VMware Cloud Community
loflof
Contributor
Contributor
Jump to solution

Adding a physical hard disk to ESX 3 server

Hi,

We are quite new to the ESX 3. (hosted on a PowerEdge 2950) & have quickly discovered that there wasn't enough disk in the server.

Currently this server uses 4x300GB SCSI disks in a HW RAID5 array.

We would like to add a new 300GB hard disk, however there is no documentation out there. We have been made aware that the process may not be as simple as adding a physical hard disk to a native Linux box.

Would anyone know where we can obtain a step by step method on how to do this safely?

Thanks in advance.

Loflof

Reply
0 Kudos
1 Solution

Accepted Solutions
msmenne17
Enthusiast
Enthusiast
Jump to solution

- You mention hardware controls (with regards to

adding the new disk to the RAID5 as more LUN space),

I presume that these controls can be accessed at boot

time? - Are the risks of data loss using this method

high?

There are always risks of doing anything. The risk here is relatively low as manufacturers have built in tools to catch errors. That said, it is possible to spread corruption without the raid controller catching it. I've seen it happen and the results are disasterous.

- Wouldn't recreating the LUN destroy the data

presently on the server?

That is correct. All of the data on the RAID set would have to be backed up and restored.

- What method would I use If I decided to install

this new disk as an independent entity (not part of

the present RAID 5 array)?

You would simply create a new VMFS volume. I would recommend this. You would then have a volume with NO redundancy in case of a failure.

Thanks

Loflof

Michael

View solution in original post

Reply
0 Kudos
3 Replies
Texiwill
Leadership
Leadership
Jump to solution

Hello,

This depends on what you want to do. If you want to add it to the RAID5 as more LUN space then you can do that using the hardware controls.

If you want to increase the space of your VMFS after doing the first, that is another story and you should not increase the space but possibly use extents within ESX or recreate the VMFS LUN. I like the later actually.

Please give more details on how the you want to use the Drive and the partitioning of the Raid 5? Is this iSCSI or SAN, or just local storage?

Best regards,

Edward

--
Edward L. Haletky
vExpert XIV: 2009-2023,
VMTN Community Moderator
vSphere Upgrade Saga: https://www.astroarch.com/blogs
GitHub Repo: https://github.com/Texiwill
Reply
0 Kudos
loflof
Contributor
Contributor
Jump to solution

Thanks for you response:

\- It's all local storage. (there are at present 5 disks of 300GB each in a RAID 5) array

\- You mention hardware controls (with regards to adding the new disk to the RAID5 as more LUN space), I presume that these controls can be accessed at boot time? - Are the risks of data loss using this method high?

\- Wouldn't recreating the LUN destroy the data presently on the server?

\- What method would I use If I decided to install this new disk as an independent entity (not part of the present RAID 5 array)?

Thanks

Loflof

Reply
0 Kudos
msmenne17
Enthusiast
Enthusiast
Jump to solution

- You mention hardware controls (with regards to

adding the new disk to the RAID5 as more LUN space),

I presume that these controls can be accessed at boot

time? - Are the risks of data loss using this method

high?

There are always risks of doing anything. The risk here is relatively low as manufacturers have built in tools to catch errors. That said, it is possible to spread corruption without the raid controller catching it. I've seen it happen and the results are disasterous.

- Wouldn't recreating the LUN destroy the data

presently on the server?

That is correct. All of the data on the RAID set would have to be backed up and restored.

- What method would I use If I decided to install

this new disk as an independent entity (not part of

the present RAID 5 array)?

You would simply create a new VMFS volume. I would recommend this. You would then have a volume with NO redundancy in case of a failure.

Thanks

Loflof

Michael

Reply
0 Kudos