StarCluster - Mailing List Archive

Re: bug report for updated Starcluster release 0.94.3 - Windows 7

From: Justin Riley <no email>
Date: Thu, 12 Dec 2013 13:59:31 -0500

Hi Emilia,

If you're using pip there's no need to download StarCluster by hand -
pip will handle fetching the downloads. However, by default pip will not
upgrade packages if they're already installed unless you tell it to do
so using the -U option. Please give the following command a try and let
me know if it works for you:

$ pip install -U starcluster
$ starcluster --version

(should be 0.94.3)

Thanks!

~Justin

On Thu, Dec 12, 2013 at 05:13:38PM +0000, Emilia Vynnycky wrote:
> Hello Justin
> Many thanks for the reply.  That's helpful to know.  I've since been able
> to launch further clusters and have not had similar problems.
>  
> However, I am a bit unclear about how to install the latest version of
> starcluster under Windows 7 if an earlier version has already been
> installed. I thought that I had installed the latest version, but the old
> version number keeps appearing.
>  
> In brief, I downloaded the file starcluster-0.93.tar.gz and unzipped to a
> folder on my hard disk.  I then went to the folder and ran pip install
> startcluster [note that easy install starcluster didn't work previously -
> see another post].  Is there a different method for upgrading the version
> under Windows 7? 
>  
> Thanks
> Emilia
>  
>  
>  
>  
>  
>  
>
> >>> Justin Riley <jtriley_at_MIT.EDU> 12/12/2013 16:56 >>>
> Hi Emilia,
>
> The exception occured because of a failure to properly fetch the
> instance's userdata. I'm going to update the code to raise a better
> error when this happens. If this happens again try waiting a few seconds
> and re-running your start command and passing the --no-create (-x)
> option. That will re-use the existing cluster instances and try again.
>
> Also, from the crash report you sent it looks like you're using 0.94.2.
> Could you please double check the version you're using and try upgrading
> again if it's not 0.94.3? Thanks!
>
> ~Justin
>
> On Wed, Nov 27, 2013 at 12:34:37PM +0000, Emilia Vynnycky wrote:
> >    Hello
> >    I've just installed the updated version of Starcluster (version
> 0.94.3).
> >    
> >    However, when I tried to access the cluster that I had set up in the
> >    previous version, I received the following bug report.  Could someone
> >    advise on what I should do please?   The cluster had been terminated
> >    before I ran this.
> >    
> >    In case it's relevant, I'm running this under Windows 7 and following
> >    previous problems (see posting dated 14/11/13), I used the command
> pip
> >    install starcluster to install the latest version.
> >    
> >    Thanks in advance
> >    
> >    Emilia
> >    
> >    ---------- SYSTEM INFO ----------
> >    StarCluster: 0.94.2
> >    Python: 2.7.5 (default, May 15 2013, 22:43:36) [MSC v.1500 32 bit
> (Intel)]
> >    Platform: Windows-7-6.1.7601-SP1
> >    boto: 2.18.0
> >    paramiko: 1.12.0
> >    Crypto: 2.6
> >    
> >    ---------- CRASH DETAILS ----------
> >    Command: starcluster-script.py start thibela-build
> >    
> >    2013-11-27 12:20:43,398 PID: 8868 config.py:567 - DEBUG - Loading
> config
> >    2013-11-27 12:20:43,413 PID: 8868 config.py:138 - DEBUG - Loading
> file:
> >    C:\Users\user\.starcluster\config
> >    2013-11-27 12:20:43,430 PID: 8868 awsutils.py:74 - DEBUG - creating
> >    self._conn w/ connection_authenticator kwargs = {'proxy_user': None,
> >    'proxy_pass': None, 'proxy_port': None, 'proxy': None, 'is_secure':
> True,
> >    'path': '/', 'region': None, 'validate_certs': True, 'port': None}
> >    2013-11-27 12:20:45,539 PID: 8868 start.py:197 - INFO - Using default
> >    cluster template: smallcluster
> >    2013-11-27 12:20:45,539 PID: 8868 cluster.py:1660 - INFO - Validating
> >    cluster template settings...
> >    2013-11-27 12:20:47,316 PID: 8868 cluster.py:982 - DEBUG - Launch
> map:
> >    node001 (ami: ami-9b9ebaf2, type: m1.small)...
> >    2013-11-27 12:20:47,316 PID: 8868 cluster.py:982 - DEBUG - Launch
> map:
> >    node001 (ami: ami-9b9ebaf2, type: m1.small)...
> >    2013-11-27 12:20:47,365 PID: 8868 cluster.py:801 - DEBUG - Userdata
> size
> >    in KB: 0.54
> >    2013-11-27 12:20:47,365 PID: 8868 cluster.py:1677 - INFO - Cluster
