Issues with Rackspace - building roles fails often and unable to launch MySQL server

[Jonas Colmsjo]

I've tried a number of times to start MySQL and Percona servers in Rackspace. I've tried both CentOS and Ubuntu.

I've mainly tested Rackspace UK but also a few times in Rackspace US. 

Example of errors:

* Build apache and Percona server (got same error) - Received RebundleFailed event from server. Reason: 'int' object has no attribute 'encode'

* I've been able to build MySQL and Percona roles on Rackspace UK but I'm unable to launch them - they are stuck in status initializing:

root@f5f02425-1034-4221-b734-92e7accce9db:~# cat /var/log/scalarizr.log 

2012-01-04 19:40:59,951 - INFO - scalarizr - [pid: 11629] Starting scalarizr 0.7.147

2012-01-04 19:41:00,423 - INFO - scalarizr - [pid: 11638] Starting scalarizr 0.7.147

2012-01-04 19:41:00,424 - INFO - scalarizr.config - State: importing

2012-01-04 19:41:00,551 - INFO - scalarizr.handlers.mysqlproxy - Updating mysql-proxy backends list

2012-01-04 19:41:02,172 - INFO - scalarizr.messaging.p2p.consumer - Building message consumer server on 0.0.0.0:8013

2012-01-04 19:41:02,177 - INFO - scalarizr.handlers.lifecycle - Server will be imported into Scalr

2012-01-04 19:41:02,228 - INFO - scalarizr.snmp.agent - [pid: 11688] Starting SNMP server on 0.0.0.0:8014

2012-01-04 19:42:03,398 - INFO - scalarizr.config - State: rebundling

2012-01-04 19:42:04,391 - INFO - scalarizr.handlers.rebundle - Lookup server 31.222.138.197 on CloudServers

2012-01-04 19:42:05,282 - INFO - scalarizr.handlers.rebundle - Creating server image. server id: 10090223, image name: 'percona-ubuntu1004-1-20120104194203'

2012-01-04 19:42:07,093 - INFO - scalarizr.handlers.rebundle - Checking that image 10498954 is completed

2012-01-09 08:59:34,092 - INFO - scalarizr - [pid: 876] Starting scalarizr 0.7.148

2012-01-09 09:00:04,360 - INFO - scalarizr - Server was started after rebundle. Performing some cleanups

2012-01-09 09:00:04,361 - INFO - scalarizr.config - State: bootstrapping

2012-01-09 09:00:04,605 - INFO - scalarizr.handlers.mysqlproxy - Updating mysql-proxy backends list

2012-01-09 09:00:06,221 - INFO - scalarizr.messaging.p2p.consumer - Building message consumer server on 0.0.0.0:8013

2012-01-09 09:00:06,227 - INFO - scalarizr.handlers.lifecycle - Starting initialization

2012-01-09 09:00:06,529 - INFO - scalarizr.snmp.agent - [pid: 1085] Starting SNMP server on 0.0.0.0:8014

2012-01-09 09:00:07,204 - INFO - scalarizr.config - State: initializing

2012-01-09 09:00:15,063 - ERROR - scalarizr.handlers - MysqlProxyHandler accept() method failed with exception

2012-01-09 09:00:15,063 - ERROR - scalarizr.handlers - argument of type 'NoneType' is not iterable

Traceback (most recent call last):

  File "/usr/lib/pymodules/python2.6/scalarizr/handlers/__init__.py", line 160, in __call__

    if handler.accept(message, queue, **self._accept_kwargs):

  File "/usr/lib/pymodules/python2.6/scalarizr/handlers/mysqlproxy.py", line 131, in accept

    return mysql.BEHAVIOUR in message.behaviour and \

TypeError: argument of type 'NoneType' is not iterable

root@f5f02425-1034-4221-b734-92e7accce9db:~# 

 Any tips?

/BR

Jonas

[itdept]

I am also having a very big problem with rackspace roles being created. Let me know if you solve this.

[Dale-Kurt Murray]

Confirmed, I too am having this issue

[Nick Toursky]

Hi guys,

The problem described in the first message of this thread was fixed in the recent versions of Scalarizr.
If you still experience similar problems, please provide log entries which will help us to pin-point the problem

Thanks,
Nick

On Wed, Feb 1, 2012 at 14:52, Dale-Kurt Murray <dalekur...@gmail.com> wrote:

Confirmed, I too am having this issue

--
You received this message because you are subscribed to the Google Groups "scalr-discuss" group.

To view this discussion on the web visit https://groups.google.com/d/msg/scalr-discuss/-/iQrgcqOEorQJ.

To post to this group, send email to scalr-...@googlegroups.com.
To unsubscribe from this group, send email to scalr-discus...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/scalr-discuss?hl=en.

[itdept]

What information will help you?

My roles on rackspace are always stuck at "installing-software"

Then in the logs it just spams "Active"
2012-01-31 22:47:41
Server status: ACTIVE
2012-01-31 22:47:30
Server status: ACTIVE
2012-01-31 22:47:20
Installing software [ OK ]
2012-01-31 22:47:20
Starting importing to Scalr in background [ OK ]

I have tried looking in so many different places at the moment to try and locate why this is constantly failing. The only time I got it to work was when I build a cent os 5.5 machine with only MySQL selected.

If i say select Apache, on every distribution it gets stuck in this type of loop.

[itdept]

*Not cent 5.5 I meant cent os 5.6, sorry

[itdept]

I also see this error quite a lot

Unhandled exception occurred during processing. com.rackspace.cloud.servers.nucleus.client.exceptions.NucleusClientException: 500 Error: (HTTP 400)

[Nick Toursky]

And what is going on on the instance during this time? You can check /var/log/scalarizr.log.

--
You received this message because you are subscribed to the Google Groups "scalr-discuss" group.

To view this discussion on the web visit https://groups.google.com/d/msg/scalr-discuss/-/qLZXbC9kdOwJ.

[Dale-Kurt Murray]

@Nick there are no logs being created in /var/scalr/app/logs and there is no /var/log/scalr path.

When creating a role on Rackspace I get this from the web interface

Bundle Task - http://cl.ly/1C1x1f2E2Z0V2O280D3M
Bundle Task > Log - http://cl.ly/3n1A0B0R122T00061r3O

Even though the server instance is launched on Rackspace and active, nothing happens thereafter to continue the building of the role.

I checked the Core Settings to confirm that the event handler and IP address is setup correctly. At the moment there is no FQDN so I'm using the IP address only

[Igor Savchenko]

Seems like cronjobs are not running. Can you please double check that
cronjobs have been configured and constantly executed.

Regards,
Igor

> --
> You received this message because you are subscribed to the Google Groups
> "scalr-discuss" group.
> To view this discussion on the web visit

> https://groups.google.com/d/msg/scalr-discuss/-/78-vPNcFqTIJ.

[Dale-Kurt Murray]

I checked, you were correct the cronjobs were not running on the server. Problem fixed