CategoryNetApp

How to check if there is packet drop in Netapp filer

Command: netstat -I e4a -d

Output:
Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Collis Drop
e4a 9000 none none 2007724074 1 2991841358 0 0 0

 

0 is the one you are looking for.

Proper way to delete snapvault relationship

I have problem with my backup in Netapp Protection Manager, the backup using snapvault technology just hang and unable to run any backup at all. This is a bug according to them and will be fixed in 8.0.2(hopefully). I always wanted to know the proper way to remove snapvault relation shop and i am doing this today.
Note: create case in Netapp to assist you in case you are uncertain of anything, this guideline is just for information sharing and my own references.


########################################################
To list out the dfpm data set for job id(check from source snapshot)
########################################################
C:\Documents and Settings\admin1>dfpm dataset list -m 23953
Id Node Name Dataset Id Dataset Name Type Name
---------- -------------------- ---------- -------------------- --------------- -------------------------------------------------------
20538 filerSource 23953 VOL A Dataset new volume filerSource:/volumeA
23955 filerDest 23953 VOL A Dataset new volume filerDest:/volume_A_Dataset_new_backup
24400 filerDest 23953 VOL A Dataset new volume filerDest:/volume_A_Dataset_new_backup_1

C:\Documents and Settings\admin1>dfpm dataset list -x 23953
Id: 23953
Name: VOL A Dataset new
Policy: Backup Policy For PG t28
Description:
Owner:
Contact:
Volume Qtree Name Prefix:
DR Capable: No
Requires Non Disruptive Restore: No

Node details:

Node Name: filerSource
Resource Pools:
Provisioning Policy:
Time Zone:
DR Capable: No
vFiler:

Node Name: filerDest
Resource Pools: PG DRNAS BACKUP 02
Provisioning Policy: No Dedupe Policy for Secondary
Time Zone:
DR Capable: No
vFiler:

==========================

After

C:\Documents and Settings\admin1>dfpm dataset snapshot list 23953
Id Name Unique Id Volume Timestamp Versioned Dependencies % of Total Blocks
---- --------------- ------------ --------------- -------------------- ---------- --------------- ----------
75019592 dfpm_base(dataset-id-23953)conn1.3 1306675183 filerSource:/volumeA 29 May 2011 21:19:43 No SnapVault 33% (19%)
83770083 dfpm_base(dataset-id-23953)conn1.0 1310908962 filerSource:/volumeA 17 Jul 2011 21:22:42 No SnapVault,acs 13% (11%)
86355690 dfpm_base(dataset-id-25559)conn1.1 1311686499 filerSource:/volumeA 26 Jul 2011 21:21:39 No Busy - SnapVault 3% (2%)
86355689 dfpm_base(dataset-id-25559)conn1.0 1311772921 filerSource:/volumeA 27 Jul 2011 21:22:01 No Busy - SnapVault,acs 0% (0%)
86377105 nightly.0 1311782476 filerSource:/volumeA 28 Jul 2011 00:01:16 No None 0% (0%)

########################################################
To delete snapshot in source from OM databse
########################################################
C:\Documents and Settings\admin1>dfpm dataset snapshot delete 23953 "filerSource:/volumeA" 75019592
Dataset dry run results
----------------------------------
Do: Delete snapshots 'dfpm_base(dataset-id-23953)conn1.3' of volume filerSource:/volumeA (20538).
Effect: Selected snapshots will be deleted.
Following snapshots have applications dependent on them:
dfpm_base(dataset-id-23953)conn1.3 - SnapVault

Started job 13340 to delete snapshot(s) of the volume 'filerSource:/volumeA' of the dataset 'VOL A Dataset new' (23953).

C:\Documents and Settings\admin1>dfpm dataset snapshot delete 23953 "filerSource:/volumeA" 83770083
Dataset dry run results
----------------------------------
Do: Delete snapshots 'dfpm_base(dataset-id-23953)conn1.0' of volume filerSource:/volumeA (20538).
Effect: Selected snapshots will be deleted.
Following snapshots have applications dependent on them:
dfpm_base(dataset-id-23953)conn1.0 - SnapVault, acs

Started job 13341 to delete snapshot(s) of the volume 'filerSource:/volumeA' of the dataset 'VOL A Dataset new' (23953).

