Most often it means that another instance of RPM is already running (notice, it could also be something like background upgrade checker, not necessary rpm command only). This is usually transient condition and goes away automatically. If you verified that no other invocation of RPM-related programs is active, it could mean unclean shutdown of previous RPM run (like kill -9). This leaves RPM database inside of open transaction. Files __db.000 and others are effectively Berkely DB transaction logs. Those files are normally simply removed on reboot, so brute force is to just manually do the same (rm /var/lib/rpm/__db.*). More prudent approach is to try to recover database; if you have Berkely db-utils installed, try "db_recover -h /var/lib/rpm/"; this should replay logs and remove them. This could be db45_recover or similar, because multiple versions of db-utils can coexist. Finally the worst case is when installation script tries to invoke RPM from within another RPM invocation. But I would expect this will be seen on every system then, so hopefully it is not the case. ... View more

The main difference is that you can keep more snapshots on destination. But usually snapmirror is for disaster recovery (to ensure data replication); for backup snapvault is more appropriate. ... View more

VSM is exact mirror of source anyway. So if you update every hour, at 11:00 you will have the same corruption dated 10:30 on destination too. You can simply revert to snapshot earlier than 10:30 on source as long as you still have it. VSM is not really intended for such sort of things. To protect against logical corruption local snapshots are just fine. ... View more

Is it VSM or QSM? ... View more

I was about to suggest the same ☺ I do not expect anything bad. Flash is sometimes completely reformatted during normal upgrade as well. Also before advent of UFM, using PC to reformat CF card and copy Data ONTAP binary was more or less sanctioned action. Thank you for testing and sharing you experience! ... View more

As guarantee is set to “none”, allocated size does not really matter from space consumption PoV. It could be safety measure in case your source volume grows (just guessing). ... View more

Well … I presumed “download” step had been performed because 7G kernel seems to be present on boot media, just environment variables are skewed. But yes, knowing exact procedure that was followed would help. ... View more

Did you try set-defaults at loader prompt? Otherwise could you describe step by step procedure you followed to downgrade kernel (or better, terminal capture for it)? ... View more

Why do you think that space is reported incorrectly? Please show exact command and its output where this happens. ... View more

I mean, I do not understand, why you would want actual raid group size 11 while having raidsize option 15. This is just confusing to anyone looking at configuration. Yes, it is possible to increase raidsize at any time without downtime or data loss. ... View more

You can always increase raid group size later when you need to add more disks. Personally I think that having raidsize option different from actual raid group size is just confusing, but it is of course up to you how you handle it. ... View more

“aggr create aggr-name -r raid-size …” where raid-size is raid group size. Every disk not used for aggregate is spare disk. NetApp does not automatically reserves anything. It is up to you to use correct number of disks and leave enough spares. ... View more

Vtic means that initiator is logged in on partner and works via interconnect. ... View more

From technical PoV you must have FCP enabled and active on both controllers to get FCP failover for host. I am not sure whether you even will be able to properly configure FCP in HA pair without running FCP on both heads. Other protocols may work. I do not know because I never tested it. But the real question is support. As far as I know such configuration is not supported. If you really need it, you need contact your NetApp representative and support and request official statement whether this is supported. May be as special release (PVR). ... View more

NearStore were dedicated low(er) cost backup appliances. Today they no more exist as separate product. Free NearStore license simply enables more concurrent backup sessions (and I believe is prerequisite for A-SIS license as well) and can be installed on any FAS. ... View more

Both filers in HA pair need to have identical licenses. ... View more

8.1P2 was recently released and contains large number of performance related bug fixes. May be worth to try. ... View more

As indicated by http://support.netapp.com/NOW/cgi-bin/bol?Type=Detail&Display=585280, you also can reallocate directory. Wow! reallocate start -o -f -p /vol/<volname>/<dirname> ... View more

Disk auto-assignment is controlled by option disk.auto_assign. “disk assign all” assigns only unassigned disks. ... View more

This is covered in great details in TR-3490. ... View more

It does not really matter which cable you unplug first as long as you do it one at a time ☺ After physically attaching shelf you need to assign disks in new shelf to controller(s) using “disk assign”. It could be that disks will be assigned automatically if the whole stack is currently owned by single controller. ... View more

Current DSM does all necessary settings so host utilities are not required for it (IIRC they actually skip settings if DSM is detected). From DSM release notes: The Windows Host Utilities are no longer required. The Windows Host Utilities components that enable you to configure Hyper-V systems (mbralign.exe and LinuxGuestConfig.iso) are now included with the DSM. While no longer required, installing the Windows Host Utilities on the same host as the DSM is still supported. ... View more