In this case you probably want to look at Metro Cluster. I do not think it is supported otherwise. ... View more

What interface secondary will use is determined entirely by routing table on secondary. This is just standard TCP/IP connection using the same rules. ... View more

Yes, you can migrate any volume non-disruptively to any other aggregate. Alternative is to setup LS mirrors for root volumes and promote mirror copy to be new root. LS mirrors are best practice anyway. ... View more

i believe your consultant set the environment up wrong.   There is no need to create snapvault sched's on the primary.   That's not what documentation says and not how I was told SV works. You must create snapshots on primary, because they are what secondary transfers. Schedules on secondary only say when and which snapshots to transfer - they do not cause snapshots themselves be created on primary.   Now you of course can have different retention policy on primary and secondary, but you still need at least one snapshot of each kind (hourly, nightly, weekly etc) that you want to transfer. I'm not aware of option to "create snapshot, transfer, delete snapshot". ... View more

Efficiency (if you mean deduplication and/or compression) is per volume, not per LUN.   For thin provisioning disable space reservation: [-space-reserve {enabled|disabled}] ... View more

This could well be an ASUP bug (it happened in the past). Open case with NetApp and ask them to verify. ... View more

As always: Disk Shelf Firmware ... View more

If you come around to doing it, please post a note here. Thank you! ... View more

Yes to both. You can find details about supported configurations on HWU, as usual. ... View more

Wow! Hat off ... Care to document it (blog or similar)? ... View more

Do you have a link to description how to do it? ... View more

It is possible only in 7.x version (not sure if it is still available). 8.x version does not support HA pair anymore.  ... View more

SnapDrive uses Single File Snap Restore (SFSR) for ISCSI RDM lun restore, and SFSR is quite fast. Hmm ... that's not what I experienced. In my experience SFSR was always painfully slow. Other users commented on this as well. It is possible that it has improved in recent Data ONTAP versions though. ... View more

Again - if takeover was successful, HA configuration is OK (at least to extent you need it). Takeover does not happen because of HA problems - HA problems may cause failures in takeover, not takeover itself. You need to investigate why takeover happens.  ... View more

Not sure I understand the question. You connect new controller and install it. Why do you need to bother with old controllers that you are going to replace right away? ... View more

It is not clear - is takeover successful or not? If it is successful, your HA configuration is obviously OK and you need to troubleshoot what is causing takeover. ... View more

The cluster interconnects don't really failover Nitpicking - they do failover inside of single node to another cluster interconnect port(s). Not sure about switchless cluster, but if one LAN controller loses physical connectivity, second should do as well so both should failover. ... View more

Max SATA raid group size is 20 disks, so 23 may not be possible. ... View more

If you are going to replace controller in near future, it makes no sense (from my point of view) to buy licenses for old controller. It may be possible to transfer them, but you have to speak with NetApp about it, and better in advance. Normally each license is issued to specific serial number.   I have no idea how you contact NetApp in your region, did you try to look at www.netapp.com for contact information? ... View more

You always need licenses associated with controller serial number to remain compliant. Starting with 8.2 licenses are also physically tied to serial number, so after head swap you have 60 days of grace period during which you must install matching licenses. ... View more

What is in your nsswitch.conf? Is filer set to uset /etc/passwd for users and in which order? ... View more

1. Create a new snapmirror relationship to transfer the volume. 2. Once completed, remove the relationship. You need to stop access from client for the duration of final snapmirror update to ensure all changes are transferred to new volume. If you can afford it, "cifs terminate" is the most simple way to do it. 3. Then remove the old share and create a new share on the new volume. If you stopped CIFS anyway in previous step, you could try to simply rename volumes (rename old volume away and give new volume old name) then start CIFS. I think that if CIFS is not running volume names in share definitions won't be updated in this case so when CIFS start it will pick up new volume. If not, you have nothing to lose 🙂 Then I need to reallocate the free space of the old volume to another volume: Controller 1: Vol_01 to the Controller 2: Vol_01. ¿How can i do this? You can't relocate part of aggregate. If old volume was the only one in aggregate, then destroy volume and aggregate, zero spares, unassign disks on old controller (disk assign -s unowned), assign disks on new controller. Make sure to unset disk.auto_assign option while doing it, otherwise disks may be unexpectedly assigned back. ... View more

Did you try to set filer management interface in SD (snapdrive config set -mgmtpath)? See "Multiple subnet configuration" in SD manual. ... View more