┊・hybrid-cloud

[keen spruce]

I have deployed it for 2 customers. it does work well. only catch i would say is it relies on the application for the file locking feature. Some apps dont support this e.g. notepad. in that scenario its very easy to overwrite content if the file is open by multiple users.

[bright bear]

That repo is gone 😦

[lapis oyster]

Yes seems to be deleted or made private

[placid talon]

I guess it would have to be a strange corner case if flexcache in an ONTap Edge setup couldn't do a better job.

[primal mirage]

having some performance issue with FSxN in EDA workload

vol 0 read ops is more than any other volume on the instance

[desert flume]

That's not an issue per se.

(not without more context)

In rare instances is vol0 traffic ever a problem.

[primal mirage]

we have the instance set to max of 80k IOPS allowed by fsxn

and throughput is set to 1GB

[desert flume]

Well yes, you're only doing 57 IOPS...

[primal mirage]

when performing this test

yeah 40 iops should not have that much latency

vol 0 is doing closy to 97k iops

[desert flume]

Ok... so the problem isn't high vol0 iops, the problem is high latency on reg_test.

No, that's 97kbytes/sec...

[primal mirage]

yeah Latency is a big problem

oh yea you are correct

i am reading that wrong

[desert flume]

Volume Vserver Ops Ops Ops Ops (Bps) (Bps) (us)

---

vol0 FsxI1-02 57 15 15 23 97832 11335 216
vol0 FsxI1-01 25 1 2 16 12488 6862 110

So do some troubleshooting to identify the bottleneck. I know it's virtualized, but this still applies.

https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/ONTAP_9_Performance_-_Resolution_Guide

[primal mirage]

been troubleshooting for couple of weeks now

will have to ask customer to open a case with AWS i guess

[desert flume]

Yep..then it will come to one of us perf TSEs lol.

[primal mirage]

these are flexcache destination

[desert flume]

Oh that changes things then...

[lapis oyster]

Has it been warmed at all?

[primal mirage]

yes cache is warmed

[desert flume]

Is the origin in the same location or is it at a different site?

[primal mirage]

Origin on-prem

looked at the Origin stat as well

latency is 5ms

same time when these are running

they have a direct connect circuit as well

so that is also not a bottleneck

[desert flume]

Have you gotten any packet traces to confirm the latency is 5 ms?

And is a case open yet?

[primal mirage]

no cases are opened

[desert flume]

Yeah let's start there.

[primal mirage]

yeah we did we have harvest grafana on the source

so we looked those stat

[desert flume]

AWS will definitely open a case and then the cloud TSEs will consult.

[primal mirage]

yeah i am going to bring that up in today's meeting with customer

[desert flume]

It would also help to know more about the workload.

[primal mirage]

open a case with aws

it's a EDA workload

heavy meta data intensive

running small regression test

[desert flume]

EDA...hmmm.

[hollow warren]

Open a case with AWS for them to trigger a perfASUP

[desert flume]

I'm trying to think what else to check.

[hollow warren]

In any case the question of the data is warm or not

[desert flume]

Yeah it would be good to get PAs for origin and cache clusters.

[primal mirage]

i have AWS architects on the call as well with customer

[desert flume]

Once AWS opens a case it can be used for the source cluster too.

[primal mirage]

ok cool

i will ask them to open a case any way

[desert flume]

Who is the customer?

[hollow warren]

Ifran, reach out to Yuval

[desert flume]

(you can ping me internally if you want)

[primal mirage]

yeah i will ping you internally

will do Aviv Thanks

[primal mirage]

Looks like prepopulate is causing the high IOPS

and Latency issue

[desert flume]

Interesting.

[hollow warren]

Prepopulate isn’t very efficient to say the least, you should warm the data with another method like fio/cat/etc

[smoky shore]

What's the advantage of SPOT by NetApp over AWS native Spot Fleets?

@sharp patrol there are definitely some multi cloud advantages above and beyond some of the native management options Spot has that Fleets doesn’t.

Integrations with our data mgmt platform, stateful workloads with trident, and much more, all in a single easy to use platform.

[rancid pilot]

Good Morning, can someone help me understand the difference between "NetApp Cloud Volumes ONTAP for AWS" and "NetApp Cloud Volumes Service for AWS"

[neon badge]

at a high level -
CVO = you manage an SVM basically.
CVS = you just provision and consume space.

https://www.netapp.com/blog/cloud-volumes-ontap-the-benefits-of-ontap-across-azure-aws-and-gcp/

@inland oak that is a huge trident logo in your blog buddy. 🙂

[rancid pilot]

Thanks MP. That is a very easy to grasp way to look at it. I do recall in CVO you had to provision an EC2 instance. CVS seems nice and easy to create volumes for any instance and no infrasructure to manage then .

[desert flume]

CVS I think is not available now.

[neon badge]

CVS - is out there - Google has it.
Azure it's actually "Azure NetApp Files" which is first party from Azure
AWS has CVS, but going towards FSxN

[desert flume]

Oh sorry I was thinking for AWS only.

Yeah CVS/ANF are nice because it's an A700/A800 next door to the cloud provider data center.

[hollow warren]

On AWS you have or CVO or FSxN

[desert flume]

FSxN is nice because it's a super fast CVO.

[hollow warren]

The entire deployment is orchestrated and automated by Cloud Manager

FSxN and CVO are the same. They have the same performance levels

The difference is the leve of management and supported features

[rancid pilot]

CVS for AWS - https://www.netapp.com/video/?1711740101719949106

[half summit]

this is extremely old

[half summit]

I don't understand why this is not mentioned anywhere publicly. (I remember reading it in some earlier docs that for ANF NetApp is using AFF A700s but currently this info is nowhere to be found...)
For customers when I talk about it this definitely is an "aha" moment for them. "oh my data actually is on NetApp hardware, nice"

[smoky shore]

It was Microsoft’s edict not to discuss the hardware backing it. I say it in almost all of my cloud videos.

We wanted to scream it from the rooftops

[smoky shore]

Ironically Pure has recently tried to use this against us as if to say “we’re not doing cloud, it’s just NetApp arrays.”

Missing the forest for the trees, Mr CEO man.

[neon badge]

I thought it was... "there is no cloud, it's just someone elses system?"

https://tenor.com/view/the-s-impsons-bush-homer-simpsons-homer-hiding-gif-4325800

[hollow warren]

I wish we could share the magnitude of NetApp in the different leading cloud providers

[rancid pilot]

NA should be keeping their content updated. I find the website and content overall pretty good, though sometime like this time, I find trying to figure out what the difference between all the products difficult.

