[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: 回复: 回复: Openshift Origin Manually deploy fail. Follow oo_deploy_guide_comprehensive



+++ ┍ζυκε┙ [16/12/13 13:15 +0800]:
I'm sorry for my mistake, I use rhel. And mcollective.log only has DEBUG message.

This is good to know.  Comments inline.

Hopefully others on this list know more about the state of Origin on
RHEL.  If you are new to OpenShift you might consider installing it on
Fedora first and becoming familiar with all the components before
trying to install Origin on RHEL.



when I run "oo-accept-broker -v" on broker, it say /ect/openshift/console.conf must have session_secret, so I copy the SESSION_SECRET in /ect/openshift/broker.conf to console.conf.
Then oo-accept-broker run without error.


and I run "oo-accept-node -v" on node,  it takes two error:
1.ruby193-mcollective is not run
2.SELinux context for ruby193-mcollective iis incorrect

On RHEL you must use ruby193-mcollective.



So I show some additional messages below:
   When I follow the guide to deploy, the command is "service mcollective start" but not ruby193-mcollective. At that time, CLI say unrecognized service.

Which guide are you following?  You are likely following the
instructions for Fedora.

   So I install mcollective 2.2.1 by "yum install mcollective", and go on deploying.


Is there any influence?
I stop mcollective and start ruby193-mcollective service.
My mcollective conf file has set in /opt/rh/ruby193/root/etc/mcollecitve/server.conf.

This is the configuration file:

/opt/rh/ruby193/root/etc/mcollective/server.cfg

On your system check the value of 'logfile'.  It will likely be
/var/log/ruby193-mcollective.log.  Check that file to ensure the
OpenShift Agent is loading (and check the other settings in that
config file for errors).  You should see a line in the logs like:

Found openshift at /opt/rh/ruby193/root/usr/libexec/mcollective/mco
llective/agent/openshift.rb

After that line you should see it loading the cartridge metadata.	

I run "oo-accept-node -v" again, now it left "error 2".

To fix the selinux error try this:

semanage fcontext -a -e / /opt/rh/ruby193/root
restorecon -R /opt/rh/ruby193/root

I will go on solving this error, and hope that you can give me some suggestions. Thanks help a lot.


I run "oo-diagnostics -v" in the broker and node host, it seems that it may be the node host problem.
Broker:
   1.Node host not found. But "mco ping" can work.
   2.FAIL: rescue in block in run_tests
      error running test_broker_certificate: #<noMethodError: undefined method 'scan' for nil: NilClass>
Node:
   1.SELinux context for ruby193-mcollective is incorrect
   2.mcollective not start.  And I start ruby193-mcollective instead now.
   3.test_mcollective_context
      FAIL: test_mcollective_context
     Mcollectived is not running in the expected SELinux context, which
     may result in node execution failures. Please check that the correct
     context is set on /usr/sbin/mcollectived and that the correct SELinux
     policies are loaded.
       Expected: system_r:openshift_initrc_t:s0-s0:c0.c1023
       Found: unconfined_r:unconfined_t:s0-s0:c0.c1023
                  unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
                  unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
                  unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
                  unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023


By the way, I reboot my broker host, and use /rest/cartridges again. I can see python cartridges that I install in the node ^-^. When I install new cartridges, do I need to reboot or restart broker host every time ?

Whenever you install a new cartridge you will need to run the
following:

# On the Nodes
service ruby193-mcollective restart

# On the Brokers
oo-admin-broker-cache --clear --console

__________________________________________________________________________________


+++ ┍ζυκε┙ [15/12/13 16:14 +0800]:
thanks for help, it make better than before, but openshift still not work normally.


I re-deploy openshift.
I can use "mco ping" without some error, and now I can see localhost/console/application_types website.


But it occur some strange error:
 In the "Choose a type of application" website, I find some cartridges are here but I doesn't install in the previous section.
  Drupal, WordPress, Spring, CakePHP, Django,etc....I only install the php cartridge.
 So I skip the other cartridges and try to create a php application, it fail.
 I receive the info:
   "Maintenance in progress
    The OpenShift service is temporarily unavailable while we perform server maintenance"
 There is no application created on the node host.
 Then I install the python cartridge to check whether /application_types will change, it still the same as before.


