I am a newbie and am just getting my feet wet on StarCluster and AWS
I have successfully installed 0.93.3 on Windows and associated it with my
AWS login. I can create clusters and I have used the documented way of
created an EBS store for use by my application.
My problem is that none of the instances in the cluster will mount the EBS
and the master complains about it having more than one partition. It was
created according to the method here:
http://web.mit.edu/star/cluster/docs/latest/manual/volumes.html#create-and-format-a-new-ebs-volume–
and to the best of my knowledge correctly so. I think I am doing all
the
things I am supposed to do, but obviously I am missing something. The
configuration file is attached but it does not seem to be causing
StarCluster any problems.
The runtime output is however not so great:
C:\Windows\System32>starcluster start imagingcluster
StarCluster - (
http://web.mit.edu/starcluster) (v. 0.93.3)
Software Tools for Academics and Researchers (STAR)
Please submit bug reports to starcluster_at_mit.edu
>>> Using default cluster template: smallcluster
>>> Validating cluster template settings...
>>> Cluster template settings are valid
>>> Starting cluster...
>>> Launching a 3-node cluster...
>>> Creating security group _at_sc-imagingcluster...
Reservation:r-953beddd
>>> Waiting for cluster to come up... (updating every 30s)
>>> Waiting for all nodes to be in a 'running' state...
3/3 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
100%
>>> Waiting for SSH to come up on all nodes...
3/3 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
100%
>>> Waiting for cluster to come up took 1.383 mins
>>> The master node is ec2-46-137-154-117.eu-west-1.compute.amazonaws.com
>>> Setting up the cluster...
>>> Attaching volume vol-ae98c1c6 to master node on /dev/sdz ...
>>> Configuring hostnames...
3/3 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
100%
!!! ERROR - volume has more than one partition, please specify which
partition t
o use (e.g. partition=0, partition=1, etc.) in the volume's config
>>> Creating cluster user: None (uid: 1001, gid: 1001)
3/3 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
100%
>>> Configuring scratch space for user(s): qadmin
3/3 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
100%
>>> Configuring /etc/hosts on each node
3/3 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
100%
>>> Starting NFS server on master
>>> Configuring NFS exports path(s):
/home /mr_data
>>> Mounting all NFS export path(s) on 2 worker node(s)
!!! ERROR - command 'mount /mr_data' failed with status 32 |
0%
!!! ERROR - command 'mount /mr_data' failed with status 32
2/2 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
100%
>>> Setting up NFS took 0.145 mins
>>> Configuring passwordless ssh for root
>>> Configuring passwordless ssh for qadmin
>>> Shutting down threads...
20/20 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
100%
>>> Configuring cluster took 0.889 mins
>>> Starting cluster took 2.375 mins
A clue could be that when I sshmaster in, the /dev looks very odd as sda
etc. are missing:
root_at_master:~# ls -l /dev/s*
lrwxrwxrwx 1 root root 8 2012-08-23 18:44 /dev/shm->/run/shm
crw------- 1 root root 10, 231 2012-08-23 18:44 /dev/snapshot
lrwxrwxrwx 1 root root 15 2012-08-23 18:44 /dev/stderr->
/proc/self/fd/2
lrwxrwxrwx 1 root root 15 2012-08-23 18:44 /dev/stdin->/proc/self/fd/0
lrwxrwxrwx 1 root root 15 2012-08-23 18:44 /dev/stdout ->
proc/self/fd/1
However is seems that the disks are not sda1/sdz but xvd1 and xvdz:
root_at_master:~# ls -l /dev/x*
ls -l /dev/x*
brw-rw---- 1 root disk 202, 1 2012-08-23 18:44 /dev/xvda1
brw-rw---- 1 root disk 202, 400 2012-08-23 18:45 /dev/xvdz
Would that trip up StarCluster?
BTW, does anybody know why the output from the ssh session via Python SSH
are all filled by garbage control characters? The outputs above I had to
clean manually for it to be readable) – remember: total newbie.
All and any help is much appreciated.
Regards,
Anders Nielsen
Received on Thu Aug 23 2012 - 15:11:14 EDT