[hollow warren]

Which specific products are you referring to?

[inland oak]

😄 I forgot about this blog post... I think I just gave this to someone as a word doc or something... Trident is HUGE

[primal mirage]

Quick Question? do we have a document that shows difference in CVO between the hyperscalers AWS vs. GCP vs. Azure

specifically how a flashcache is available in Azure and AWS but not in GCP

[desert flume]

I don't know if we have a document per se. Are you just wanting to know why that is?

[primal mirage]

Customer wants a side by side comparison and any documentation as to why if there are any variation between the hyperscale

[desert flume]

Hmm, you might reach out to a CSA for that.

[primal mirage]

well do i reach out to all 3 of the CSA for their expertise i would think a PM or TME for CVO should have the comparison

FSxN does it do user quotas like ONTAP?

not able to see any documentation around user quota on FSxN

this is the only document i am able to find on quota https://docs.aws.amazon.com/fsx/latest/ONTAPGuide/limits.html

[primal mirage]

similar to the doc for ONTAP quota https://docs.netapp.com/us-en/ontap/volumes/quota-targets-types-concept.html

[random marsh]

In 2-nodes HA CVO, Do I need to pay double for the provisioned size because the same amount of EBS volumes will be allocated on both nodes in HA? For instance, If I provisioned 40TB space, then I need to pay the price for 80TB, correct?

[primal mirage]

i would run the CVO price Calculator from cloud.netapp.com for a particular cloud provider and check the difference between HA and non-HA

[hollow warren]

Hi @random marsh ! You never need to pay twice. If it's CVO by Capacity then the allocated EBS isn't relevant for the licenses, it's just the provisioned ONTAP capacity.

There's a difference between single and HA for the functionality, not the allocated EBS

https://cloud.netapp.com/pricing

[hollow warren]

No

You should compare by the documentation

[random marsh]

@hollow warren It does look like I need to pay double for the storage, upon the CVO calculator: https://cloud.netapp.com/aws-calculator. If I provision 40TB, I will pay $2,048 on the storage for single-node CVO, but $4,096 for 2-nodes HA CVO. So, as I said, for HA, each node has 40TB space, that would be 2x40TB, but the user can only 40TB. Please let me know if my understanding is correct?

[hasty panther]

Hi all, in which channel should I post Trident CSI questions?

[sage nest]

Hi @hasty panther, #┊・astra🔒 is the channel you’re looking for.
Edit: I see you found it 🙂

[hasty panther]

Thx

[scenic night]

Hi gang, I am getting 403 error with https://activeiq.netapp.com/catalog/internal/api-reference/activeiq-public/capacity-forecast-v2/#get-v1-capacity2-usage or "GET /v1/capacity2/forecast" but "GET //v1/capacity2/storage or GET /v1/capacity2/prediction/summary" works just fine. Any idea why I am getting the 403 error? I am internal employee btw

[frosty spade]

I could be completely misunderstanding your ask here, but the first link you pasted is a documentation link not an API service URL. "https://activeiq.netapp.com/catalog/internal/api-reference/activeiq-public/capacity-forecast-v2#get-v1-capacity2-usage" goes to the API catalog documentation. As the base URL for API is api.activeiq.netapp.com

[agile lynx]

Hello everyone! We can't deploy NetApp CVO-HA resources. I've tried Cloud Manager and TF provider and getting the same error. I believe I've managed to get the exact API that causes the error:

I've found the API which causes this:
curl -k --location --request GET "https://cloudmanager.cloud.netapp.com/occm/api/gcp/ha/metadata/regions" --header "Content-Type: application/json" --header "x-agent-id: $CONNECTOR_ID" --header "Authorization: Bearer $USER_TOKEN" | jq -r

Output:
{
"message": "Numeric value (9.2233720368547758e+18) out of range of int (-2147483648 - 2147483647)\n at [Source: (StringReader); line: 320, column: 42] (through reference chain: com.netapp.oncloud.simplicator.client.gcp.GcpRegionsResponse["items"]->com.fasterxml.jackson.module.scala.deser.GenericFactoryDeserializerResolver$BuilderWrapper[0]->com.netapp.oncloud.simplicator.client.gcp.GcpRegionResponse["quotas"]->com.fasterxml.jackson.module.scala.deser.GenericFactoryDeserializerResolver$BuilderWrapper[60]->com.netapp.oncloud.simplicator.client.gcp.GcpQuotaResponse["limit"])",
"causeMessage": "SimplicatorBadRequestException: Numeric value (9.2233720368547758e+18) out of range of int (-2147483648 - 2147483647)\n at [Source: (StringReader); line: 320, column: 42] (through reference chain: com.netapp.oncloud.simplicator.client.gcp.GcpRegionsResponse["items"]->com.fasterxml.jackson.module.scala.deser.GenericFactoryDeserializerResolver$BuilderWrapper[0]->com.netapp.oncloud.simplicator.client.gcp.GcpRegionResponse["quotas"]->com.fasterxml.jackson.module.scala.deser.GenericFactoryDeserializerResolver$BuilderWrapper[60]->com.netapp.oncloud.simplicator.client.gcp.GcpQuotaResponse["limit"])"
}

Could someone suggest please?

[hollow warren]

Did you try from the UI?

What does the timeline says?

You should try in the in app chat in Cloud Manager

[agile lynx]

Yes, I tried. UI says the same. The API call is just the next step of my investigation ...

[hollow warren]

Sounds like a support case is in order

Which version are you deploying?

We removed 9.11.1 this week so it might be the cause

[agile lynx]

I know, so I tried 9.10.1. When you try 9.11 the error looks straightforward 'Couldn't find version' or something like that

[hollow warren]

Can you try another version?

[agile lynx]

I think I can. But what about the direct API call. Shouldn't it be working?

https://docs.netapp.com/us-en/cloud-manager-automation/cm/wf_aws_cloud_md_get_regions.html#get-regions-for-high-availability-pair

[random marsh]

Can we reduce the size of storage capacity in FSx ONTAP after created?

[smoky shore]

Yes. Check here for "Updating a Vol Config"

https://docs.aws.amazon.com/fsx/latest/ONTAPGuide/managing-volumes.html

API here: https://docs.aws.amazon.com/fsx/latest/APIReference/API_UpdateVolume.html

[random marsh]

@smoky shore I am sorry, but I am looking for a method to DECREASE the storage capacity for the entire FSx for ONTAP after it is already provisioned. I am not looking for one to decrease the size of volumes. Make sense?

[smoky shore]

@hollow warren might know, but he's in Tel Aviv, so odd hour for him. I'm sure you can, it's just not something I've ever done, personally.