########################################################
To delete snapshot at destination, take few hours.
########################################################

[root@linux tiwlim]# rsh filerDest snapvault stop filerDest:/vol/volume_A_Dataset_new_backup_1/t28lp_ip
Stopping /vol/volume_A_Dataset_new_backup_1/t28lp_ip is permanent.
The secondary qtree will be deleted.
Further incremental updates will be impossible.
Data already stored in snapshots will not be deleted.
This may take a long time to complete.
Are you sure you want to do this? y
Snapvault configuration for the qtree has been deleted.

ndmpcopy duration

I have to migrate a volume from filerA to filerB , instead of using snapmirror, i will give ndmpcopy a chance.
Like snapmirror it is a block level transfer(lot faster than rsync). The benefit for this would be the single file/directory that we are able to move. snapmirror transfer entire volume while ndmpcopy, we can copy a single directory within a volume, and yet still using block level.

Below is just the example output i have for references:
I am copying from a FAS3170 to FAS960
Directory size: 1.5TB
~45MBps!!!!
Time taken: 9 hours 15 minute
Ndmpcopy: filerA: Log: DUMP: dumping (Pass V) [ACLs]
Ndmpcopy: filerA: Log: DUMP: 1563223738 KB
Ndmpcopy: filerB: Log: RESTORE: RESTORE IS DONE
Ndmpcopy: filerB: Log: RESTORE: The destination path is /vol/vol_restore/re store/
Ndmpcopy: filerB: Notify: restore successful
Ndmpcopy: filerA: Log: DUMP: DUMP IS DONE
Ndmpcopy: filerA: Log: DUMP: Deleting “/vol/fc_loan/../snapshot_for_ backup.195” snapshot.
Ndmpcopy: filerA: Notify: dump successful
Ndmpcopy: Transfer successful [ 9 hours 15 minutes 47 seconds ]
Ndmpcopy: Done

Netapp autosupport undelivered

Recently found my autosupport mail was undelivered.  The config was follow exactly with the working one. So after digging for few days and finally i found the root cause. there is one file *.ems.gz is about 30MB and caused the exchange server to block it. (Thank god 1 of the triggered mail received)

Netapp support had send me the instruction to rotate the log files:

priv set advanced
syslog rotate_syslog
priv set

ta.da……..problem resolved and you are able to receive all autosupport now!

Snapmirror time taken

I have a mission to upgrade my single FAS 3020 to cluster FAS 3050.

With the help from NetApp professional service team, i managed to

  • Removed the 3020 head and replaced with new 3050 head and re-own the disk.
  • Setup cluster environment.

Here is the detail for my 3020:

Ontap 7.2.4

FC disk shelf x 6

4 port-channel network cable

My ultimate goal:

  • Upgrade to Ontap 7.2.6.1P2
  • Upgrade all Firmware(disk,disk shelf….etc)
  • Setup cluster(filer A & B)
  • snapmirror 2.8TB over to filer B.

The upgrade finished successfully and smoothly. The challenging part will be snapmirror. We have 2 days down time and was hoping the snapmirror to finish ASAP.

NetApp told me the rough calculation is :

~200GB / hour = 14 hours for the snapmirror if everything goes well.

During the snapmirror ,  there were no activities on both filer except snapmirror.

The destination(filerB) CPU load was ~45%(sysstat -m) and the speed was about 220GB/hour.

I am so glad that the snapmirror for 2.8TB finished within 12hours. Oh ya, the filer B is also configured with 4 port channel network.

I searched for so many sites and there is no article/blog mentioning about the snapmirror time, you may refer to NetApp for the data or just refer to my post 🙂

© 2019 Thinkway

Theme by Anders NorénUp ↑