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

Re: Openshift Origin, Proxy errors on /console/application_types.



-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

This is a CentOS 6.4 box, so yes, RHELish.

On 11/19/2013 10:06 AM, Luke Meyer wrote:
> Fedora doesn't need the puppetlabs repos to install recent puppet,
> so it's not a problem there. This is restricted to the RHEL
> section.
> 
> ----- Original Message ----- From: "The Dude"
> <michael mcconachie hotmail com> To: "Nathan"
> <lagern lafayette edu>, "Luke Meyer" <lmeyer redhat com> Cc:
> users lists openshift redhat com Sent: Tuesday, November 19, 2013
> 10:03:39 AM Subject: RE: Openshift Origin, Proxy errors on
> /console/application_types.
> 
> 
> Awesome, glad you go it. I remember running into that one too --
> but I have only added that line for RHEL'ish boxes. Do we do it for
> FC based distros too?
> 
> 
>> Date: Tue, 19 Nov 2013 09:26:52 -0500 From: lagern lafayette edu
>>  To: lmeyer redhat com Subject: Re: Openshift Origin, Proxy
>> errors on /console/application_types. CC:
>> users lists openshift redhat com
>> 
> SUCCESS!
> 
> It turns out that I did in fact add the exclusions, but i did not
> add them to the puppet-deps repo. This should probably be more
> clear in the install docs, it just says to add the exclude line "to
> the puppet repo entries in the file
> /etc/yum.repos.d/puppetlabs.repo", not where, or which repo. I made
> a guess, and guessed wrong. At any rate.....
> 
> 
> I updated the repo file, including that exlude line on the deps
> repo as well as allllll of the others (i decided to be thorough)
> and then did a yum downgrade mcollective* activemq. It did its
> thing, and now mcollective starts right up.
> 
> I can now access the pages that perviously proxy errord.
> 
> THANKS A TON for all of the help!
> 
> 
> 
> On 11/19/2013 06:14 AM, Luke Meyer wrote:
>>>> Oh yeah; I remember that problem. What happened in my case is
>>>> that puppetlabs released a newer version of activemq in its
>>>> repos, and that's what is installed on your system. And it's
>>>> expecting some different config which specifies that
>>>> parameter.
>>>> 
>>>> What you need to do is add the exclude from the puppet doc: 
>>>> http://openshift.github.io/documentation/oo_deployment_guide_puppet.html
>>>> 
>>>> 
>>>> 
> Add exclude=*mcollective* activemq to the puppet repo entries in
> the file /etc/yum.repos.d/puppetlabs.repo
>>>> 
>>>> ... and then yum downgrade activemq. I don't *think* you need
>>>> to do anything about mcollective now that ours is SCL-ized...
>>>> 
>>>> 
>>>> ----- Original Message ----- From: "Nathan T. Lager" 
>>>> <lagern lafayette edu> To: "Luke Meyer" <lmeyer redhat com>
>>>> Cc: users lists openshift redhat com Sent: Monday, November
>>>> 18, 2013 10:29:40 PM Subject: Re: Openshift Origin, Proxy
>>>> errors on /console/application_types.
>>>> 
>>>> I'll bet activemq is the problem.
>>>> 
>>>> I started paging through the openshift origin comprehensive
>>>> guide, and found this in my /var/log/activemq/activemq.log
>>>> while though through the activemq section.
>>>> 
>>>> 2013-11-18 22:24:43,995 [erSimpleAppMain] ERROR
>>>> XBeanBrokerFactory - Failed to load: class path resource
>>>> [activemq.xml], reason: Invalid bean definition with name 
>>>> 'org.apache.activemq.xbean.XBeanBrokerService#0' defined in
>>>> class path resource [activemq.xml]: Could not resolve
>>>> placeholder 'activemq.data' in string value
>>>> "${activemq.data}"
>>>> 
>>>> Sure, enough, activemq isnt running! [root strife activemq]#
>>>>  service activemq status ActiveMQ Broker is not running.
>>>> 
>>>> ----- Original Message -----
>>>>> From: "Luke Meyer" <lmeyer redhat com> To: "Nathan" 
