StarCluster - Mailing List Archive

Re: parallel ssh support?

From: Ryan G <no email>
Date: Thu, 16 Apr 2015 13:37:24 -0400

Turns out this is supported by StarCluster in the config file using
USERDATA_SCRIPTS

On Thu, Apr 16, 2015 at 10:52 AM, Ryan G <ngsbioinformatics_at_gmail.com>
wrote:

> I've discovered AWS allows you to upload user data aka a script that is
> run when an instance us brought up. Perhaps this is the way to configure
> in parallel. I'll look into this more if someone hasn't already.
>
> On Mon, Apr 13, 2015 at 2:53 PM, MacMullan, Hugh <
> hughmac_at_wharton.upenn.edu> wrote:
>
>> I concur. Threading for this? It's pretty non-intensive, just
>> time-consuming. And adding/removing nodes can be pretty painful, as well,
>> as it monkeys (serially) with /etc/hosts, keys, etc.
>>
>>
>>
>> -Hugh
>>
>>
>>
>> *From:* starcluster-bounces_at_mit.edu [mailto:starcluster-bounces_at_mit.edu] *On
>> Behalf Of *Ryan G
>> *Sent:* Monday, April 13, 2015 2:43 PM
>> *To:* starcluster_at_mit.edu
>> *Subject:* [StarCluster] parallel ssh support?
>>
>>
>>
>> Hi all - Whenever I launch a cluster with a large # of nodes, I have to
>> wait quite a bit of time because each node has to be configured
>> separately. The commands in my custom plugin are the same regardless of
>> machine. I was wondering if there was any support in StarCluster to use
>> parallel-ssh to run the plugin on all the nodes at once? This could save a
>> lot of start-up time....
>>
>> Ryan
>>
>> _______________________________________________
>> StarCluster mailing list
>> StarCluster_at_mit.edu
>> http://mailman.mit.edu/mailman/listinfo/starcluster
>>
>>
>
Received on Thu Apr 16 2015 - 13:37:27 EDT
This archive was generated by hypermail 2.3.0.

Search:

Sort all by:

Date

Month

Thread

Author

Subject