[random marsh]

Yeah, let's wait for Aviv to confirm. I was told by a knowledge person we cannot, but just cannot find documents to approve

[smoky shore]

The alternative would be to stand up a new FS, move the data, spin the other down

[red leaf]

Hi David. Pretty sure shrinking the FSxN total size is still on the roadmap. Sorry. You will need to follow Nick's method. Adding your interest here helps get it on the radar for the product management team.

[random marsh]

Thank you! That should be the answer

[hollow warren]

@random marsh I can confirm.
You should try using Cloud Manager for setting up the replications or increase the FS size.

[random marsh]

@hollow warren or Anybody, I have questions about Backup Tool in built-in FSx ONTAP, Can I set up schedules to automatically back up a certain of NFS shares? Can you please send a document on details? Thank you!

[hollow warren]

It’s a backup by AWS Backup that can protect and recover on a volume level

The backup runs on 00:05

[random marsh]

Can we back up data to a different region using FSx Backup or AWS Backup?

[hollow warren]

I’m not familiar with this feature availability for AWS Backup for FSxN

Do you have such a requirement?

[warm hamlet]

Hello everyone
I have a permission problem when registering EKS in Cloudmanager.
I checked the eks audit log to investigate this issue, but I couldn't find the K8S user that cloudmanager uses to access EKS.
Has anyone succeeded in registering eks in cloudmanager?
[DOC]
https://docs.netapp.com/us-en/cloud-manager-kubernetes/requirements/kubernetes-reqs-aws.html#review-networking-requirements
[ error message ]
Unauthorized, Your Kubernetes Cluster needs additional permissions. See which permissions are needed here:
{kind: "Status", apiVersion: "v1", metadata: {}, status: "Failure", message: "Unauthorized",…}

[cerulean carbon]

FSxN Backup only works within a region at this time. Cross region backup/restore is said to be coming soon.

[random marsh]

@cerulean carbon Need your help to please clarify on following:

1. I believe FSxN backup can back up data to a different AZ in the same region. Right?
2. Based on my reading, AWS backup can COPY data across regions. Right? If yes, Can FSxN backup do that as well?

[gray fjord]

Hello - I am planning on deploying a NetApp CVO on AWS. I have a requirement of bringing in an IP I am using on-premises. If I use a multi-node CVO instance, all the cluster IPs will be using floating IPs which are not part of the VPC CIDR block. But if I use single-node CVO instance, I can add a secondary CIDR block(with my on-prem IP) to the VPC and use it as the SVM's IP. Is there anyway I can use a multi-node instance and still use VPC CIDR for the SVM's endpoint?

[desert flume]

I don't believe so @gray fjord

[smoky shore]

@spice barn you have some new flair...

Glad to have you here with us!

[spice barn]

🫠🫠🫠

Thanks Nick! For everyone here, I work on the FSX ONTAP product team so please don’t hesitate to reach out if you have questions

[frosty spade]

Hey Cloud Experts! 🤙 We have a customer post on the Community Forum that I wanted to see if anyone can help answer, https://community.netapp.com/t5/ONTAP-Discussions/NetApp-CVO-Network-configuration/td-p/438686 it's related to AWS CVO

[random marsh]

@spice barn Chris, Glad to hear you work on FSx ONTAP team. We are evaluating if we can combine two CVO's in AWS to a FSx ONTAP instance. To measure if this FSx ONTAP can bear workloads from these two CVO's when calculating what IOPS or throughputs we should assign to the FSx ONTAP, can we in general or roughly combine IOPS's and Throughputs (avg or max) collected from these two CVO's and as the basis of the resources to create the FSx ONTAP?

[spice barn]

David, I will check internally for you, but I think your approach makes sense. You could then pick an SSD size & throughput configuration to match

[desert flume]

What is your workload profile now?

[random marsh]

One is for Oracle DB and the other is for DB backups

[desert flume]

It depends on your I/O profile and how many IOPS/MBs you have.

[random marsh]

How do I create or use a Azure disk with the encryption of SSE with CMK? As you can see in the KB https://kb.netapp.com/Advice_and_Troubleshooting/Cloud_Services/Cloud_Volumes_ONTAP_(CVO)/How_to_map_an_Azure_Managed_disk_to_a_CVO_disk, the CVO uses disks with the encryption of "SSE with PMK" which is by default. The need here according to the Security requirement here is to use disks with "SSE with CMK". What can we do to achieve that? Does NetApp support that.

[cerulean carbon]

The official documentation is located at https://docs.aws.amazon.com/fsx/latest/ONTAPGuide/protecting-data.html There is no mention of where the backup is stored but assuming some type of S3, it's probably spread across all AZ's in the region. There is no FSxN Backup replication at this time. Only SnapMirror/SnapVault replication, which then could be backedup in a different region. But new features are always on the way!

[cloud basin]

hello All, anyone has tried to use velero with trident for restoring a backup on a new cluster ?

Is this place correct for asking about trident and velero ?

[lapis oyster]

You’ll want to try the #┊・astra🔒 channel. I’m not sure how many will have tried this combination though. We have Astra Control for backup/DR/data management that works with Trident.

[cloud basin]

@lapis oyster thanks but I think astra is a different solution . We can think to use it but we take a lonk time to acquire licenses/subscriptions, meanwhile we need a free solution

[lapis oyster]

The Astra channel covers both Astra Trident and Astra Control products

[spice barn]

Hi David, a couple of answers for you:

1. For a Multi-AZ file system, FSxN backups will be in the same region as the ONTAP instance but will be spread across different AZs. For Single-AZ file systems, FSxN backups will be in the same AZ as the ONTAP instance.
2. We do not currently support copying or moving FSxN backups to other locations

Feel free to ping me directly if you want; I am on the FSxN product team (and used to work for NetApp). And thanks Richard for bubbling this up!

[west frost]

Hi ! I'm trying to add a CentOS Linux with SAP HANA to Cloud Insight but there is no Collector for CentOS.
Someone tried to add it manually?

[frosty spade]

Possibly I have misunderstood you, but we do have a Collector agent for CentoOS/RHEL -https://docs.netapp.com/us-en/cloudinsights/task_config_telegraf_agent.html#rhel-and-centos

[west frost]

sorry my bad ! i was thinking in CentOS but Hana runs in Suse Linux
CI doesn´t have a collector for Suse it´s that correct?

[frosty spade]

Hey Gonza, I don't see anything in our docs, but will let someone with more experience with Cloud Insights chime in on your query.

[west frost]

Thanks Ross .

