LincolnBlogs

I had the occasion to need some extra filespace on the LVM’d root partition of a RHEL(or CentOS) vm. This is how to do it:

First off, VMWare allowed me to create a second HDD on the fly while the vm was running (YAY!)

Once that was done, I rooted into the server and:

# echo “- – -” > /sys/class/scsi_host/host#/scan
(partprobe should also do the trick ??)

# fdisk -l
(Just to see that the new disk is available – in this case /dev/sdb)

# fdisk /dev/sdb
(create a new partition here)

# mkfs.ext3 /dev/sdb1

# vgs
(list the volume groups here)

# pvcreate /dev/sdb1
(add new physical volume)

# vgextend VolGroup00 /dev/sdb1
(extend my default volume group from the vgs command)

# vgs
(check to see pv and vg has another volume now)

# lvextend /dev/VolGroup00/LogVol00 /dev/sdb1
(extend my / volume by the entire size of /dev/sdb1)

# resize2fs /dev/VolGroup00/LogVol00
(resize filesystem to match vol size increase)
(requires a 2.6 kernel to resize while fs running)

That was it! And remember that all this was done on the fly, on the root partition and filesystem, WHILE the vm was running.

Everyone wants to know what kinds of crazy hardware a computer geek has kicking around his house. If you are actually reading this post, I know your curiosity will drive you to find out. Here is what this geek uses at home.

A few weeks ago I saw that an Alpha of Google Chrome was available on Linux. I could hardly wait to try it so I quickly loaded it onto my Ubuntu 8.04 box and took it straight out for a spin. First impressions were that it was clunky looking and I immediately noticed that it would not handle html authorization. That is something I NEED to use (fail), so I just filed it away and went back to using my beloved Firefox.

Fast forward to today when I noticed once again that there was a package update for GC and I decided that since I had seen a few of those go through I should probably give it another spin. All I can say is “WOW”. This thing is shaping up to be absolutely fantastic. It’s blazingly fast and handles things with speed and grace that Firefox barfs on. The only thing I have had any trouble with on it so far at all is an ornery java application I occasionally have to use, but then again, it was difficult to get it working under Firefox as well and I haven’t spent *any* time on trying to get it to work so far.

If you haven’t tried GC yet, you really should. You are missing out! You can find the directions for getting it at:
http://dev.chromium.org/getting-involved/dev-channel#TOC-Linux

You have heard me tell you before that cssh (Cluster ssh) is one of my most favorite tools as an admin. There is only 1 problem I have with it, and that was getting it installed on CentOS or RedHat. Not I mostly use it on Ubuntu, and that’s just a quickie apt-get install clusterssh away, but on RedHat and CentOS it’s VERY difficult to figure out how to get it going. You can find the package without too many troubles, but there are a pile of dependencies that seem hard to fulfill. Well, after some digging, here’s a sure-fire way to get it running on RedHat5.x and CentOS 5.x:

wget http://yum.trixbox.org/centos/5/old/perl-Tk-804.027-4.i386.rpm
wget http://packages.sw.be/perl-X11-Protocol/perl-X11-Protocol-0.56-1.el5.rf.noarch.rpm
wget http://dl.atrpms.net/all/clusterssh-3.21-4.el5.i386.rpm
rpm -Uvh perl-Tk-804.027-4.i386.rpm
rpm -Uvh perl-X11-Protocol-0.56-1.el5.rf.noarch.rpm
rpm -Uvh clusterssh-3.21-4.el5.i386.rpm

Do that on a command line as root and you’ll have a functional cssh. Now there is some wierd error that ends up on my console when this is running, but it does appear to function correctly. Have at it folks!

A Happy and safe Halloween to you all from the Fessendens!

Just got a note from a friend who is using the script now. Seems that I inadvertently left my own MAK in the curl script line. I just fixed and re-uploaded the file. All should be well now. Again, you can find the script at:

http://lincgeek.org/linc/docs/tivo/getmytivo.tgz

Thanks Allan!

By the way, while I was walking around the mall, I noticed that Suncoast Video (or at least my local one) has the 10 movie set of Star Trek movies (on dvd) for only $45 on sale. That’s a fantastic deal, and if you don’t yet have the full set, run out and get it right away before they are gone. My store only had 2 sets when I got there. Of course I felt obligated to pick one up

I bet you all can guess what I am watching today!

Happy new year everybody!

Yes, even I have a new years resolution, and it’s one that all of you should have too. This particular one is one that Dann and I know all too well…

DO YOUR BACKUPS!

You see I woke this morning to the horrible realization that I could not contact my web server, had no DNS resolution, no dhcp server and no mail among other things. These are all services that are handled by my trusty old thinkpad server. And it has been dutifully doing it’s job for 7 months now without a single backup

Lucky for me, a simple reboot started things working well again, however, it also brought me to the realization that I didn’t have any backups whatsoever on my now irreplaceable server. So, it was my friend and yours, tar, to the rescue.

I just happen to have a NAS here with some space on it, so I mounted that on my workstation and wrote a little shell script, that in part, looks a lot like this:

ssh root@server “tar czvf – /etc /home /var/www /var/lib/mysql” > `date +%Y%m%d`_server.tgz

What this does is ssh as root to my server and starts the tar process that is in the Parentheses. That tar command is making a compressed copy of each one of the filesystems listed there, and those are the important ones to me in this case, and sending this info through stdout back to the machine where the ssh command started. As the tar file makes it’s way back to my workstation, where I started the ssh command from, all that info is being redirected into a file on my NAS. The date command in there simply spits out, in this case, 20080101, and the rest of the filename is appended. So what I get is a file called 20080101_server.tgz and it gets all the tgz data from my server pumped into it. Viola, instant, or at least reasonably fast, backups.

You see, this is not very hard at all. It only takes a very few minutes of your time and is well worth the aggravation if you were to lose something important. It’s also important to note that I have, for a very long time, used the “round robin” approach to my home backups. That is to say that I save different machines backups to different machines. That way, if any single machine goes down, I will still have either a good working copy, or a good backup of that machine available. Backing everything up to one single location with no duplicate copies makes me a little nervous, but then again, I haven’t done *any* backups for the last 7 months.

So anyway, here is you wakeup call today. Take 20 minutes and DO YOUR BACKUPS! You’ll thank me later.