ONTAP Discussions

SSD Scrubbing on Netapp AFF systems

Novicer
3,906 Views

Hello. I'd like to understand deeper about SSD scrubbing. Tried to read up online. Understand the concept and stuff but couldn't find enough information about how exactly Netapp handles it. 

We have a netapp with two aggregates that are fully flash. The one with more volume count, more IO load seems to be not complaining on the AIQUM. It shows up as green but the other aggregate with more free space, lesser volumes compared to the former and less IO overall seems to show up with a warning symbol that its performance level is over loaded. 

This looked strange and the only difference I could see is that the later had its scrubbing schedule run more than 6 months ago, the former ran last month. Nothing else seem different. Any help advice on how to navigate and find the problem would be much appreciated. Thanks!

1 ACCEPTED SOLUTION

paul_stejskal
3,872 Views

So the "other" aggregate is named aggr0. I can safely assume that is likely your root aggregate. RAID scrubs are simply a read of every block from A-Z that cycles and runs daily. https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/What_are_the_important_changes_to_RAID_scrub_in_Data_ONTAP_8.3.x_or_la...

 

UM is just saying "your disk performance is terrible and overloaded" but it apparently isn't being smart enough to know it's raid scrubs. This should be an informational alert at best. Nothing is wrong here.

 

Please open a case against UM so we can file a bug. If you want, link this thread with it. I can't find a bug right off.

 

 

View solution in original post

6 REPLIES 6

Novicer
3,884 Views

Added an image for better understanding of the question

paul_stejskal
3,873 Views

So the "other" aggregate is named aggr0. I can safely assume that is likely your root aggregate. RAID scrubs are simply a read of every block from A-Z that cycles and runs daily. https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/What_are_the_important_changes_to_RAID_scrub_in_Data_ONTAP_8.3.x_or_la...

 

UM is just saying "your disk performance is terrible and overloaded" but it apparently isn't being smart enough to know it's raid scrubs. This should be an informational alert at best. Nothing is wrong here.

 

Please open a case against UM so we can file a bug. If you want, link this thread with it. I can't find a bug right off.

 

 

Novicer
3,765 Views

I didn't know if it was ok to share the aggregate and netapp names. That is why I hid them. But the aggregates in concern here aren't root aggregates. They are standard aggregates serving data to customers consisting of only SSD disks. But I hear your opinion. I can open a case and ask if it can be registered as a bug. Will reference this page and also will post the ticket number here for your reference. Thanks!. 

Novicer
3,758 Views

Case number 2008611448 . For your reference

paul_stejskal
3,741 Views

Thank you. I see your case. You've got 4 nodes (3-6), which nodes have the aggregates with the alert?

Novicer
3,693 Views

netapp06_04_aggr0_STD and netapp06_03_aggr0_STD. Both SSD aggregates serving data.

Public