I find that the broker error_log show "No request sent, we did not discover any nodes", but "mco ping" is find.
And I use REST API(broker/rest/cartridges) to find cartridges, it only show "cron" and "php5.3" cartridges but not "python".
I have no idea what to do, please help

Can you clarify if you are using RHEL, CentOS, Fedora or something
else?

Since you are running 'mco ping' I'm going to assume you are using
Fedora.  If that is the case it could be that something is causing the
openshift MCollective agent not to load.  Check
/etc/mcollective/server.cfg and see where your log file is configured.
Then look there and make sure the openshift agent is loading.  You can
try 'tail -f $logfile' and restart the service to see if there are any
errors at start up.



__________________________________________________________________________________


If I recall correctly, because you're using the nightly the mcollective config is now located in /opt/rh/ruby193/

On Sat, Dec 14, 2013 at 2:45 PM, ┍ζυκε┙ <yuanzilu foxmail com> wrote:

Hi, I need some help!

When I follow the http://openshift.github.io/documentation/oo_deployment_guide_comprehensive.html to deploy origin manually, I suffer some error.

I deploy broker and node separated. When I finish the guide, the broker is not work well. I can see the console web site, but when I click "create your first application" it will wait for a long time and reach proxy error finally. I find that it should activemq not work correctly:

when I use "mco ping", it will print

transmit to localhost failed: Broken pipe

connection.receive returning EOF as nil - resetting connection.

connect to localhost failed: Broken pipe will retry(#0) in 5




the /var/log/openshift/broker/httpd/error_log will look like above error.

As a result, when I finish section 9(conf node mcollective), the mco ping is ok and it can receive the node's response.But it turn fail when I finish all section.

  and the /var/log/activemq/activemq.log will show the below info again and again

///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////

2013-12-13 20:13:52,450 | WARN | Exception occurred processing:

CONNECT

content-type:text/plain; charset=UTF-8

content-length:0

passcode:*****

login:mcollective

: java.lang.SecurityException: User name [mcollective] or password is invalid. | org.apache.activemq.transport.stomp.ProtocolConverter | ActiveMQ Transport: tcp:///0:0:0:0:0:0:0:1:41684 61613

2013-12-13 20:13:52,451 | WARN | Transport Connection to: tcp://0:0:0:0:0:0:0:1:41684 failed: java.io.IOException: User name [mcollective] or password is invalid. | org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ Transport:tcp:///0:0:0:0:0:0:0:1:41684 61613

2013-12-13 20:13:54,450 | INFO | Stopping tcp://0:0:0:0:0:0:0:1:41684 because Failed with SecurityException: User name [mcollective] or password is invalid. | org.apache.activemq.broker.TransportConnection | ActiveMQ BrokerService[broker.example.com] Task-36

2013-12-13 20:13:57,459 | WARN | Failed to add Connection ID:localhost.localdomain-33070-1386935489311-3:14, reason: java.lang.SecurityException: User name [mcollective] or password is invalid. | org.apache.activemq.broker.TransportConnection | ActiveMQ Transport: tcp:///0:0:0:0:0:0:0:1:41685 61613

//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////

I doublecheck my work, it not worng.

I follow the guide exactly, and the mcollective conf is correct

I try to solve this problem by editing /etc/activemq/activemq.xml, jetty-realm.properties, /etc/mcollective/client.cfg on broker or /etc/mcollective/server.cfg on node, but it fail

What can I do to solve this problem?thanks for help!!

I have tried oo-install and puppet depolyment, but they not work, so I have to try manual guide and get the problem.

I use the repo that the guide offer(https://mirror.openshift.com/pub/openshift-origin/nightly/rhel-6/dependencies/x86_64/), so the ActiveMq version is 5.8 and mcollective version is 2.2.1

_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

.


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]