>>>>> <lagern lafayette edu> Cc: users lists openshift redhat com
>>>>> Sent: Monday, November 18, 2013 2:22:17 PM Subject: Re:
>>>>> Openshift Origin, Proxy errors on
>>>>> /console/application_types.
>>>>> 
>>>>> Yep, that's probably the issue. The question is why... if
>>>>> mongod is actually listening and DNS is sending you to the
>>>>> right addr then you ought to get at least a speedy
>>>>> rejection if not success.
>>>>> 
>>>>> You might try it with telnet / sniff packets with tcpdump /
>>>>> check the mongod logs in /var/log/mongodb
>>>>> 
>>>>> ----- Original Message ----- From: "Nathan" 
>>>>> <lagern lafayette edu> To: "Luke Meyer" <lmeyer redhat com>
>>>>> Cc: users lists openshift redhat com Sent: Monday, November
>>>>> 18, 2013 1:12:00 PM Subject: Re: Openshift Origin, Proxy
>>>>> errors on /console/application_types.
>>>>> 
>>>> Finally having found a way to reproduce the connection
>>>> refused error. I started a netstat -anp via watch, and
>>>> watched for connections to show up that didnt look like they
>>>> were getting accepted, adn then ran the oo-accept-systems
>>>> command again.
>>>> 
>>>> I found this: tcp 0 0 10.1.1.35:20374 10.1.1.35:27017
>>>> TIME_WAIT
>>>> 
>>>> netstat -tulnp says thats mongod. tcp 0 0 10.1.1.35:27017
>>>> 0.0.0.0:* LISTEN 1710/mongod
>>>> 
>>>> 
>>>> Red herring? Or am I on to something?
>>>> 
>>>> 
>>>> 
>>>> On 11/18/2013 12:09 PM, Luke Meyer wrote:
>>>>>>> BTW http://127.0.0.1:8080/ will give you a 500 error
>>>>>>> (or with /broker) - this is normal, there's no route at
>>>>>>> those locations. But that error should come back
>>>>>>> instantly, not after a long timeout. Try this for a
>>>>>>> more representative URL:
>>>>>>> http://127.0.0.1:8080/broker/rest/api
>>>>>>> 
>>>>>>> Also run oo-accept-broker on the broker for more
>>>>>>> possible troubleshooting hints.
>>>>>>> 
>>>>>>> ----- Original Message ----- From: "Luke Meyer" 
>>>>>>> <lmeyer redhat com> To: "Nathan" <lagern lafayette edu>
>>>>>>> Cc: users lists openshift redhat com Sent: Monday,
>>>>>>> November 18, 2013 11:59:56 AM Subject: Re: Openshift
>>>>>>> Origin, Proxy errors on /console/application_types.
>>>>>>> 
>>>>>>> Check /var/log/openshift/broker/development.log and see
>>>>>>> if there are helpful error messages. Perhaps the
>>>>>>> connection to the DB or activemq isn't working. Or
>>>>>>> perhaps just as likely, mcollective is making a request
>>>>>>> for information from the nodes and not getting anything
>>>>>>> back (try out oo-accept-systems on the broker and
>>>>>>> follow the previous thread 
>>>>>>> https://lists.openshift.redhat.com/openshift-archives/users/2013-November/msg00079.html).
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> 
> ----- Original Message ----- From: "Nathan"
>>>>>>> <lagern lafayette edu> To: "Luke Meyer"
>>>>>>> <lmeyer redhat com> Cc:
>>>>>>> users lists openshift redhat com Sent: Monday, November
>>>>>>>  18, 2013 11:35:23 AM Subject: Re: Openshift Origin,
>>>>>>> Proxy errors on /console/application_types.
>>>>>>> 
>>>>>>> Thanks!
>>>>>>> 
>>>>>>> This is very helpful! I've done some troubleshooting.
>>>>>>> 
>>>>>>> First, the configs you pointed out all look sane.
>>>>>>> 
>>>>>>> If i curl -k http://127.0.0.1:8118/console, or 
>>>>>>> https://cloud.mydomain.com/broker/rest they both appear
>>>>>>> to work, or, at least, they dont error. They give me a
>>>>>>>  web-server error, stating that i havent authenticated,
>>>>>>> which I assume is normal.
>>>>>>> 
>>>>>>> 
>>>>>>> However, if i try to hit http://127.0.0.1:8080/broker,
>>>>>>> it sits forever (i assume it will eventually time out,
>>>>>>> though it's been a long time). Further along that line,
