Re: ALUA Configuration
That is correct. I had host A running ESXi 5.1 VMW_SATP_LSI, and host B running ESXi 5.5 VMW_SATP_ALUA -- both hosts were connected to same controller in an HA (dual path) configuration. Host B keeps...
View ArticleRe: Unable to unmount
vfk I've already checked all the pre-reqs in that KB article but it hasn't fixed the problem. RegardsEoin
View ArticleRe: Unable to unmount
I presume you have tried to unmount this datastore a number of times. I find that sometimes they will unmount from all but one host but then when I try to unmount a second time it then unmounts from...
View ArticleRe: SMB 2.1 and SMB 3.0
It most scenarios you will not see performance improvement between SMB 2.0 and 3.1, unless you are using:SMB Scale-Out SMB Direct (SMB 3.0 over RDMA) SMB Multichannel Directory Leasing BranchCache...
View ArticleBizarre growth of disk usage when moving thin provisioned VMs between Datastores
Example: CentOS-VM with 20GB Thin Provisioned Virtual Disk on Nexenta-Hosted NFS-Datastore:Used Space as seen in vShpere Client: 2,36 GB Clone of the VM through vCenter Server to another...
View ArticleRe: Safest way to delete storage
Just to make it clear, VMFS partition can be deleted from vCenter side, but the LUN itself has to be unmapped & destroyed on the storage side anyway. Regards,Radek
View ArticleClik here to view.
Re: Safest way to delete storage
Thank you Radek. To be very honest, I am still a bit confused. I have read some blogs tonight that suggest deleting a datastore in VCenter can still cause an APD. To me that does not make any sense. I...
View ArticleRe: Safest way to delete storage
I think it makes sense to me though:1) unmount the datastore if you want to get rid of the LUN (followed by deleting it on the array side)2) delete the datastore if you want to get rid of its content,...
View ArticleRe: Safest way to delete storage
I would also like to add, that if you do the steps to Unmount: 1. Unmount Lun2. Detach Device3. Have your SAN then revoke access to your hosts4. re-scan your HBAs there is a step #5 that I only see...
View ArticleRe: Bizarre growth of disk usage when moving thin provisioned VMs between...
That is bizarre, NFS by default it this provisioned datastore, when you clone the VM it should retain the format as the source vdisk, thin provisioned, unless you have changed it during the clone. What...
View ArticleRe: Bizarre growth of disk usage when moving thin provisioned VMs between...
Found the answer which is as simple as: compression of the zfs-folder on the storage server. I had (falsely) assumed storage-side compression/deduplication would be transparent from a vsphere...
View ArticleRe: Safest way to delete storage
Thank you very much Radek. I will proceed with the Unmount/Detach method.
View ArticleClik here to view.
VNX SAN CPU hike between 4.15 AM - 4.45 AM
Hi All, ESXi Host version - VMware ESXi 5.1.0 build-1612806 We use EMC VNX FC SAN for our prod. cluster and we observer SAN CPU hike goes up to 75% during 4.15 AM - 4.45 AM ONLY. We are more concern...
View ArticleRe: VNX SAN CPU hike between 4.15 AM - 4.45 AM
Possibly anti-virus scan or virus definition updates, or other agent downloading updates???
View ArticleRe: VNX SAN CPU hike between 4.15 AM - 4.45 AM
Thanks for your reply. But I forgot to mention that I worked with Anti virus vendor to exclude Windows page file, Exchange DB's and SQL DB's from the scan. We have to keep this in mind that this...
View ArticleDell Compellent and using the UNMAP primitive
I have many Luns which need to have space reclaimed, and the UNMAP does not seem to work due to how the Dell Compellents work. And I really don't want to go through deleting and recreating over 50+...
View ArticleRe: Dell Compellent and using the UNMAP primitive
Take a look at this blog post: Storage: Starting Thin and Staying Thin with VAAI UNMAP » boche.net – VMware vEvangelist
View ArticleRe: Dell Compellent and using the UNMAP primitive
I believe I may have read the blog before. Looking at his graphs, it appears he is doing the unmap *before* the Compellent takes any Replays(snapshots), i.e. moves his Active data to Read-only replay....
View ArticleRe: VNX SAN CPU hike between 4.15 AM - 4.45 AM
This is a half hour windows, I would suggest performing esxtop in replay mode for that time windows using vm-support, and this way you can replay the esxtop and examine which lun/vm is generating the...
View ArticleRAID System
May I know the maximum number of disks that RAID 0,1,2,3,4,5,6 systems use? Thanks,M
View Article