> >    template settings are valid
> >    2013-11-27 12:20:47,365 PID: 8868 cluster.py:1498 - INFO - Starting
> >    cluster...
> >    2013-11-27 12:20:47,365 PID: 8868 cluster.py:1008 - INFO - Launching
> a
> >    2-node cluster...
> >    2013-11-27 12:20:47,367 PID: 8868 cluster.py:982 - DEBUG - Launch
> map:
> >    node001 (ami: ami-9b9ebaf2, type: m1.small)...
> >    2013-11-27 12:20:47,367 PID: 8868 cluster.py:1036 - DEBUG - Launching
> >    master (ami: ami-9b9ebaf2, type: m1.small)
> >    2013-11-27 12:20:47,367 PID: 8868 cluster.py:1036 - DEBUG - Launching
> >    node001 (ami: ami-9b9ebaf2, type: m1.small)
> >    2013-11-27 12:20:47,551 PID: 8868 awsutils.py:247 - INFO - Creating
> >    security group _at_sc-thibela-build...
> >    2013-11-27 12:20:50,388 PID: 8868 cluster.py:801 - DEBUG - Userdata
> size
> >    in KB: 0.55
> >    2013-11-27 12:20:51,003 PID: 8868 awsutils.py:432 - DEBUG - Forcing
> >    delete_on_termination for AMI: ami-9b9ebaf2
> >    2013-11-27 12:20:51,959 PID: 8868 cluster.py:847 - INFO -
> >    Reservation:r-a44340dd
> >    2013-11-27 12:20:51,959 PID: 8868 utils.py:624 - INFO - Waiting for
> >    instances to propagate...
> >    2013-11-27 12:20:53,334 PID: 8868 cluster.py:1298 - INFO - Waiting
> for
> >    cluster to come up... (updating every 30s)
> >    2013-11-27 12:20:53,561 PID: 8868 cluster.py:1255 - INFO - Waiting
> for all
> >    nodes to be in a 'running' state...
> >    2013-11-27 12:20:53,971 PID: 8868 cluster.py:711 - DEBUG - existing
> nodes:
> >    {}
> >    2013-11-27 12:20:53,971 PID: 8868 cluster.py:719 - DEBUG - adding
> node
> >    i-da130da1 to self._nodes list
> >    2013-11-27 12:20:54,890 PID: 8868 cluster.py:719 - DEBUG - adding
> node
> >    i-c4130dbf to self._nodes list
> >    2013-11-27 12:20:55,198 PID: 8868 cli.py:307 - ERROR - Unhandled
> exception
> >    occured
> >    Traceback (most recent call last):
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\cli.py",
> >    line 274, in main
> >        sc.execute(args)
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\commands\start.py",
> >    line 215, in execute
> >        validate_running=validate_running)
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\cluster.py",
> >    line 1485, in start
> >        return self._start(create=create, create_only=create_only)
> >      File "<string>", line 2, in _start
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\utils.py",
> >    line 111, in wrap_f
> >        res = func(*arg, **kargs)
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\cluster.py",
> >    line 1508, in _start
> >        self.setup_cluster()
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\cluster.py",
> >    line 1516, in setup_cluster
> >        self.wait_for_cluster()
> >      File "<string>", line 2, in wait_for_cluster
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\utils.py",
> >    line 111, in wrap_f
> >        res = func(*arg, **kargs)
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\cluster.py",
> >    line 1301, in wait_for_cluster
> >        self.wait_for_running_instances()
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\cluster.py",
> >    line 1256, in wait_for_running_instances
> >        nodes = nodes or self.get_nodes_or_raise()
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\cluster.py",
> >    line 731, in get_nodes_or_raise
> >        nodes = self.nodes
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\cluster.py",
> >    line 721, in nodes
> >        if n.is_master():
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\node.py",
> >    line 884, in is_master
> >        return self.alias == "master"
> >      File
> >   
> "C:\Python27\lib\site-packages\starcluster-0.94.2-py2.7.egg\starcluster\node.py",
> >    line 140, in alias
> >        aliases = aliasestxt.splitlines()[2:]
> >    AttributeError: 'NoneType' object has no attribute 'splitlines'
>
> > _______________________________________________
> > StarCluster mailing list
> > StarCluster_at_mit.edu
> > [1]http://mailman.mit.edu/mailman/listinfo/starcluster
>
> References
>
> Visible links
> 1. http://mailman.mit.edu/mailman/listinfo/starcluster




Received on Thu Dec 12 2013 - 13:59:33 EST
This archive was generated by hypermail 2.3.0.

Search:

Sort all by:

Date

Month

Thread

Author

Subject