December 13, 2016 at 3:33 pm
Hi,
SQL 2014 STD Edition.
We have been trying to set-up Log Shipping on a 2 node cluster which is using Clustered Shared Volumes for the storage configuration.
The issue is we can only get the TLOG backups to go one or other of the physical cluster node names as the Network path to backup folder destination.
Obviously in the event of a fail over on the secondary site, it means we have to manually change the log shipping to point to the other physical node.
We want to use the SQL Virtual host name, so that it doesn't matter which node is active, but this does not seem to be able to work.
So insteqad of using
\\SQLP1\C$\ClusterStorage\SQLBACKUPS or \\SQLP2\C$\ClusterStorage\SQLBACKUPS if we failover
I want to use
\\SQLV1\C$\ClusterStorage\SQLBACKUPS which is the virtual name that is active.
Hope it makes sense! Appreciate any help.
Cheers
December 13, 2016 at 4:10 pm
You cannot have a local drive defined as a share in the cluster. You need to create a new drive - add that drive to the cluster - then create a shared folder on that drive that can be used for your log backups.
I would also caution you are using the hidden admin share (c$, d$, e$, etc...) as those require the account access the folder to have an account on that server. Normally you would be running SQL Server with a domain service account that is not a member of the local administrators group.
Jeffrey Williams
“We are all faced with a series of great opportunities brilliantly disguised as impossible situations.”
― Charles R. Swindoll
How to post questions to get better answers faster
Managing Transaction Logs
December 13, 2016 at 4:25 pm
Jeffrey Williams 3188 (12/13/2016)
You cannot have a local drive defined as a share in the cluster. You need to create a new drive - add that drive to the cluster - then create a shared folder on that drive that can be used for your log backups.I would also caution you are using the hidden admin share (c$, d$, e$, etc...) as those require the account access the folder to have an account on that server. Normally you would be running SQL Server with a domain service account that is not a member of the local administrators group.
Thanks for the reply Jeffrey.
This isnt actually a local drive. C$\ClusterStorage\SQLBACKUPS is how we have to reference the Clustered Shared Volume.
SQLBACKUPS is like a link to the CSV and is how Windows references it and how it shows up under Windows Explorer.
I have added a screenshot to try and show what I mean.
I do however, take your point about the domain service account!
Cheers
December 14, 2016 at 12:31 pm
This does not look like a Microsoft Failover Cluster configuration.
The file system on those files make me suspect you are using some type of clustering file system that allows access to the same storage from multiple nodes. If that is the case then you should have that same folder as a mount point off the C:\ drive from all nodes in the cluster.
Instead of trying to access that folder through a UNC path and the C$ hidden admin share you should be able to just use C:\SQLBACKUPS\...
from either node.
To be able to access from outside the cluster based on the currently active node - then you need to create a shared folder in Cluster admin that references the mounted volume. Not sure how to do that with your clustering software...
Jeffrey Williams
“We are all faced with a series of great opportunities brilliantly disguised as impossible situations.”
― Charles R. Swindoll
How to post questions to get better answers faster
Managing Transaction Logs
December 14, 2016 at 4:39 pm
Jeffrey Williams 3188 (12/14/2016)
This does not look like a Microsoft Failover Cluster configuration.The file system on those files make me suspect you are using some type of clustering file system that allows access to the same storage from multiple nodes. If that is the case then you should have that same folder as a mount point off the C:\ drive from all nodes in the cluster.
Instead of trying to access that folder through a UNC path and the C$ hidden admin share you should be able to just use C:\SQLBACKUPS\...
from either node.
To be able to access from outside the cluster based on the currently active node - then you need to create a shared folder in Cluster admin that references the mounted volume. Not sure how to do that with your clustering software...
Hi Jeffrey,
It is a Failover Cluster, we are using clustered-shared-volumes as described [/url]
I can't use C:\SQLBACKUPS as I need to reference the fail over cluster which is on a different server to the primary.
I actually want to use the SQL Virtual Name, but am having to use the Physical host name (1 of the 2 physical nodes) which will obviously mean I have to change this if the cluster fails over to the other physical node.
I see this as as a bug or limitation in Clustered Shared Volumes (which is supported with SQL 2014 onwards) and Windows Failover Clusters?
Regards
January 4, 2017 at 5:01 am
Cluster Shared Volumes are a feature of Windows Server and integrate with Windows Server Failover Clusters, hence their name of Cluster Shared Volumes.
They were originally employed to provide storage configurations for Hyper-V cluster nodes, they have been expanded to support other Windows Server applications, SQL Server from 2014 on.
There are quite possibly many bugs and certainly still limitations, have you checked the CSV documentation for creating and provisioning shares
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
January 4, 2017 at 4:23 pm
Perry Whittle (1/4/2017)
Cluster Shared Volumes are a feature of Windows Server and integrate with Windows Server Failover Clusters, hence their name of Cluster Shared Volumes.They were originally employed to provide storage configurations for Hyper-V cluster nodes, they have been expanded to support other Windows Server applications, SQL Server from 2014 on.
There are quite possibly many bugs and certainly still limitations, have you checked the CSV documentation for creating and provisioning shares
Hi Perry,
Yes, I appreciate they are new, particularly with regards to SQL server. I was unable to find anything specifically about this limitation
as the CSV documentation is aimed more at Windows rather than SQL server.
Thanks for replying.
January 5, 2017 at 6:38 am
somone else asked about CSV's in this post
If you want to have a shared location made available then you should create a shared folder cluster resource and place into either a new cluster role or an existing cluster role
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
January 5, 2017 at 4:44 pm
Perry Whittle (1/5/2017)
somone else asked about CSV's in this postIf you want to have a shared location made available then you should create a shared folder cluster resource and place into either a new cluster role or an existing cluster role
Hi Perry,
The issue isn't creating the shared folder, its getting SQL Server to recognise the SQL Virtual host name as part of the shared location so that failover doesn't require any configuration changes to log shipping.
At the moment, it will only accept the physical node name which means if there is failover at the receiving cluster, we have to change the log shipping configuration to point to the other physical node.
Regards
May 28, 2020 at 11:41 pm
I realize I'm over 3 years too late to help the original poster, but had this issue and may have insight to new people. I'm using Windows Server 2016 and SQL Server 2019, cluster shared volumes, and log shipping. Using File explorer, I navigated to the mounted volume, opened properties and went to the "sharing" tab. This revealed a network path that can be used whichever node is owner.
May 29, 2020 at 12:58 am
Thanks!
I was the original poster and 3 years later this info is still useful to know. Thanks for posting!
May 29, 2020 at 4:27 am
This was removed by the editor as SPAM
May 29, 2020 at 8:26 am
under windows 2016 CIFS shares have changed quite a bit, you now need to setup a new file server role and then provision a redundant fileshare from there.
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
Viewing 13 posts - 1 through 12 (of 12 total)
You must be logged in to reply to this topic. Login to reply