Scheduled Snapshot create failed

[edp_citycarburoil]

Hi all,

I have a problem with snapshots.

I have a VOLUME (125 GB - 20% snapshot reserve) that contains a LUN (100 GB) for SQL data directly attacched to Windows Server 2008 x64.

Once I started to put data on that LUN, snapshot system told me that "SCHEDULED SNAPSHOT CREATE FAILED" and the most important error is:

"lun_SQL_data - hourly.0 : Snapshot operation failed: No space left on device."

I really don't know what can I do, please give me an advice!

Thankyou very much!

SCHEDULED SNA and I rea

[aliyazici · Accepted solution]

Hi ,

Try

vol options lun_SQL_data fractional_reserve 5

then take a snapshot.

Regards.

[edp_citycarburoil]

I don't know what this string does but it worked

THANK YOU VERY MUCH!

[lukasz_borek]

Generally (following netapp best practice) :

vol_size = lun size * 2.2

fractional reserve = 100% and turn off snapshot reservation.

Now, when you take first snapshot it will take 100% lun size...you need additional space for overrides.

Have a look at : http://www.netapp.com/us/library/technical-reports/tr-3483.html

[edp_citycarburoil]

But now I don't need to do it anymore, right?

[edriel]

Hello,

Because you changed your fractional reserve from 100% to 5% there is a chance that you're LUN can go offline. With the change you say to the volume the there will be reserved 5% of the volume for changes (in other words when the LUN is changed for 5% this changes can be saved). However, when you have 6% or chances and you're volume is 100% full and the autogrow of the volume is off then you're volume goes offline.

You need to be carefull when changing your fractional reserve.

Regards,

Eelco

[aliyazici]

Elco is right,

When you decrease fractional_reserve , there is a chance LUN can go offline.

To prevent this issue, you can set "snap autodelete" and/or "vol autosize"

Example:

snap autodelete lun_SQL_data on

vol autosize lun_SQL_data -m 250g -i 2g on

To choose which one is first:

vol options lun_SQL_data tyr_first snap_delete

or

vol options lun_SQL_data tyr_first volume_grow

Regards.

[edp_citycarburoil]

Mhh I'm not sure if I understand right.

This command should grow the LUN maximum to 250GB by 2GB of each step?

vol autosize lun_SQL_data -m 250g -i 2g on

These commands tell netapp: "When the space reserve is full, TRY to delete or grow (depending on what command I run)" Right?

vol options lun_SQL_data tyr_first snap_delete

or

vol options lun_SQL_data tyr_first volume_grow

Thank you!

[edp_citycarburoil]

Just understood

RTFM!!!

Thank you all!

ps: Do I have to do it also on the other netapp? I think so...

[edp_citycarburoil]

***********OLD******************When I have to issue the same commands on the other netapp, how can I do that?

I think I have to issue a SNAPMIRROR BREAK and the resize the volume... right?

After that I can do a SNAPMIRROR INITIALIZE correct? ***********OLD******************

Everything ok

I done it by breaking snapmirror, disabling fixed size of volume and then with the previous command I changed the size after that... snapmirror initialize

thank you all guys (once again)

Message was edited by: edp_citycarburoil

[upendra]

Dear all,

Before some days i created 3 TB thik volume and from there i created 2 TB thik lun but from i chose the snap reserve option 0 and then from second day i getting daily below alert message.

HA Group Notification from storage  (SCHEDULED SNAPSHOT CREATE FAILED) CRITICAl.

please can anyone suggest, do i need to do stop this error...????????????/