>>>>>>> if i take off /broker, it connects, but gives an
>>>>>>> internal server error.
>>>>>>> 
>>>>>>> Whatever is supposed to serve /broker seems to be my 
>>>>>>> problem. This gives me something to troubleshoot at
>>>>>>> least! thanks!
>>>>>>> 
>>>>>>> 
>>>>>>> On 11/18/2013 08:04 AM, Luke Meyer wrote:
>>>>>>>> The main httpd should proxy back to an internal
>>>>>>>> address for the console and broker. Look in 
>>>>>>>> /etc/httpd/conf.d/000002_openshift_origin_broker_proxy.conf
>>>>>>>>  for e.g. ProxyPass /console
>>>>>>>> http://127.0.0.1:8118/console ProxyPass /broker
>>>>>>>> http://127.0.0.1:8080/broker
>>>>>>> 
>>>>>>>> That's what it *should* look like.
>>>>>>> 
>>>>>>>> The other main "proxy" point is from the console to
>>>>>>>> the broker. Look in
>>>>>>>> /etc/openshift/console{,-dev}.conf which should have
>>>>>>>> something like: 
>>>>>>>> BROKER_URL=https://broker.example.com/broker/rest
>>>>>>> 
>>>>>>>> ... only it should be pointing to the name for your 
>>>>>>>> broker. Check that this address is correct, the host
>>>>>>>>  resolves in DNS, and is reachable from itself (curl
>>>>>>>> -k https://...). Something in that chain is probably
>>>>>>>> the issue. If the request gets hung up looking for
>>>>>>>> something that can't be retrieved, the proxy request
>>>>>>>> eventually times out and you get a 502.
>>>>>>> 
>>>>>>>> ----- Original Message ----- From: "Nathan T. Lager"
>>>>>>>>  <lagern lafayette edu> To: 
>>>>>>>> users lists openshift redhat com Sent: Sunday,
>>>>>>>> November 17, 2013 3:03:46 PM Subject: Openshift
>>>>>>>> Origin, Proxy errors on /console/application_types.
>>>>>>> 
>>>>>>>> I just installed a new origin, using the puppet
>>>>>>>> method. It appeared to complete without error, but
>>>>>>>> I'm having some troubles.
>>>>>>> 
>>>>>>>> First, /etc/openshift/htacces wasnt generated, i had
>>>>>>>> to do that on my own. Not a big deal.
>>>>>>> 
>>>>>>>> I created an admin user, and logged in, all appears
>>>>>>>> well, however, if i try to do.. well, just about
>>>>>>>> anything, i end up at a 502 proxy error after a very
>>>>>>>> long time.
>>>>>>> 
>>>>>>>> I see the following in my browser:
>>>>>>> 
>>>>>>>> Proxy Error
>>>>>>> 
>>>>>>>> The proxy server received an invalid response from an
>>>>>>>>  upstream server. The proxy server could not handle
>>>>>>>> the request GET /console/application_types.
>>>>>>> 
>>>>>>>> Reason: Error reading from remote server
>>>>>>> 
>>>>>>>> And a bunch of these in my logs.
>>>>>>> 
>>>>>>>> ==> openshift/broker/httpd/error_log <== connect to 
>>>>>>>> cloud.mydomain.com failed: Connection refused -
>>>>>>>> connect(2) will retry(#12) in 30.0 connect to
>>>>>>>> cloud.mydomain.com failed: Connection refused -
>>>>>>>> connect(2) will retry(#6673) in 30.0
>>>>>>> 
>>>>>>>> I cant figure out what its trying to connect to
>>>>>>>> though. Anyt insight here?
>>>>>>> 
>>>>>>> 
>>>>>>>> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
>>>>>>>> Nathan Lager, RHCSA, RHCE, RHCVA (#110-011-426)
>>>>>>>> System Administrator 11 Pardee Hall Lafayette
>>>>>>>> College, Easton, PA 18042
>>>>>>> 
>>>>>>> 
>>>>>>>> _______________________________________________ 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
>>>>>>> 
>>>>>>> 
>>>> 
>>>>> 
> 
> 
>> 
>> _______________________________________________ users mailing
>> list users lists openshift redhat com 
>> http://lists.openshift.redhat.com/openshiftmm/listinfo/users


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.15 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlKLgEoACgkQsZqG4IN3suk1PACeKOXrkEJWbBuhELLvjNfS1lP0
WfcAn3nHLsdjaV+AaYduII1nZN0Jt38k
=uCxp
-----END PGP SIGNATURE-----


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