About aborzenkov

aborzenkov · ‎2012-04-09

The question is unrelated to Data ONTAP. You have to ask your operating system vendor whether re-exporting of NFS is supported. It is not supported with stock Linux kernel NFS implementation (although rumors are it is possible to make it work and that there are various user-space implementations that do it).

aborzenkov · ‎2012-04-08

The possibility to update (as opposed to manual uninstallation before installing new version) would be more user friendly ☺

aborzenkov · ‎2012-04-06

In 7G snapmirror.conf update usually took effect after couple of minutes (I believe, it was checked every minute). Under 8.0.2 snapmirror.conf changes apparently go unnoticed. I edited snapmirror.conf to disable schedule and continued to get messages on console about inability to transfer (why exactly is unrelated) every minute after that. Only after I quiesced and resumed transfer (QSM) did these messages stop. Is it a known issue? Is there BURT for it?

aborzenkov · ‎2012-04-06

At least messages from SnapMirror and some others appear twice. The real problem is, the messages are partially overlapped with text mixed up which makes it hard to read. Here is an example: Sat AprSat Apr 7 09:06:01 MSK last message repeated 2 times 7 09:06:01 MSK last message repeated 2 times Sat AprSat Apr 7 09:07:01 MSK [fasr3: replication.dst.err:error]: SnapMirror: destination transfer from sapdr-con:/vol/saplog_sns_r3p/SNS to /vol/saplog_sns_r3p/SNS : source contains no new data; suspending transfer to destination. 7 09:07:01 MSK [fasr3: replication.dst.err:error]: SnapMirror: destination transfer from sapdr-con:/vol/saplog_sns_r3p/SNS to /vol/saplog_sns_r3p/SNS : source contains no new data; suspending transfer to destination. This makes it rather hard to follow when you are sitting on console. Anyone has noticed it? What are the sources of these messages? Any BURT on it? Add. Checking /etc/messages - it contains no duplicated messages; I get them once per minute (as modified by throttle). /etc/syslog.conf does not exist (in DOT root at least). This is vanilla system factory preinstalled and updated to 8.0.2P6.

aborzenkov · ‎2012-04-06

I recently had question from customer who listed multiple mail serves in autosupport.mailhost but apparently only one of them was ever tried (and did not work, so no ASUP was sent). Are multiple mail servers supported? The version is 8.0.2 but would be interested for earlier versions as well.

aborzenkov · ‎2012-04-06

If you have snapshots on this volume, you may need to delete snapshots which still has “copies” of these LUNs.

aborzenkov · ‎2012-04-06

There is no procedure to reboot shelf. The only possibility is to power it off and on. This procedure is disruptive and requires that both controllers are shut down. You should open case with NetApp support and ask them to investigate the problem.

aborzenkov · ‎2012-04-05

The minimal size of root volume for different platforms is specified in Storage Management Guide. Additionally, if you try to declare volume as root (vol options root) it will complaint if volume is too small. FlexVol can be then easily extended.

aborzenkov · ‎2012-04-05

So you have here 360+G your LUN occupies. We found it ☺

aborzenkov · ‎2012-04-05

More likely it is due to “A-SIS = 97377440KB”, which is “Amount of disk space used by deduplication metadata in the aggregate”. I bet on active deduplication ☺ Rahul, could you additionally show output of “df -s” (I always forget this option, sorry).

aborzenkov · ‎2012-04-05

I always was under impression, that “Occupied size” is exactly the actual amount of data on LUN. I may be wrong; do you know what this field means?

aborzenkov · ‎2012-04-05

Interesting. Are you using deduplication on this volume?

aborzenkov · ‎2012-04-04

please, show ouput of lun show -v df -h df -rh aggr show_space

aborzenkov · ‎2012-04-04

Duplicated IP sounds like quite possible cause.

aborzenkov · ‎2012-04-04

snapshot is always volume based and includes everything on this volume.

