About aborzenkov

aborzenkov · ‎2014-10-31

Indeed. You have 56 good disks (including spares) and one "broken". Do you have ASUP from the time before disk failure?

aborzenkov · ‎2014-10-31

Data ONTAP will not automatically grab a larger spare drive to replace a failed smaller disk because it would result in the larger spare disk being downsized in this case. Yes. And where is the problem? https://kb.netapp.com/support/index?page=content&id=3013616&locale=en_US

aborzenkov · ‎2014-10-31

Commands do not work because you have hardware based disk ownership, so disk assignment is moot. This is single node system so pool should not really matter either. Did you try manually starting "disk replace" for failed disk? Unfortunately to say anything one would need to fetch logs and analyze them. I suggest you open case with NetApp to investigate.

aborzenkov · ‎2014-10-31

Disk size is OK in this case - NetApp selects disk of same size or larger as replacement. Given that disk types and RPM seem to match, the only explanation would be different pool. Please, paste output of "disk show -v". Please use CODE tags so it is readable.

aborzenkov · ‎2014-10-30

Yes, it is possible to upgrade from A to AE. It will require downtime.

aborzenkov · ‎2014-10-30

Do not you need base cluster license first?

aborzenkov · ‎2014-10-29

There is already WFA workflow that does it.

aborzenkov · ‎2014-10-29

"Occupied space" (from memory, but close) in "lun show -v" output.

aborzenkov · ‎2014-10-29

Multinode cluster must have dedicated redundant interconnect over LAN. Internal HA interconnect cannot be used for this.

aborzenkov · ‎2014-10-29

Do not forget to set loader variable to indicate c-mode before reinstalling ... This is the main step 🙂

aborzenkov · ‎2014-10-29

There is 7MTT (7-Mode Transition Tool). Not sure whether it "better" but it helps to migrate filer configuration which is not possible with rsync.

aborzenkov · ‎2014-10-29

this WP says NVRAM caching uses 120B header for the incoming data. please read again. Consider a simple 8KB NFS write request. ... Data ONTAP merely has to log the 8KB of data

aborzenkov · ‎2014-10-29

NVRAM just have the journals/logs not the actual data Sorry, this is wrong. NVRAM contains actualt client requests as received from client.

aborzenkov · ‎2014-10-29

I have noticed that the /etc/passwd file is empty. It cannot be synced with the AD. Is there some options to find this. No. And never existed. /etc/passwd is for local users; if you want to use AD you need to turn on LDAP for user UNIX lookup.

aborzenkov · ‎2014-10-27

There is limit on number of concurrent replication streams, you may have hit it.

aborzenkov · ‎2014-10-27

- As I have an Autogrow setted to 3.0 TB may i rely on 320 GB + 500 GB = 820 Gb of space left? No. Volume cannot grow beyond free space available in contraining aggregate. - How do i resize the CIFS Volume? Does the process destroy data currently on the CIFS share or can it be done anywhere without problems?l It can be resized without problem. "vol size vol-name +500g".

aborzenkov · ‎2014-10-26

Disk reassign is possible either in maintenance or in takeover mode only.

aborzenkov · ‎2014-10-26

Each filer model has fixed hardware configuration; it is not possible to change it (except adding supported PCI adapters).

aborzenkov · ‎2014-10-23

the same goal That obviously depends on goal. For a start, QSM does not replicate existing snapshots nor can it be cascaded. Replicating non-qtree data is possible, but reverting such replication - not. Etc ... It is up to you to decide whether QSM can fulfill your requirements.

aborzenkov · ‎2014-10-22

No. But while in 8.1.x snapmirror from 64 to 32 is possible, I do not think it was ever considered as permanent solution - more as transition before migrating to 64 bit.

aborzenkov · ‎2014-10-22

VSM destination must be the same or higher version. It is not supported to do volume snapmirror to lower version.

aborzenkov · ‎2014-10-20

If under "VSM" you mean cluster mode SVM - yes, two SVMs are completely independent and use separate Kerberos setup. It is per SVM.

aborzenkov · ‎2014-10-18

It is manual using cf forcetakeover -d.

aborzenkov · ‎2014-10-15

Head swap is possible online using ARL (aggregate reallocation) since 8.2.1; unfortunately, document I have is marked as internal so you will need to contact your NetApp representative to discuss it.

aborzenkov · ‎2014-10-14

A user decided to abort a change to a file and begin the edit again by fetching an older version of the file from a snapshot. However, when he looked in the ~snapshot folder (accessing the volume from a Windows desktop) all the versions of the file in 6 different snapshot folders (4 x hourly + 2 x daily) were the same as the latest version of the file. It is not clear what "latest version" means nor when user started changing the file. So there simply is not enough facts to even start guessing.

Re: There are not enough spare disks. /aggr0/plex0/rg2:

Re: There are not enough spare disks. /aggr0/plex0/rg2:

Re: There are not enough spare disks. /aggr0/plex0/rg2:

Re: There are not enough spare disks. /aggr0/plex0/rg2:

Re: Does FAS3220A upgradable to FAS3220AE?

Re: About Simulator cDot 8.2 Licenses show

Re: 7MTT to Support LUNs

Re: block level reporting...

Re: Migrating FAS 2240-4 7mode filer to CDOT [SAN Attached]

Re: upgrade FAS2240 to cDoT

Re: Data migration from 7-mode to Cluster mode filers

Re: Quick NVRAM questions

Re: Quick NVRAM questions

Re: CIFS SID to UID , Linux permission denied.

Re: Snapmirror to multiple peers

Re: Extending CIFS Share

Re: old shelf -> new controller

Re: Memory Upgrade FAS3160?

Re: Snapmirror from OnTap 8.1.4P2 to OnTap 7.3.7P2

Re: Snapmirror from OnTap 8.1.4P2 to OnTap 7.3.7P2

Re: Snapmirror from OnTap 8.1.4P2 to OnTap 7.3.7P2

Re: Setup of NFS Kerberos Realm

Re: Metro Cluster Failover

Re: Expanding 2-node 3220 Cluster to 4-node 3220/8040 Cluster

Re: Multiple snapshot folders appear to have identical contents