Speech:Spring 2016 Systems Group


 * Home
 * Semesters
 * Spring 2016
 * Proposal
 * Report
 * Information - General Project Information
 * Experiments - List of speech experiments

Groups

 * [Systems Group]
 * Experiment Group
 * Tools Group
 * Data Group
 * Modeling Group

Group Member Logs

 * Michael Salem
 * Neil Champagne
 * Saverna Ahmad
 * Aaron Miller

In-House Useful Resources for the Systems Group

 * Fall 2014 Redhat notes by Forrest
 * 2015 Systems Group Log
 * 2015 Systems Group Proposal
 * Software specs/configs
 * Hardware specs/configs

Tasks

 * Remove every Dell PowerEdge 1750 server in the rack except for Caesar, Idefix, and Obelix
 * Install five Dell PowerEdge 1950 servers to replace the older units
 * Install Redhat on the five new PE1950s
 * Configure the networking on the new servers in the same fashion as the previous servers. The new names are as follows
 * Majestix
 * Miraculix
 * Obelix2
 * Idefix2
 * Asterix
 * Mount Caesar's NFS on the new servers except for Majestix, which will NOT have NFS mounted
 * Configure Rome to be IRC Server and backup server

Week 1
Emailed Bruce to fix DNS issues. Working on getting Slack up and running to allow easy collaboration amongst group members. Received feedback and know which servers we should not disconnect. Our group is more familiar with RedHat and Linux.

Week 2

 * Get access to server room
 * Remove/replace 1750s with 1950s

This week was when we got familiar with the equipment in the server room and began the removal of the older units. Confusion arose almost immediately upon looking at the rack, as the labeling for the units was almost entirely inaccurate. Luckily, we didn't remove any units until we fully verified which unit was which. Aaron logged in as root on Caesar, and ran this command: dmidecode|grep –A3. This command gave us some crucial information on the nature of the CPU, which allowed us to match it with a server model.

Week 3

 * Get access to server room
 * Remove/replace 1750s with 1950s
 * Correct labeling
 * Install Redhat

This week was more hands on activity with following our plan to change the servers. All of the new servers were installed and labeled. Four of the new units displayed a ROMB Battery warning on the front display. The actual error was E1211 which indicates a RAID battery error. Once the 1750s had been replaced by the new units, we began the installation of Redhat. This process went perfectly on Asterix, and failed miserably on all other servers. Majestix, Idefix, and Miraculix were not booting at all. The error given was "disk not found". Currently, Red Hat is only running on Asterix. Our task for the upcoming week is to have the batteries changed and install Red Hat on the rest of the servers.

Results: FOBS access was given for the server room which needs to be picked up at the security desk.

Week 4

 * Troubleshoot ROMB errors and Obelix's CLI-only problem.
 * Install Redhat

It turns out that ROMB low battery was not the cause of the servers failing to boot. The problem was that the disks being used in the servers all have previous or foreign configs on them. These need to be cleared before an operating system can be installed on these servers. This process is detailed on the Hardware Setup Page. This problem took ages to figure out. I'm glad we finally overcame that obstacle. Obelix was installed with a CLI-only. This is a problem. Reinstallation of this server has begun. Installation of Redhat on the other servers has also begun.

Week 5

 * Install Redhat on Miraculix and Majestix
 * Configure networking on Asterix

Networking is actually very straightforward, and went well on Asterix following the software setup documentation Installation has completed on Obelix, beginning installs on the other servers this week.

Week 6

 * Setup Networking on the remaining servers
 * Begin mounting Caesar to the drones

Now that Redhat is finally installed, we can begin the networking configuration for the remaining four servers. Trying to figure out how mounting Caesar will work... Very worried due to the delicacy of the entire operation.

Week 7

 * Mounting NFS

We finally dug through the previous semesters logs, and found that Forrest Suprenant has a fantastic guide with info on NFS mounting. The mounting process from here was easy.