Speech:Summer 2014 Mitchell Dezak


 * Home
 * Semesters
 * Summer 2014

Week Ending June 4th, 2014
Just went in to meet with Erol. We went over a few things that we will be doing over the course of the semester, but didn't get much done otherwise. It was good to get some planning done and get a look at the lab and what I will be working on.

Week Ending June 18th, 2014
For this week I went into the lab and began working on the hardware. Professor Jonas wants to take parts off of the extra servers next to the servers that we use for Speech to make them run at maximum power. To figure out how much power each one of the servers has, I am going through them one by one to determine what each machine is made up of. There are four machines. Each one of them is a Dell Poweredge 2650, so they are all the same when you are looking at them.

To determine the differences in the machines I found the amount of processing power they had. As it turns out, machine 3 had 3200 MHz. That is actually more than Caesar has, so Caesar could be seeing some changes soon. I also checked how many hard drives were in each machine. Some of the hard drives did not say clearly on them how many GB of memory were stored on them. This was a bit frustrating but its not all that important right now.

We also tried to boot up each machine but were unsuccessful. Common errors included:

Power PS1 (E08F1) RPM Fan 4 (E0412)

There were a couple other errors but these were the two that occurred on just about every machine. When an error occurs the screen appears orange. We want it to be blue.

My plan for the upcoming week is to work inside the machine and determine how many CPUs are in each one and begin oiling the fans. Somewhere along the line I would like to work out those errors, but for now this is more important and we want to get the parts from these machines over to the machines for Speech.

Week Ending June 25th, 2014
There was still some Inventory for the servers left to do from last week. I had not gone inside the servers to see what was in them. I still have some left. I went into the first two servers and checked how many CPUs were in them and how much RAM there was. Apparently there is not a lot. That did not come as much of a surprise.

The real task this week, along with the continued inventory, was oiling the fans. most of the fans in the first machine looked and felt as though they had been oiled already. In fact, I know that some of them had already been oiled. The second machine hadn't had any fans oiled yet. I did all of the fans in that one. There were seven in the second machine opposed to the five in the first machine. Because there is only one CPU in the first machine and two in the second.

Most of the fans barely needed any oiling, but it did help at least a little bit. I found online how the fans are numbered inside the Poweredge 2650.



With this I found which one was fan 4, which was giving an error. This fan was also barely spinning. It was in need of a good oiling. The oil helped it out quite a bit. It still does not spin as well as the other ones. The fan speed could have something to do with the error appearing. That goes with the fan error on all machines.

Next week I want to get into machines three and four. From there I will finish inventorying them from the inside. I will also finish oiling the fans in those two machines. I also want to try to figure out any errors that are appearing in the machines and try to get rid of those.

Week Ending July 2nd, 2014
I went into the last couple of machines and and finished documenting the inventory. Machine three had a little bit more RAM than the other machines did, but other than that, it was basically the same inventory wise. Both had two CPUs and seven fans.

I also finished oiling the fans in all of the machines. the first machine still had a couple of fans that needed oiling that I wasn't sure about so I went back and checked. The third machine had a couple of very bad fans inside of it. They were fans 3 and 4. I would consider them unusable. They would barely move before I oiled them. After I tried oiling them, there was hardly any improvement. There was a lot of rust inside of them ad they have obviously seen a lot of wear. This machine really just needs a couple of new fans.

This is the machine that is coming up with two fan errors, so it's pretty obvious with fans like that why those two fan errors are happening.

Week Ending July 9th, 2014
Couldn't Access the Server Room

Week Ending July 16th, 2014
Couldn't Access the Server Room

Week Ending July 23rd 2014
7/22- Haven't been able to get into the lab for awhile since I have not had access. But did get a key card yesterday and was able to to get inside again.

Began cleaning the servers starting from the bottom up. I ran into a problem right from the start. These servers are Dell Poweredge 1750s. The parts inside of them are different from the 2750s that I have been cleaning in the past weeks. The fans in particular, which need to be oiled, are much different. There is actually nowhere to oil these fans. At least nowhere that I could find. They also have two separate fans inside of them which is a different setup than the 2750s. Luckily, all the fans were in pretty good shape. They all moved pretty well.

I plan on continuing cleaning and whatever inventory I have left this week in order to catch up on what I have missed the passed couple weeks.

Week Ending July 30th, 2014
Continued with the cleaning. The only servers I am not able to clean at the moment are Obelix and Caesar.

Jared is putting Fedora on the servers and is currently using Obelix right now. That isn't a problem. We still have a couple of weeks left and I am sure he is almost done with that. For now, I will just skip over that one and move onto Asterix.

Professor Jonas and I will also be using one or two of the 2750's that I cleaned already. It will depend on the shape that they are in. In my opinion, between the four of them there are enough good parts to put on two of them. I almost forgot about the orange screen that they had. I think new fans should take care of those. There were a couple of other errors to be figured out that I can work on as well.

Next week I need to figure out those errors that I just remembered as well as finish cleaning the last few servers, including Obelix if I can.

August 7-22

Finishing up with the server cleaning. Cleaned the last of the servers. The last ones I had to do this week were Automatix, Asterix, Obelix, and Caesar.

Caesar I went in and did last. I made sure with Professor Jonas that it was okay to turn off because it is the main server. It is a Dell 2650 so it is just like the 4 servers that I cleaned not too long ago that we plan on using in the future. All the fans in it were in great condition, but could use some maintenance of course.

As for the other servers, we won't be using those just yet. We will be using the two best ones that are only having fan issues, or at least one of them.

My inventory for the other servers can be found here: http://foss.unh.edu/projects/index.php/Speech:Summer_2014_Hardware#Inventory