Blog :: Configuration

Expanding Scrutinizer’s disk: how to add more storage to your deployment

mikel

So you have added storage to your fresh Scrutinizer deployment. Or you’re giving your existing Scrutinizer deployment extra space. If you add a second disk, this is a straightforward process. Our awesome developers have come up with an easy way for expanding Scrutinizer’s disk.

Scrut_util.exe
Set partitions /dev/(disk)

Example:

Set partitions /dev/sdb

This will automatically go through everything for you.

But what if you were to expand your existing disk? This is a little more involved. If you would like help, there is a link at the bottom of this blog that will take you to our Support page. Feel free to open a ticket. Now let’s begin…

To get started, verify that the disk space is there.

fdisk –l
Expanding Scrutinizer's disk--verify available disk space
Fdisk showing 214gb in but only 2 partitions, sda1 and sda2, where sda2 ends at 108gb

We will need to add a third partition. For this, we’ll use GNU Parted.

Partitioning

To start, run Parted.

parted /dev/sdb

To print the existing partitions, enter print.

 print
Expanding Scrutinizer's disk--printing existing paritions

Now make a new partition. To do this, enter:

mkpart

This will run through a wizard. It will ask you 4 things.

  1. Partition type? (Primary partition)
  2. File system type? (ext4)
  3. Start? In this case you would enter 108gb, but if you are adding another partition again, look at your fdisk –l and see where the previous partition ends.
  4. End? -0 will use all available space.
Making a new partition

Once done, type print again.

 Print

If everything looks good, type quit and then reboot Scrutinizer.

 shutdown –r now 

Expand the volume group

Once you reboot, add the new partition to the volume group using vgextend.

vgextend vg_scrut /dev/sda3

You can use vgs to show the volume group.

vgs vg_scrut
Using vgs to show the volume group

Now we can see that we have 99.42GB free to apply to the database volume. All but .01 can be used. To apply this:

lvextend –-size +99.41g /dev/mapper/vg_scrut-lv_db

And to check to see if it worked:

lvdisplay /dev/mapper/vg_scrut-lv_db

Home stretch!

Now that we have applied the additional storage to the database, we need to unmount, check the disk for errors, then resize the file system.

Let’s start by stopping all running services.

scrut_util.exe
services all stop

Next, what I like to do is run a df –h and unmount these paths.

unmount /var/db/fast/scrut_tmp/collector
unmount /var/db/fast/pg_stat_tmp 
unmount /dev/mapper/vg_scrut-lv_db 
Unmount the paths

Then we need to check the disk for errors. We use fsck to check the disk integrity.

fsck –f /dev/mapper/vg_scrut-lv_db

Just one last thing… we need to resize now using resize2fs.

resize2fs /dev/mapper/vg_scrut-lv_db

That’s it!

Remount and start all the services.

mount –a
scrut_util.exe
services all start 

If you’re still reading this, it means one of two things. Either you were able to add more storage on your own, or you came across a hiccup. If you came across a hiccup, please open a support ticket to reach out to myself and the rest of the support team.