Dar backup scripts still running on server – pending OS upgrade.

Well, after I pruned down the size a bit on the CRAB server.  Here’s a summary of why that had to be done:

BEFORE:

Filesystem                                                 Size  Used Avail Use% Mounted on
/dev/mapper/vg_cap-lv_root        50G   40G    7.3G  85%       /
tmpfs                                                           1.9G  204K  1.9G   1%        /dev/shm
/dev/sda1                                              485M   45M  415M  10%   /boot
/dev/mapper/vg_cap-lv_home  129G   67G   56G  55%       /home
/dev/sdb1                                              1.8T  856G  886G  50%  /mnt/storage
/dev/sdc1                                               1.8T  796G  945G  46% /mnt/backup

AFTER:

Filesystem                                              Size  Used Avail Use% Mounted on
/dev/mapper/vg_cap-lv_root      50G   21G   27G  45%     /
tmpfs                                                         1.9G  352K  1.9G   1%    /dev/shm
/dev/sda1                                           485M   45M  415M  10%  /boot
/dev/mapper/vg_cap-lv_home  129G   17G  106G  14%  /home
/dev/sdb1                                               1.8T  924G  818G  54% /mnt/storage
/dev/sdc1                                                1.8T  924G  818G  54% /mnt/backup

You will see the problem areas highlighted in red, and the end product in green.  I can probably still trim the system / partition a bit, but it’s MUCH better, and should pose no problems with the OS upgrading process.  The /boot partition will be fine. 😉

The pruning will be fairly ‘transparent’ to users as I used some pretty slick symbolic links to ‘move’ the content to the storage drive.

The dar backup script for the top level of the storage drive has run about three days, as it’s fairly large.  I expect it to complete sometime today, or overnight, and expect to start the upgrade with the Fedora 14 dvd.  If all goes well, I might just go ahead and push on with Fedora 15.  I really don’t foresee any issues through Fedora 16, but not sure with 17, as it moves a few directories like /sbin and /bin, merging their content under /usr/sbin and /usr/bin for compatibility.  Time will tell. 😉

Server OS drive ‘imaged’ today.

Today, I imaged the server’s OS drive, since the last image (not backup, but ‘image’) that I saved was a year ago.  Before I was able to boot the system with the Clonezilla boot disk, I had to replace the DVD Rom drive, since it was dead.  After that was finished, I upgraded the kernel and kernel headers.  The length of time the clonezilla image process took reminded me that I need to cut the size of / and /home.  😀  So, I’m starting on that one now.

Once that is finished to my satisfaction, I shall upgrade the Fedora 13 to 14 initially, and if things go well, to 15 and/or 16.  (I would like to get to at least 17 where the directories /bin, /sbin are ‘merged’ into /usr/bin/ and /usr/sbin/ for compatibility.)

Until I return to work on 8 Sep 2014, I would expect that there might be extended periods of server ‘downtime’ as I have decided that the best way to upgrade this legacy Fedora install would be from DVD media.

I will try to reduce inconvenience, but it may be noticeable.  😉

New Camera Server

CamServer VNC Screenshot.

This is the new camera server that was just put into service today.  Compaq Ultra Small Form Factor DC-7700, running at 3.40GHz with 3GB of RAM and a 160GB Sata Hdd.  It is backed up using Ghost 15.

I see no real need to attach a  display to this box, as I can remote into it and do anything that I need to.

Only things left to do, are to move it to the other servers (at the moment, it’s not where it’s going to end up), and get the other cameras up and running.  😀

See, give a computer geek time off, and he feels the ‘need’ to improve the network infrastructure, heh.  Let’s see: yesterday, I upgraded the CRAB server’s storage drive, today I set up the camera server…..  Thinking about possibly upgrading the server’s Fedora OS sometime soon.  😉

Server upgrade and cloning has begun!

Server

This morning, I removed the backup drive, installed the new drive in the hot swap bay, and started prepping the drive at about 0900 Hrs.  Instead of using fdisk to partition the disk, I used another tool that is a bit easier, more flexible, and should be included in most every modern Linux distributions:  gparted.

So, as root user, I did the following:

# parted /dev/sdc mkpart primary 1m 100%
Information: You may need to update /etc/fstab.

# parted /dev/sdc “unit b print”
Model: ATA ST2000DM001-1CH1 (scsi)
Disk /dev/sdc: 2000398934016B
Sector size (logical/physical): 512B/4096B
Partition Table: msdos

Number  Start             End                 Size                 Type     File system  Flags
1      1048576B  2000398843903B  2000397795328B   primary

Notice that the start is at the 1 Meg spot, and not at 63, the way fdisk defaults.  This ‘aligns’ the partition structure and should result in higher performance.  Now, it was time to create the filesystem.  Here is the output:

# mkfs.ext4 /dev/sdc1
mke2fs 1.41.10 (10-Feb-2009)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=1 blocks, Stripe width=0 blocks
122101760 inodes, 488378368 blocks
24418918 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=0
14905 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968,
102400000, 214990848

Writing inode tables: done
Creating journal (32768 blocks): done
Writing superblocks and filesystem accounting information: done

This filesystem will be automatically checked every 35 mounts or
180 days, whichever comes first.  Use tune2fs -c or -i to override.

By now, it was 0915 Hrs, and it was time to mount the new drive and start ‘cloning’ as before mentioned using ‘cp -rp’.

It’s now about 1015 Hrs, and the cp process has run an hour or so, and I see that it’s clipping right along at about a rate of 225GB/hour.  😉

# df -h

/dev/sdb1             917G  764G  107G  88% /mnt/storage
/dev/sdc1             1.8T  224G  1.5T  13% /mnt/new

At this rate, I don’t expect it to take more than a couple more hours.  Then, I will have to do the ‘disk swap dance’.  😉

Tech, music, and general ramblings…..