[smoky shore]

I reached out to the PM for a response. Either way we’ll get you an answer Gonza

[smoky shore]

@west frost ok here’s the answer for ya. Cloud Insights uses the telegraf agents for collecting some 3rd party stuff.

Here’s the SUSE telegraf agent: https://packagehub.suse.com/packages/telegraf/

…and here’s a doc on configuring telegraf agents with CI: https://docs.netapp.com/us-en/cloudinsights/task_config_telegraf_agent.html

[west frost]

thanks Nick.

[smoky shore]

Anytime!

[mental flax]

@west frost Hey Gonza, here are the steps I would follow

Install telegraf via suse package or rpm
Get the Cloud Insights agent configuration file, there are two ways to do that:
1. (Easier) Install the agent on a supported OS, copy the configuration file from `/etc/telegraf/telegraf.d/cloudinsights-default.conf`
2. Use the installer snippet to get the configuration file out of the install script
  - Follow steps for the "Preparing the Telegraf configuration file" section in the blow post below
  - Link: https://community.netapp.com/t5/Tech-ONTAP-Blogs/Monitor-Everything-Automated-deployment-of-Cloud-Insights-agents-using-Ansible/ba-p/167712#toc-hId--733935034
Then replace items in the configuration file for your host:
- `agent_node_uuid`
- `agent_node_ip`
- `agent_node_os`
Finally, make sure there is no conflict with the main telegraf.conf file:
- mv /etc/telegraf/telegraf.conf /etc/telegraf/telegraf.conf.bkup
- touch /etc/telegraf/telegraf.conf
- chmod 644 /etc/telegraf/telegraf.conf
Then restart the telegraf service

You can also review the install script from the install command, its basically does the above process.
Let me know if you have any other questions.

[west frost]

Thanks folks ! i will try to install Telegraf tomorrow.

[ionic linden]

Hi all, we have deployed some NetApp CVOs (so its not a fully managed solution, Azure NetApp files wasn't an option apparently for us) in Azure and wondering what people use for monitoring CVO's for day to day stuff, such as monitoring, capacity mgmt, ALERTING (v important obviously!). Any real world advice welcome. I know Unified Manager pretty well for on prem filers. Is that still valid in Azure? Cloud Insights possibly? But our 3rd party has said its main purpose is to determine if on-prem data is a good fit for cloud and not applicable for our needs.......however the marketing on it doesn't seem to suggest that to me, and its more of a SaaS version of UM maybe? Having a SaaS option to monitor some CVOs would be my preference. But any pointers or ammo to use against our 3rd party is welcome! We are also using NetApp Cloud Backup too for all of this, so if something also tied into that monitoring wise, even better....Many thanks.

[desert flume]

Same as on prem! Cloud Insights is our flagship monitoring product and can do quite a bit of stuff! You might request a demo.

[ionic linden]

@desert flume awesome, many thanks. That is good to hear. I am making them setup the free version regardless of what they are saying, and I believe it starts in trial mode so I can play with the full features too. (although I feel they really didn't want to set this up for us and are trying to make us use snmp monitoring to an on-prem SolarWinds instance instead which I just find bizarre). But Cloud Insights did sound far better than they described it from what I was reading.....plus its SaaS - ticks my box! But flagship monitoring product clearly sounds like its the right way to go if we can. Do you know where I can request a demo please, as we are only are liaising on all things NetApp via the 3rd party only and I would like a true demo of its power! I think that would be really useful too to sell it to my business. Obviously we have just the 3 x CVO's so far but all being well, that will expand and we'll have the right tools in place to grow it out. Thanks.

[desert flume]

https://cloud.netapp.com/ is our main cloud page. If you click "Products" then "Cloud Insights" there is a nice "Start Free Trial" button. 🙂

[random marsh]

@spice barn Hi Chris, I just have the chance to reply by asking a couple of follow-ups:

1. Weather it is Multi-AZ FS, or Single-AZ FS, FSxN backups will be saved in a type of S3 object storage, my understanding is S3 data spread across all AZ's in the region(or maybe in difference region? i am not sure), and can be preserved if the AZ containing the FS failed, which means I can restore data in the volume from S3 in any of cases. Right?
2. My understanding is AWS Backup can copy data from one region to the other, since FSxN is a AWS service, then AWS Backup can do to FSxN backup data too, right?
   In this video https://www.youtube.com/watch?v=otFjYm1l7Do, at about 4:57 minutes, she stated, with AWS Backup, you can copy your Amazon FSx FS backups across AWS regions, , or both, helping you meet business continuity...
   We prefer to use Single-AZ FS for saving cost, and if meanwhile we also can somehow store the backups to a different AZ or region, we will then feel safer.
   I am slow, Thanks for helping me to get out of the confusion!

[idle token]

A fsxontap question: Can I add the FSX ONTAP File System into ActiveIQ UM for monitoring? I understand I can use AWS CloudWatch, but prefer to use AIQUM if I can. Thanks!

[distant palm]

Pretty sure I know the answer to this one, but figured I'd ask just in case -- is there any recommendation or guidance (for or against) backing up the CVO instances themselves (e.g. through AWS or Azure backup of the VM itself?)

[desert flume]

You can SnapMirror the data elsewhere.

[distant palm]

Yeah for sure, and this use case is actually a SM target-- but in general the customer was asking "should I backup the VM in Azure too" and I'm quite sure the answer is "no" but wanted to double-check in case I'm missing something

[desert flume]

Depends on the risk model.

[distant palm]

Interesting, so is it technically supported to backup and restore the CVO VM instance directly then?

[green ginkgo]

I’d think it very unlikely to be supported, no. WAFL doesn’t do crash consistent very well at all. Cloud Volumes ONTAP automatically creates configuration backup files that contain information about the configurable options that it needs to operate properly.
By default, Cloud Volumes ONTAP backs up the files to the Connector host every eight hours.

[distant palm]

My thought as well. Good to know on the connector backup though, hasn’t realized that!

[hollow warren]

VM backup won’t help for Cloud Volumes ONTAP.

We take snapshot for root/boot devices and in the worse case we can use it for recovery

Even if it means spinning on a new cluster and mount that snapshot to get ONTAP configuration

*Cloud Manager is taking this snapshot during deployment and during upgrades

PS I don’t remember a case of a disk failure in Azure

[low obsidian]

Hi guys,
Today I recognized a change in cloud manager.
I tried to setup a CVO HA in Azure. In the past I could activate a checkbox if the connection to the storage account should be secured by SSL. This option is gone and the deployment now fails, as my organization forces SSL for storage account connections by policy.
Is there a way to activate this in cloud manager again?
We are using Cloud Manager 3.9.22 Build: 2

[desert flume]

What version of ONTAP?

