use the BMC FAS200 did not have any BMC. Unfortunately. I could use some at the time ... View more

Probably not worth new thread ... in the ATTO PDF I read: Go to the ATTO FibreBridge Description page on the NetApp Support Site at support.netapp.com a. Access the ATTO Web site using the link provided Could you please provide direct link to this "ATTO FibreBridge Description" page on NOW. I cannot find it on NOW; the only material available is the single PDF document we are discussing. BTW, Lisa, the E-Mail link you provided is somewhat messed up ... View more

How exactly do you set it? Could you copy'n'paste command and output? The smallest possible file on NetApp is 4K, so you should be able to set number of files to number of 4K blocks which is around 268,435,456 for 1TB volume. I have customers with more than 50M files in a volume. You can't access volume via FCP/iSCSI. If you have license, you can create LUN though. ... View more

There was discussion on toasters about shelf unplug and the answer (from NetApp employee) was - the only safe way is with both heads down. It may appear to work with takeover/giveback but you are still risking panic after days or weeks. Now, moving shelf connection to another adapter basically means "unplugging" all shelves on old one. So I am not sure. I would simply switch off to be on safe side. ... View more

This could be possible with QSM. QSM allows you to have independent snapshot schedules on primary and secondary. Actually you must have snapshot schedule on secondary to actually preserve more than just the latest baseline snapshot. It is not possible with VSM. VSM replicates the whole volume with all existing snapshots. Destination is read-only, so no additional snapshots can be created. Destination is always exact image of source at the time of last fully transferred snapshot. ... View more

Shelf hot plugging is supported and I have done it multiple times. ... View more

cf.takeover.on_disk_shelf_miscompare is ignored for software based disk ownership (should should be default today). So you are correct – as long as filer does not have online aggregate on the shelf, it should not panic and imitate failover. ... View more

SATA shelves are supported in stretch MetroCluster. Rules are the same – SATA and FC must be on different loops. Check MC support matrix on NOW for more information. ... View more

But that assumes power was cut off completely. What happens if power is returned after shutdown started? So that filer never completely loses power? ... View more

In this case you will have to recreate all aggregate and volume configuration manually. Share configuration is part of root volume so – assuming you created the needed same volumes – this will be restored automatically once root volume is restored. You will need to manually configure access to your DRP filer in NetWorker server as well. Also do not forget that volume options (like fractional reserve) will not be restored, you need to configure them manually. I am not sure about snapshot settings (like snap reserve) to be honest, you may need to test it. Rough sequence as I see it - Configure volumes as you need. They need not be identical to original filer, actually, as long as paths required for shares exist - Configure filer in NetWorker - Restore need volumes back; do not restore root volume as yet - Restore root volume and reboot It likely may need one or more corrections as you go ☺ ... View more

You can and this is described in Active/Active guide. You connect one adapter to In port on channel A on one end of stack and another adapter to Out port on channel B on another end of stack. ... View more

I briefly reviewed Active/Active guide and I do not see where they mention “Loop A” at all ☺ They speak about connecting adapters to shelf channel A or B. But you are right in that A or B applies to shelf. ... View more

OK, so you do have software based disk ownership. In this case my original statement holds true – it does not matter (technically) how you connect shelves, filer always finds own disks. The only consideration is consistency and ease of maintenance. In case of SANOWN enabled nothing should be displayed. You can verify it be checking whether command “disk show” exists at all. If SANOWN is disabled, this command is disabled as well. “Loops” in this context means all shelves connected to an FC port. You can daisy chain multiple shelves, supported number is 6 on single loop. With advent of MP-HA and SAS notion of “loop” became a bit blurry, so what is often used today (in NetApp documentation) is “stack”. It has the same meaning – it means shelves which are daisy chained and connected to the same filer adapter(s). Each adapter forms a single loop. Each shelf has two channels – channel A and B. With single path connection you connect FC port on filer to channel A (although as mentioned with SANOWN it does not really matter). With multipath connection you connect one FC port on filer to channel A and another FC port on the same filer to channel B. So you have two “loops” A and B connected to the same “stack” of shelves. Each stack has own loops, connected to different FC ports. In your case you have 4 loops, forming 4 single path stacks each consisting of single shelf. ... View more

Actually now when I looked more closely on “storage disk show -p” output, half of FC ports on each head are connected to channel A and half of ports are connected to channel B. And sorry for confusion, you are right. For hardware based disk ownership (default on FAS3000) disk ownership is indeed determined by channel; filer connected to channel A owns all disks in a shelf. I have worked with software based disk ownership for so long that I tend to forget it. ☺ ... View more

You have four loops. Whether head is connected to channel A or channel B is completely irrelevant as long as you keep it consistent and it suits you. In the times of hardware based disk ownership and single path connections I tended to connect one head to channel A and another head to channel B, exactly as had been done in your case. Personally I’d rather implement MP-HA by creating 2 stacks, one FC and one SATA, and enable software based disk ownership. It needs the same amount of FC ports but gives better resiliency. But it may require additional equipment (cables and/or SFP). ... View more

FAS3020 does not support SAS shelves. You probably mean FC. There is no terminate switches since ESH2 modules. You have normal single path connection. I would do it differently but this is working configuration. ... View more

Could you describe in more details your DRP environment? Do you have some (tape) backup and try to restore it on completely blank filer? ... View more

Was not release model changed so that today RC correspond to earlier GA and today GA correspond to earlier GD? From  "PB DataONTAP Release Model": RC releases are ready for installation on production systems. The GA designation simply implies that all NetApp specific Data ONTAP certifications with client applications and platforms are completed and the release has customer usage with no critical product defects. ... View more

For some reasons it finds old version of host utilities. It could be the cause of your problem. ... View more

You seem to have old host utilities version. According to IMT SDU 4.2 for AIX requires at least HU 4.2. ... View more

Are WWPW of local adapters correctly found? What is result of sanlun fcp show adapter -v sanlun fcp show adapter -c And on filer igroup show -v <the name of your AIX igroup ... View more