Comments for Research Computing @ USF http://rc.usf.edu/blog Because sometimes it takes more than a "high-end" PC to get the job done. Mon, 03 Nov 2008 05:10:21 -0500 http://wordpress.org/?v=2.8.4 hourly 1 Comment on Perl-Fu: Automating ILOM SSH Sessions by brs http://rc.usf.edu/blog/?p=10&cpage=1#comment-25 brs Mon, 03 Nov 2008 05:10:21 +0000 http://rc.usf.edu/blog/?p=10#comment-25 You wouldn't be using Net::Telnet to connect (or nc/telnet as the backend), you'd use SSH as the backend and Net::Telnet to handle console I/O. Its perfectly doable. You wouldn’t be using Net::Telnet to connect (or nc/telnet as the backend), you’d use SSH as the backend and Net::Telnet to handle console I/O. Its perfectly doable.

]]> Comment on Perl-Fu: Automating ILOM SSH Sessions by biffsocko http://rc.usf.edu/blog/?p=10&cpage=1#comment-24 biffsocko Sat, 01 Nov 2008 03:39:55 +0000 http://rc.usf.edu/blog/?p=10#comment-24 most networks don't allow telnet most networks don’t allow telnet

]]> Comment on Cluster Build-Out, 2008 by chow http://rc.usf.edu/blog/?p=11&cpage=1#comment-23 chow Fri, 17 Oct 2008 03:21:18 +0000 http://rc.usf.edu/blog/?p=11#comment-23 Cool, Nice to see some pics of that new server cluster. I like to think I helped a little with the "physical" computing aspect of the project.. =) Cool, Nice to see some pics of that new server cluster. I like to think I helped a little with the “physical” computing aspect of the project.. =)

]]> Comment on Perl-Fu: Automating ILOM SSH Sessions by aastaneh http://rc.usf.edu/blog/?p=10&cpage=1#comment-21 aastaneh Sat, 24 May 2008 23:26:28 +0000 http://rc.usf.edu/blog/?p=10#comment-21 More on this: Apparently you can also accomplish the same goal using Net::Telnet; but instead of using a perl sub as the child, the ssh binary is called instead. Our solution's advantage is that it's all perl-based. To each his own, I suppose. More on this:

Apparently you can also accomplish the same goal using Net::Telnet; but instead of using a perl sub as the child, the ssh binary is called instead. Our solution’s advantage is that it’s all perl-based. To each his own, I suppose.

]]> Comment on Librarian-NG: Solve your unresolved linkage dependancies FAST. by aastaneh http://rc.usf.edu/blog/?p=6&cpage=1#comment-19 aastaneh Mon, 28 Jan 2008 17:07:52 +0000 http://rc.usf.edu/blog/?p=6#comment-19 Yes it does, as of last Friday. I'll create one more update to it and that should be sufficient for a informal distribution. Packaging for Sourceforge will come later when more things are sorted out. Yes it does, as of last Friday. I’ll create one more update to it and that should be sufficient for a informal distribution. Packaging for Sourceforge will come later when more things are sorted out.

]]> Comment on Librarian-NG: Solve your unresolved linkage dependancies FAST. by brs http://rc.usf.edu/blog/?p=6&cpage=1#comment-10 brs Fri, 25 Jan 2008 00:38:18 +0000 http://rc.usf.edu/blog/?p=6#comment-10 Does that package include the fix for shared object files? Does that package include the fix for shared object files?

]]> Comment on Kickstart and ssh by aastaneh http://rc.usf.edu/blog/?p=3&cpage=1#comment-3 aastaneh Fri, 18 Jan 2008 16:29:01 +0000 http://rc.usf.edu/blog/?p=3#comment-3 This approach is highly effective when dealing with chassis with absolutely no PS2 plugs or VGA output. A big example of this is the Sun Bladecenter, which have neither. This is the only way to get into the innards of the blades during an install session. Another approach could be to spawn a shell in the installation environment, at which point VNC is an acceptable point of access. However, I have not yet found a means to do so. This approach is highly effective when dealing with chassis with absolutely no PS2 plugs or VGA output. A big example of this is the Sun Bladecenter, which have neither. This is the only way to get into the innards of the blades during an install session. Another approach could be to spawn a shell in the installation environment, at which point VNC is an acceptable point of access. However, I have not yet found a means to do so.

]]>