[low obsidian]

9.11.P3

Ah, on 9.10.P3 I can activate SSL for storage accounts

[desert flume]

Hmm.

That may be an issue with the ktlsd performance issues we've had and maybe Cloud Manager not being updated.

I'd suggest opening a case to ask about this. It may require investigation with OCCM engineering.

[low obsidian]

Ok, thx. I will do

[distant palm]

Seemingly random question Re: CVO on Azure (deployed with Terraform, but not sure that's relevant) -- it appears wether or not we enable "capacity_tier" on the first aggregate, it still created the Azure storage account/container for fabric_pool. Is it expected behavior that it will create those things even if you never intend to use tiering w/CVO?

[idle token]

On AWS FSx ONTAP, can I create a new volume via CLI or using AWS FSx ONTAP console, not by using Cloud Manager? On CVO, for anything CVO can do should be done by Cloud Manager not by CLI’s

[smoky shore]

Yes, FSxN can be administered in any of those places, assuming permissions are properly granted.

[rocky glade]

You can manage CVO volumes from Cloud Manager, System Manager and CLI

[fluid crane]

Hi can anyone help me with Step by step guide/document for GFC Installation ?

[frosty spade]

Hey @fluid crane , this is a good starting point - https://cloud.netapp.com/global-file-cache/onboarding , we also have some more information over at our docs.netapp.com site, https://docs.netapp.com/us-en/cloud-manager-file-cache/concept-before-you-begin-to-deploy-gfc.html

[fluid crane]

Thank you.

[carmine cloak]

Silly question, is ONTAP Select supported on VMware Cloud on AWS? We have a weird use case for a small amount of data. Didn't see anything in the documentation that clearly said it wasn't

[green ginkgo]

I don’t know for sure! I think Deploy might be very unhappy to be honest. I’d look at FSxN ONTAP or Cloud Volumes ONTAP for AWS instead - might be much cheaper than the VM for ONTAP select

[carmine cloak]

We were looking at FSxN ONTAP, we have a snag though as we currently have our networks stretched between on prem and VMC for our migration so we pick up a lot of latency trying to get to FSxN ONTAP

which we have a plan for most of the ONTAP workload, it's a small subset of data (<10 TB) that is on DMZ vLans that we'd like to keep there

if it's not possible, that's totally fine

[green ginkgo]

Checking out the IMT, it seems to meet requirements - vmware version match, vsan for underlaying storage. But the access you get to vmware in AWS is not bare metal as I understand it.

[carmine cloak]

Ok so there may be configuration requirements that we may not be able to make because it's a managed service?

[green ginkgo]

potentially - @cyan depot might be able to help further when he gets up. Bedtime here in Australia! best of luck with solutioning

[dreamy delta]

Hi Dev, several of my teammates came over with the acquisition that became GFC and the next evolution being Cloud Volumes Edge Cache (CVEC). Please tag me if you want me to connect you with someone who can provide you more dedicated assistance!

[cyan depot]

I don't see a reason it wouldn't work, but I haven't tried it. Assuming your vcenter account has the required permissions (detailed in the OTS docs) I would expect it to work. Its easy enough to validate with an eval instance if you have a running environment to put it in.

[green ginkgo]

Thanks @cyan depot ! @carmine cloak sounds like it’s worth a shot!

(Midnight baby feed here - back to sleep now!)

[carmine cloak]

Ok we'll give it a shot and see how it goes

thanks!

[cyan depot]

Just recall the SDDC network it created will likely default to mtu 1500. You'll need to bump whichever networks are used to the internal/HA traffic to jumbos if you do HA, and drop the MTU in deploy to 8800 to allow for NSX overhead in the packets.

[carmine cloak]

ok good to know

[smoky shore]

@carmine cloak you may want to join this Q&A Panel with VMC Prod Mgr and FSx ONTAP Prod Mgr tomorrow! https://discord.com/events/855068651522490400/1033123267403337828

[carmine cloak]

oh nice, thanks!

@smoky shore is this going to be the same as ENT308 at AWS Re:Invent?

[smoky shore]

I don't believe so. I haven't looked at the re:Invent catalog quite yet. This week, our INSIGHT event is going on and we're bringing in some key presenters to do breakout sessions and open Q&A Roundtables. This is one of them. With the rise of VMC and the GA announcement of FSx ONTAP for VMC at VMware Explore a couple of months ago, I wanted to bring in those two PM's together with me to have a chat, field questions, and give any updates on the services.

[jolly peak]

is there a cloud ontap manager image i can install in kubernetes?

[forest violet]

https://aws.amazon.com/about-aws/whats-new/2022/11/aws-launch-wizard-amazon-fsx-netapp-ontap-sap-hana-host-auto-failover/

[idle token]

Has anybody tried to add FSXONTAP service into AIUM? I was told that it won't work, and also was not mentioned in the link https://docs.aws.amazon.com/fsx/latest/ONTAPGuide/monitoring_overview.html So, would like to find out if anyone of you ever tried to do that?

[desert flume]

I think it's coming.

[idle token]

Any idea on a timeline?

[lapis oyster]

There’s some announcements at Re:invent, I don’t know if this is included but that will be the next big set of news given it’s just a few weeks away

[forest violet]

We should plan a Discord community meetup at the NetApp booth at re:Invent. @lapis oyster @smoky shore

[smoky shore]

https://tenor.com/view/elrond-egld-crypto-criptomoneda-im-in-gif-23033082

[desert flume]

I don't know.

[obtuse stump]

Hi all, I have a question about azure HA in Availability Zones. Is this coming on 9.12 if aso is there an eta for its release?

[smoky shore]

Regarding which product? Azure NetApp Files? CVO?

[hollow warren]

Cloud Volumes ONTAP Multi AZ deployment is coming with 9.12.1

Very soon

[sour stag]

does anyone know if you can lock workspaces on BlueXP to certain members only ?

[frosty spade]

Hey @sour stag Yeah, that would be my understanding. An account creates (owns) a workspace, and can then add other users to access that workspace. https://docs.netapp.com/us-en/cloud-manager-setup-admin/task-setting-up-netapp-accounts.html

[sour stag]

Ahh.. that's it.. All my users are "Account Admin" and they have access to all workspaces.. I'll have to do some shuffling around..