aborzenkov · ‎2012-04-04

Last shelf needs to be connected for dual controller, yes. Single controller does not need it. And having 10 x 20m cables does not mean having 200m distance. I have seen enough configurations where long cables were ordered "just in case" because at the time of putting it together nobody could say where equipment would be actually located. Then you have to use whatever is available on site. (I remember couple of dozens 50m FC cables between adjacent cabinets. Just to find place to stash them away was the challenge ) And at the end the real question is, would NetApp support it. I have seen mix of FC and SATA shelves in one loop, shelves connected to non-shelf port on FAS270, copper controller-to-shelf connection. It worked

aborzenkov · ‎2012-04-04

What I have not found so far - is there any limit on total distance from head to shelf. Meaning - can you actually have 10 shelves stack each connected with 20m cable?

aborzenkov · ‎2012-04-04

Was it not what I was trying to tell you all the way long? ☺ Of course I have been working with NetApp for 9 years only, so I may easily miss something

aborzenkov · ‎2012-04-04

Almost every system I have seen has aggr0 and vol0. Never had any issues on takeover due to the same names. Go figure … ☺

aborzenkov · ‎2012-04-03

During takeover virtual copy of partner filer is started with all its resources (volumes, exports, IP addresses etc), so there is no confusion. Clients continue to access the same exported resources via the same IP addresses as before.

aborzenkov · ‎2012-04-03

Robert Reynolds wrote: I tried to assign all of the remaining SATA drives owned by S2 to S1 to have all the SATA drives on S1. However when I try to do this via the console it tells me that S2 owns the drives and cannot be changed. Do I need two root aggregates and root volumes, one for the SATA and the other for SAS? To achieve what you want, you need assign SAS drives to S2 create new aggregate from SAS drives on S2 move S2 root volume from SATA aggregate onto SAS aggregate (please search KB for procedure to do it) reboot S2 destroy SATA aggregate on S2 unassign SATA disks on S2 (disk assign -s unowned) assign them on S1 add to existing aggregate on S1 If these filers are new and do not have any user data yet, another possibility would be to reassign disks the way you want them in maintenance mode and simply reinstall both heads. Would be valuable experience as well

aborzenkov · ‎2012-04-03

No. SnapDrive is unrelated product and is not related to controller failover. Correct settings (on Windows) are implemented either by Host Utilities or by MPIO DSM package (at least, recent versions; I believe in the past Host Utilities was prerequisite).

aborzenkov · ‎2012-04-03

No, it does not require OSSV. https://kb.netapp.com/support/index?page=content&id=1011627

aborzenkov · ‎2012-04-03

Yes, it can work for QSM as I already said.

aborzenkov · ‎2012-04-02

http://support.netapp.com/eservice/toolchest?toolid=540

Re: Does ONTAP 7G support re-exporting a NFS mount from a client to additional servers?

Re: OnCommand System Manager 2.0.1 is available

snapmirror does not re-read snapmirror.conf under 8.0.2

Some console messages are duplicated under 8.0.2

Does NetApp use multiple mail servers for autosupport?

Re: Destroying LUN(s) on the volume does not free up the volume space

Re: FAS3020 FC shelf reboot

Re: Best practice way to make SATA drives with CIF shares available on the network

Re: Volume usable size and lun

Re: Volume usable size and lun

Re: Volume usable size and lun

Re: Volume usable size and lun

Re: Volume usable size and lun

Re: Odd networking behavior (OnTap 7.3.6) - arp ok, ping not okay

Re: How to take Snapshot of a LUN through NetApp API's

Re: extend SAS cables more than 5 meters

Re: extend SAS cables more than 5 meters

Re: Volume and Exports in HA Controller setup

Re: Volume and Exports in HA Controller setup

Re: Volume and Exports in HA Controller setup

Re: Best practice way to make SATA drives with CIF shares available on the network

Re: Cluster Failover (CF -Takeover and CF -Giveback)

Re: snapmirror base data transfer bast practice

Re: snapmirror base data transfer bast practice

Re: snapmirror base data transfer bast practice