closing erlang connection Shared Otp Connection: org.springframework.erlang.connection.DefaultConnection@79ec41c0
2011-07-21 15:33:36,929 DEBUG [Thread-3] [HypericErlangTemplate] Closing erlang connection Shared Otp Connection: org.springframework.erlang.connection.DefaultConnection@79ec41c0
2011-07-21 15:33:36,934 DEBUG [Thread-3] [HypericErlangTemplate] Closing erlang connection Shared Otp Connection: org.springframework.erlang.connection.DefaultConnection@79ec41c0
2011-07-21 15:33:36,935 DEBUG [Thread-3] [HypericRabbitAdmin] [HypericRabbitAdmin] destroy()
2011-07-21 15:33:36,935 ERROR [Thread-3] [AutoScan] AutoScan failed for RabbitMQ
org.springframework.erlang.ErlangBadRpcException: {badrpc,{'EXIT',{undef,[{rabbit_access_control,list_vhosts,[]},{rpc,'-handle_call_call/6-fun-0-',5}]}}}
        at org.hyperic.hq.plugin.rabbitmq.core.HypericErlangTemplate.executeRpcAndConvert(HypericErlangTemplate.java:72)
        at org.hyperic.hq.plugin.rabbitmq.core.HypericRabbitAdmin.getVirtualHosts(HypericRabbitAdmin.java:75)
        at org.hyperic.hq.plugin.rabbitmq.detect.RabbitServerDetector.getServerResources(RabbitServerDetector.java:106)
        at org.hyperic.hq.autoinventory.scanimpl.NullScan.scan(NullScan.java:83)
        at org.hyperic.hq.autoinventory.Scanner.start(Scanner.java:198)
        at org.hyperic.hq.autoinventory.ScanManager.mainRunLoop(ScanManager.java:141)
        at org.hyperic.hq.autoinventory.ScanManager.access$000(ScanManager.java:41)
        at org.hyperic.hq.autoinventory.ScanManager$1.run(ScanManager.java:107)
↧

Critical installer bug

August 16, 2011, 6:57 pm
$
0
0
This is an issue in at least 4.5.2. I say critical because it essentially renders the entire HQ webapp unusable; not sure if it affects data/metric collections.

Problems start to arise when one opts to install Hyperic HQ against an existing Oracle DB. First, there is the problem of setup instructions at this link being inaccurate: http://support.hyperic.com/display/EVO/Set+Up+Oracle#SetUpOracle-ObtainOracleJDBCDriver%28HypericHQonly%29

Step 6 should read: Copy the driver jar file to hyperic-hq-installer-4.x.y/installer/lib

Second, and more importantly, is the fact that the installer fails to do anything about the JDBC driver jar not getting copied to hq-engine/hq-server/webapps/ROOT/WEB-INF/lib. Clearly, without this step the site will not work at all, as evidenced by this stack trace on server startup:

2011-08-15 19:41:38,738 WARN  [main] [org.hibernate.cfg.SettingsFactory@117] Could not obtain connection metadata
java.sql.SQLException: oracle.jdbc.driver.OracleDriver
        at org.apache.tomcat.jdbc.pool.PooledConnection.connectUsingDriver(PooledConnection.java:208)
        at org.apache.tomcat.jdbc.pool.PooledConnection.connect(PooledConnection.java:157)
        at org.apache.tomcat.jdbc.pool.ConnectionPool.createConnection(ConnectionPool.java:617)
        at org.apache.tomcat.jdbc.pool.ConnectionPool.borrowConnection(ConnectionPool.java:561)
        at org.apache.tomcat.jdbc.pool.ConnectionPool.init(ConnectionPool.java:427)
        at org.apache.tomcat.jdbc.pool.ConnectionPool.<init>(ConnectionPool.java:132)
        at org.apache.tomcat.jdbc.pool.DataSourceProxy.createPool(DataSourceProxy.java:94)
.............
Caused by: java.lang.ClassNotFoundException: oracle.jdbc.driver.OracleDriver
        at org.apache.catalina.loader.WebappClassLoader.loadClass(WebappClassLoader.java:1680)
        at org.apache.catalina.loader.WebappClassLoader.loadClass(WebappClassLoader.java:1526)
        at java.lang.Class.forName0(Native Method)
        at java.lang.Class.forName(Unknown Source)
        at org.apache.tomcat.jdbc.pool.PooledConnection.connectUsingDriver(PooledConnection.java:202)
        ... 136 more

It appears that the installer does make an attempt to copy the supplied ojdbc*.jar to the resultant lib/ directory of the server but that is insufficient for Tomcat to load the driver!

The workaround is to either copy the driver to hq-engine/hq-server/webapps/ROOT/WEB-INF/lib or hard-link it there (NOTE: By default, Tomcat does not support symlinks - at least not the way it's bundled with HQ).

I am not sure whether the Oracle JDBC driver could not be bundled with with the installer for licensing reasons, but the problem does not seem to affect MySQL installs since the connector jar is in the WEB-INF/lib already.

I am contemplating patching the tarball (server-4.5.2.tar.gz) locally to include the driver jar and repackaging it with the install here internally, since we have to do this anyway to automate installation with new HQU and other plugins. Is there a legal reason to not do this, keeping in mind the package will not redistributed?
↧
Viewing all 52618 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>