Using and Creating Mount Points in SQL Server

  • k-335975 (11/10/2011)


    I just see the letter of mount point and not underlying disks in the instance list of above counters

    you're looking for volumes not disks. They are not actual disks within Windows, they are formatted volumes

    CDA (11/10/2011)


    So, the one limitation, that I still see with mount points, is still doing any kind of logical detection of free-space.

    Say I have a stub drive (drive letter) that's 500mb

    xp_fixeddrives will ONLY see the stub drive freespace; it doesn't matter if I have mount points of 2 TB of space underneath it.

    some vendor apps fail to install, saying there's not enough free space to complete installation based on this.

    the only way I've figured out around this is to drop down into powershell and gwmi win32_volume|where-object {$_.filesystem -match "ntfs"}|ft name,capacity,freespace

    Anyone know of a way (or i'd like sql to have something built-in) to figure this out within TSQL?

    you can either use

    wmic volume get name, capacity, "free space"

    or view through explorer but you'll need to add the extra columns in the explorer window first

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • I currently have an issue with Multiple Mount Points and SCOM monitoring our SQL Servers.

    This is how our configuration looks like

    Possible values for VolumeName along with current mount points are:

    \\?\Volume{196adec4-2a1c-11e0-96b6-806e6f6e6963} C:

    \\?\Volume{cd7bb9b7-2dd0-11e0-9724-d485645a492c} E:

    \\?\Volume{cd7bb9c5-2dd0-11e0-9724-d485645a492c} E:\OLAPDATA\DATA1 E:\SQLDATA\DATA1

    \\?\Volume{cd7bb9cc-2dd0-11e0-9724-d485645a492c} E:\OLAPDATA\DATA2 E:\SQLDATA\DATA2

    \\?\Volume{cd7bb9d3-2dd0-11e0-9724-d485645a492c} E:\OLAPDATA\DATA3 E:\SQLDATA\DATA3

    \\?\Volume{cd7bb9be-2dd0-11e0-9724-d485645a492c} E:\OLAPDATA\DATA4 E:\SQLDATA\DATA4

    \\?\Volume{85ef6b9b-9add-4e88-bbfb-c970696c866f} E:\BDATA

    \\?\Volume{196adec5-2a1c-11e0-96b6-806e6f6e6963} D:

    Because multiple MP point to the same Volume, SCOM seems to have an issue. Can anyone help here?

  • Great article Perry. I have a question about adding a new instance to a SQL failover cluster. I created a 2-node test environment using Srv 2008 R2, Windows Clustering and SQL 2008 R2 Ent. I followed your article closely and I was able to create the empty application (let's call it WebProd Storage), add the storage and dependencies, create mount points under one drive letter for the application. My question is when I am ready to install a new instance (let's call it WebDev), can I use the existing WebProd drive letter and mount some additional volumes or do I have to have a dedicated empty application for each instance I install, thus another drive letter and mount the new volumes under it?

  • kellithompson (12/9/2011)


    My question is when I am ready to install a new instance (let's call it WebDev), can I use the existing WebProd drive letter and mount some additional volumes or do I have to have a dedicated empty application for each instance I install, thus another drive letter and mount the new volumes under it?

    You can use the same root drive for all of them, however, for management purposes some would choose to create another root drive and letter. It's entirely up to you 😉

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • Perry Whittle (12/9/2011)


    kellithompson (12/9/2011)


    My question is when I am ready to install a new instance (let's call it WebDev), can I use the existing WebProd drive letter and mount some additional volumes or do I have to have a dedicated empty application for each instance I install, thus another drive letter and mount the new volumes under it?

    You can use the same root drive for all of them, however, for management purposes some would choose to create another root drive and letter. It's entirely up to you 😉

    Then I must be doing something wrong. I created 2 new luns, assigned them to WebProd Storage, created the dependencies for the root drive. When I get to the Cluster Resource Group screen in the SQL installation, the WebProd Storage is not available for me to select as a resource.