[shy crystal]

Hey just a bit of feedback and wondered what peoples thoughts are. I've been looking at FSXN and in particular their documentation for monitoring. There's a reference here https://docs.aws.amazon.com/fsx/latest/ONTAPGuide/monitoring-harvest-grafana.html that harvest can be used which I thought was brilliant as we're heavy users of this on prem. After some botched cloudformation deployments (aws ticket pending, running on default 80 ports and other oddities) i've managed to stand this up on my own. Pretty excited at this point to only realise 3 out of 20 dashboards actually work due to FSXN permissions

[primal mirage]

I have a customer with single node CVO instance in GCP running about 5k nfs IOPS and latency is hitting 128ms any help is appriciated

[desert flume]

Got a case open?

I'm from the perf team @primal mirage and can look if perf archives are triggered.

Also what is your latency breakdown using "q s v l s"?

[primal mirage]

yes

messaging you in teams

[desert flume]

Ok.

tl;dr: case is escalated with CPE/dev.

[random marsh]

Thanks for sharing. Have you tried to use CloudWatch within FSxN? It can show you throughput, IOPS and Latency. Those should be good enough to me to do the performance monitoring, I am wondering what else would you be looking for and you think we should have to do the job?

[vast ether]

Hi NetApp team, when I list snapshot policies from rest api(/api/storage/snapshot_policies)) via cluster management port, only part of policies or no policies are retuned. but actually I do have some more policies on my cluster and if I list them by SDK I can get them all. do you have any insight?

[desert flume]

Probably this is the wrong channel (there is an API one), but you might open a case as it sounds like a bug.

[graceful pumice]

Hi I was trying to output data to Cloud Insight through telegraf
After waiting for over 15 minutes, I can still not find the Category I set in the config file
I ran the cloudinsight-default.conf config file
Here is the log file:

[mental flax]

Hi I was trying to output data to Cloud

[spice barn]

As Paul said correctly, it is on the roadmap to be coming soon. Please reach out to your account team (or me) for more details

[smoky shore]

Hey @spice barn will we see you at re:Invent?

[spice barn]

Negative 😭

[signal dune]

Is cloudmanager (bluexp) SSO auth broken? Since the page has been updated I haven't been able to login. Once I pass SSO auth I get an "Oops! Something went wrong" error message from Netapp (invalid_request: failed to obtain access token)

[green ginkgo]

Have you tried in a private browsing window?

[signal dune]

Yep, it's not a cookie thing. I believe the auth actually progresses because the cloud chat window says "hello Carl" (in a private window), just cloud manager doesn't load, and going back to cloud.netapp.com doesn't login either

[green ginkgo]

Hmm yeah, that doesn’t sound right

I’ll let one of the cloud team chime in

[smoky shore]

Good morning, Carl. Thanks for being here and letting us know. I’m at the airport with limited resources, but I’m able to log in to the service in Las Vegas.

I’m checking on what might be happening.

[hollow warren]

Auth is working. I suggest to work with support on this one

[signal dune]

Yeah I guess I'll have to raise a support request. It seems to have stopped working for my entire org. Only via SSO, not with another account like a support account. Could be a problem with my side.

[hollow warren]

Are you using Identity Federation? Did you try to changed to login with NSS?

[sage nest]

📢 Take the Cloud Data Services certification beta exam and get a voucher for 50% off the exam of your choice🎓

The NetApp Cloud Data Services exam [NS0-700] is designed to certify a candidate’s ability to deploy, configure and administer the NetApp cloud storage and data services solutions.

Achieve this valued certification at a 50% discount (USD $75) by taking the beta version of the exam prior to the final release, and you will receive a 50% off voucher on the exam of your choice!

Act fast—the discounted beta version and 50% off your next exam voucher will be available only until December 31, 2022.

Go to View Exams on the (https://home.pearsonvue.com/netapp/) page and enter NS0-700 or Cloud Data Services to select and schedule your beta. Pass this exam to be one of the first with the new NetApp Certified Professional Cloud Data Services credential!

Note: You will receive your 50% voucher via email when the Cloud Data Services beta ends.

Questions? @surreal pebble can help!

[hollow warren]

I wonder if I’ll pass 😅

[neon badge]

Take it 🙂 @hollow warren

That was a fun one to help write.

[main hatch]

What's the best way to learn about NetApp Cloud Data Services to get ready for the exam (for someone who has never used these services before)?

[sage nest]

@neon badge, do you want to take this one? Or @surreal pebble?

[neon badge]

it's slated to replace the Hybrid Cloud Admin exam I believe. But doesnt' look like the Topics list has been published.

https://www.netapp.com/support-and-training/netapp-learning-services/certifications/cloud-administrator/

[neon badge]

Nita or Eva can comment when the list for that will be updated.

[surreal pebble]

@main hatch Hi Jim - Here is a list of what you will be tested on in the Cloud DS exam:
Fundamentals, identify:
• Networking concepts
• Components of NetApp Cloud Data Services Solutions
• IAM Concepts
• Documentation and tools used for deployment
Implementation:
• Describe tooling used for replication or migration
• Deploy and configure NetApp Cloud products and services
• Administer the hybrid cloud
Automation:
• Identify Infrastructure as Code (IaC) tools
• Demonstrate the ability to deploy using Infrastructure as Code (IaC) tools
• Describe the concepts of storage automation
Monitoring:
• Identify monitoring components in NetApp Cloud solutions
• Demonstrate the ability to monitor Cloud storage solutions
• Describe concepts of cloud storage monitoring
Troubleshooting:
• Demonstrate the ability to troubleshoot deployments of NetApp Cloud Data Services
Describe tools used for troubleshooting cloud storage

[surreal pebble]

@main hatch The recommended training still TBD, but here is a short list of suggested courses:

NetApp Public Cloud Services Fundamentals
Cloud Volumes ONTAP Fundamentals
NetApp Cloud Manager Fundamentals

[random marsh]

On AWS FSx ONTAP console, how can I run aws CLI's? For instance, if I wanted to run "aws fsx delete-volume --volume-id fsvol-1234567890abcde", what steps do I need to go through to set up the CLI environment?

[green ginkgo]

hi there! aws commands run from your system, not ONTAP - this looks like a good overview of how to set it up - https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-configure.html

[balmy goblet]

Hi, I am ex-NetApp and current NetApp customer here. I would like to know if FSx ONTAP supports Flex groups?

[random marsh]

A better and easier way to set up CLI environment is to use AWS CloudShell icon on AWS FSx ONTAP console. Is this the interface I can run AWS command line from? (I got permission errors when I am trying to start, but this is another issue I need to talk to our AWS Admin, but I just wanted to make sure this is a right interface first)

[green ginkgo]

https://aws.amazon.com/fsx/netapp-ontap/features/ - “Each Amazon FSx for NetApp ONTAP file system can scale to petabytes in size, allowing you to store virtually unlimited data in a single namespace.” - that has got to be FlexGroups

This doc lists two caveats for using FlexGroups, so I’d say that confirms it https://docs.aws.amazon.com/pdfs/fsx/latest/ONTAPGuide/ONTAPGuide.pdf#what-is-fsx-ontap

@spice barn potentially can you include the term FlexGroups in the feature description pls?

[spice barn]

thanks for the tag, Alex! @balmy goblet yes, FlexGroups are supported.

if you are SnapMirroring from "on-prem" ONTAP into FSxN, please make sure the FSxN FlexGroup is created with the same number of constituents as the on-prem FlexGroup

[balmy goblet]

@green ginkgo @spice barn Thank you for confirming. I worked on the Infinite Volume when I was at NetApp and Flex groups was a WAFL driven initiative to address the same scale out problem space. Rooting for NetApp always.

[spice barn]

I never had any InfiniVol customers myself but I do remember the tech!

[desert flume]

I think I remember seeing your name.

[green ginkgo]

Yeah, infinite volumes was in retrospect not ideal from an architectural pov.

FlexGroups are awesome though

[neon badge]

https://tenor.com/view/everything-is-awesome-good-mood-good-vibes-lego-gif-14154321

[balmy goblet]

Infinite volume was all implemented in the networking layer (Nblade). I agree doing it in the file system is the right place. This is all possible due to the cool dblade/nblade separation cDOT offers. Kudos to the WAFL team for pulling off Flex groups, that is amazing engineering.

[balmy goblet]

thanks for the tag Alex Dhruva3219 yes

[green ginkgo]

where are/were you based @balmy goblet ?

[random marsh]

I have another question about FSxO, if you can please help me out. There are two volumes now can only be seen from AWS console as shown below. They were used for SM destinations, then late somehow removed from ONTAP. I did volume show, and snapmirror show, there are no such volumes or ones with same fsvol ID's on ONTAP. therefore they no longer exist on ONTAP. Further, I cannot delete them from AWS console. There are no any errors about deleting process, but it took forever and still shown there. My questions: Can you think of anyway somehow they are still in use? and what can I do to delete from AWS console?

If I click on that square with ?, I got the message below.

[desert flume]

Are there any snapshots left on the source? Or are they in the vol recovery queue?

[spice barn]

are they still visible now? if you do certain actions in the ONTAP CLI (e.g. creating or deleting a volume) it can take up to 60 minutes to show up in the AWS console

[random marsh]

They were now removed. Here is what happened. For easy explanations, let's call those 2 volumes as described above as "AWS volumes". There were the other two volumes in ONTAP (Let's call them "ONTAP volumes") and they were destination volumes in a SM relationship as the replacement for those "AWS volumes". These two sets of volumes happened to be with same names to each other. Could that be the reason of how these two sets of volumes got tied up together? It shouldn't be, because their UUID numbers were different, and "AWS volumes" were already removed from ONTAP as I said before. Finally, the fix was to remove ONTAP volumes and also SnapMirror relationship(the source was on-prem) first before we could delete 2 "AWS volumes" from AWS console. So, can you please explain to me internally or logically as for why?

[spice barn]

If you want to email me we can try and figure this out together offline!

[gray fjord]

Can cloudsync do a hash validation on data during migration?

[queen dock]

Can you be more specific about when during migration? What exactly is the issue we are trying to address?

[gray fjord]

Can cloudsync verify and validate data migrated for compliance reasons?

[dreamy delta]

Have you looked at Data Sense for compliance and file auditing/tracking? https://bluexp.netapp.com/netapp-cloud-data-sense

[dusky condor]

Hey! Please Netapp - fix the numerous bugs in blue xp. There are so many things which does not work correctly.
Eg

• Readding working environment and a error msg regarding azure subscription cover important fields
  *standardview dors not have checkbox on modify activated even the setting is activated (e.g standardview requests to activate by some volumes nfs but i have all activated)
  *job page of cloud backup displays 1 second an overview and will be then covered with an error: no cloud backup is activated but the backup is activated
  *eventlog in advanced view cannot be displayed
  fileexplorer in bluexp never worked this works only in sysmgr!
  *restore is not possible in degraded state but a restore must be possible also in this state
  *missing settings in modify standard view which should exist according to the docu
  *issues with autoupdate ( is not displayed under some circumstance
  *sysmgr suggest to deactivate admin account but without bluexp is broken
  *changing working env username is not possible without readding workkng environment
  *cloud backup policy cannot changed: message say: archivation is activated but it isn‘t

I could spend days just doing ticket opening, for every issue (including discussions with support)
The blue xp quality and development must definitly improved!

[smoky shore]

Thank you for reporting all of these. The core team is traveling back to Tel Aviv from Las Vegas for re:Invent, but I'm sure @hollow warren and others will be all over these when they get back. //cc @alpine hare

[limpid flame]

@ionic linden To start a trial, go to this link. If you spend some time on this page, a chat bubble will pop up in the bottom right corner. You can request a demo there. https://bluexp.netapp.com/cloud-insights

[hollow warren]

wow, it's not something we are seeing with other customers... Is it still the case? we had a release a few days ago... It might have been somehting with your Connector

[prime pilot]

If I have a volume that is 99TB, can I enable auto tiering on the volume to keep from having to convert to a flexgroup to support growth or does the volume still show the same size and the aggregate sees that space savings?

[desert flume]

You need to split it because FlexVols are limited to 100TB cold or hot.

[half summit]

I mean you could still use FabricPool to tier your cold data away. If you for example manage to move 30tb of cold data to your cloud tier you would have more empty space in your 99tb vol. If the free space is enough for you might not need to grow over 100tb.
But be aware that you need ~5% of the tiered away data for metadata.