  • Hi sorry, my bad totally missed the cluster part. Yes for a new clustered instance you will need to add another root drive and letter to the new application group

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • Got it. Thanks...

  • A little piece of missing information: You describe some of the differences between MBR and GPT, but in my opinion you lack the most important one: MBR can only host partitions up to 2TB. Obviously, if you are trying to create a partition bigger than this, you need to have a GPT disk. More importantly though, if you think your partition can possibly be resized to more than 2TB in the future, you should go for GPT to allow for this.



    Ole Kristian Velstadbråten Bangås - Virinco - Facebook - Twitter

    Concatenating Row Values in Transact-SQL[/url]

  • okbangas (12/21/2011)


    A little piece of missing information: You describe some of the differences between MBR and GPT, but in my opinion you lack the most important one: MBR can only host partitions up to 2TB. Obviously, if you are trying to create a partition bigger than this, you need to have a GPT disk. More importantly though, if you think your partition can possibly be resized to more than 2TB in the future, you should go for GPT to allow for this.

    <sigh> please read the article.

    I make a recommendation that if your disk will be under 500GB and have 1 partition use MBR otherwise consider GPT. It's not meant to be a comprehensive guide on disk types that is why I include the link to that KB. I do however cover the important aspects regarding partition constraints.

    You're just being petty, good day 😉

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • I do definitely not want to be petty, I do however think that the 2TB partition limit is an important knowledge regarding GPT vs MBR, in most cases actually more imporant than then number of partitions you can create, which you do mention. But, if this is the kind of response which I can expect if I try to point out something which could have been better, then I'll better keep it for myself.

    That said, I do realize that I should have read the article more thoroughly, but I do still think that the 2TB limit should have been mentioned.



    Ole Kristian Velstadbråten Bangås - Virinco - Facebook - Twitter

    Concatenating Row Values in Transact-SQL[/url]

  • Perry,

    This is a little off topic, but I was wondering if you or someone else on this forum have a step-by-step article on how to add high-performance mirroring on a stand-alone server with a failover cluster as the primary?

  • Is the stand alone server part of the same domain as the cluster or is it in a workgroup.

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • Yes it is. It's off-site at another location, but it's the same domain. I think I know how to do it, I will give it a whirl today and post back if I have a problem. I do have another question. Let's say my current failover cluster is on SQL 2008 R2 Ent. Can I add another node and install SQL 2008 R2 Std? I want to be able to fail the existing instances on ent over to std. Can my cluster be mixed in that way? I know there are some technical differences between ENT and STD like partitioning, but we're not utilizing that.

  • kellithompson (1/20/2012)


    Yes it is. It's off-site at another location, but it's the same domain.

    That makes things a whole lot easier. Just set mirroring up as defined in BOL. I have a script which you can use if you want it.

    kellithompson (1/20/2012)


    Can I add another node and install SQL 2008 R2 Std? I want to be able to fail the existing instances on ent over to std. Can my cluster be mixed in that way? I know there are some technical differences between ENT and STD like partitioning, but we're not utilizing that.

    You know what never even tried it and can't think why you would want to. I'm sure it would fail and cause issues somewhere along the line so don't 😉

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • Perry Whittle (1/20/2012)


    kellithompson (1/20/2012)


    Yes it is. It's off-site at another location, but it's the same domain.

    That makes things a whole lot easier. Just set mirroring up as defined in BOL. I have a script which you can use if you want it.

    Thanks Perry, I would like a script. I just wanted to test if I could get a mirror to work between my clustered instance and an offsite instance. I followed the BOL in creating a new database on my primary. Created a table and added some rows. Backed up the DB and the log fille using the proper method. Restored both w/ NoRecovery. Made sure I could telnet to both primary and mirror server on my designated port. I cannot get past the famous 1418 error w/ severity 16, state 1, line 1. I have removed all the configurations and started over 3 times and I must be missing something simple because each time I get the same error. Let me know if you have any advice.

    Thanks in advance. KT

Viewing 15 posts - 16 through 30 (of 33 total)

You must be logged in to reply to this topic. Login to reply