[vale marsh]

In addition to the suggestions already put forth, consider what you have there for snapshots. If nothing, then FabricPool/CloudTiering may help (with the metadata caveat OG1 listed). If much of the capacity is snapshots then you can try the snapshot-only policy as well. Or just go straight for the nuclear option and tier it all. 🙂

I would check IDR to insure enough data is cold to make it worthwhile. https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/How_does_the_FabricPool_inactive_data_reporting_work

[green ginkgo]

Fabricpool doesn’t make the space available in the volume, it makes it available in the aggregate, so as @desert flume said- 100TB, hot or cold

[dusky condor]

Hi, i opened tickets for that! Covered azure Subscription error is a usability thing, standardview checkbox nfs is still open as ticket, Job Page showing only self initiated Backup Jobs, should be better described and handled. For Other things therd are currently no tickets opened

Is there a plan for blue xp to release a mobile version? Currently blue xp on my iphone looks terrible and not only not optimized, it‘s more or less not usable

[half summit]

ok wait a minute... Not sure if I'm misunderstanding all this stuff? Is the following not the case? (let's talk in examples, that's usually easier)
Let's assume you have a volume sized 70TiB on one of the aggrs of your AFF (it's the only volume on that aggr). The volume is thin-provisioned aka space-guarantee=none. Currently 65TiB physical space is used inside the volume.
You now attach this aggr to an object-store and configure the tiering-policy to auto for this volume. After the default 31 days all the cold blocks are getting tiered out to the cloud tier.
Let's say about 20TiB of data was cold on that volume. I would assume that the used-size inside that volume is now 65TiB-20TiB=45TiB plus the 1TiB metadata for the cold data (~5%) --> so about 45TiB used
Basically you gained 19TiB on your AFF aggr and since the vol is thin-provisioned you also gained the 19TiB inside the volume.

Is that not the case?!

Do you mean that the space is only gained in the aggr but not in the volume? Because it sort of needs to reserve the "now-actually-empty" space of the tiered out data for when it comes back?

So to actually gain the 19TiB inside my volume I would need to increase the volume by 19TiB? So I can use the additional space we gained in the aggr....
oh boy that makes it even more complicated to wrap my head around... 😵‍💫

[green ginkgo]

In short, yes. In long - think of it this way, (simplified..) volumes are made of files, files are inodes, inodes point to blocks. But not real blocks, blocks of the aggregate. When the aggr does raid operations, the volumes don’t change. If it’s expanding, changing from 4/DP/TEC - the volume is the same set of block pointers. And it has (currently 😎) capacity for 100TB of them.

Other raid operations include.. tiering cold blocks in and out of flash pool.. or fabric pool

Again, highly simplified, I’m sure some people are screaming internally at me, especially with regards to operational domains and what is and is not a raid op, but it’s easier this way 😛

[half summit]

🤔 ... ok

So in layman's terms: When I tell my customers "activate the tiering, wait for 31 days and you will have some more free space inside your volumes"... that's wrong?
What I should tell them: "Wait until the cold data is tiered off. You now gained free space in this magic land called "aggr". But not inside your vols. You now need to increase the size of your volumes by that amount to actually gain free space inside them."

Or just use autogrow...

Or even increase the vol-size to 100TiB because it's thin-provisioned anyway

[green ginkgo]

Exact ways of using the extra space is up to the end administrator and under some circumstance those are all valid options, yes

Really the benefit is when you have overprovisioned volumes

[half summit]

The difficult part is to explain to my customers how they can easily check how much free space they have left on their vols... Many of them don't grasp the concepts of aggrs, tiers, vols, etc. They just want to know how much space they have left on their AFF.
With VMware for example they just have their NFS datastores in vSphere and check the free space there. And then they come to the conclusion "ok I have x TB left on my fast storage". Which not really works anymore in this scenario.

I guess the best way is to tell them: Ignore the "free space" you see in the NFS datastores and simply always check the free space of your aggrs in System Manager. Maybe also in ONTAP Tools... not sure if it shows aggr space or only vol space.

[carmine cloak]

Any troubleshooting guides when it comes to figuring out exactly why a Cloud Connector deployment fails to AWS from BlueXP?

[placid talon]

The "aggregate" is a logical grouping of storage devices that have physical capacity limits... The "flexible volume" is a logical abstraction that can grow/shrink within the aggregate, but is limited to 100TB (with ASA exceptions). The total allocation of "logical" flex vols can exceed actual physical capacity, i.e. overprovisioning (like on basically every storage system in existence). "Magic" in the layers between volumes and aggregate capacity (deduplication, compression, compaction, etc) can allow for more actual storage usage than exists in the aggregate also ... So one can't just ignore volumes and look at the aggregate.

Overprovisioning is used in electrical installations too, for example... you can only plug so many cords into a socket before at some point you blow the fuse... the cords are the volumes, the fuse/breaker is the aggregate. 😉

[half summit]

Thanks for the explanation but I do understand how aggrs and vols work. 🙂
The issue is that you basically move data away from your vols when you're tiering to a cloud tier but that is not really reflected in the space-used value of the volume. The customer thinks he would get more empty space inside the volume but that's not the case if I understand it correctly. Basically your volume now consists of data being inside the performance tier (locally) and of data being inside the capacity/cloud tier (remote in a objectstore bucket).

This is totally different than without FabricPool.

The reporting of free space inside your vol can not be compared anymore with the free space in your aggr (local performance tier). I'm talking here about a 1-to-1 relationship between aggr and volume, meaning having only a single volume on your aggr

[placid talon]

Yeah, the logical allocation of the volume doesn't change, just it's placement in the expanded system of FabricPool tiers

[half summit]

That's how many customer do it with NFS datastores for vSphere with AFF C190 or so. Two aggrs (one per node), having exactly one volume per aggr.

[placid talon]

well, now you can use flexgroups and hope you have a lot of dead data, hehe

when the aggregate gets 90% full (per default) things get a lot slower, hehe

[vapid scroll]

Does anyone have the docker image name for the latest version of the OCCM connector in the docker hub repo?

[idle token]

Question about the new feature: NVMe read cache support on Single-AZ file system, For those previous created FSxO before the announcement, do I need to do anything in order to use the feature? or it would be already enabled on my FSxO instance?

[hollow warren]

You need the right instances for that

[idle token]

How could I find out if the instance that I have supports NVMe on Single-AZ? What instance can support it?
When I create FSx ONTAP, I don't remember I have the option of choosing which instance or maybe my remember was incorrect?

[spice barn]

For NVMe in Single-AZ file systems, the FS must have been created after November 28th of this year and have at least 2GB/s of throughput capacity

[idle token]

@spice barn Good. With this new feature for Single-AZ, it should improve the performance on reading. Could that or anything else we can do to improve the performance on writing. Thanks!

[spice barn]

What kind of read performance are you seeing now? What is the dataset?

[vale hazel]

I have a question about NetApp ONTAP AWS.
I have an existing NetApp volume which is being used by Windows EC2 instances using NFS. I want to use the same volume to be mounted on EKS, so that the pods can read from the volume. I was able to import the volume to EKS, by using tridetnctl import volume command. Now, how can I use the volume in different namespaces?

I tried creating a new pvc and trying to import again, but I get below error
Error: could not import volume: PV pvc-a4b360c7-7e0a-4e1d-bfb9-ee4c00bf0cc8 already exists for volume vol2 (400 Bad Request) command terminated with exit code 1

[idle token]

This is how the latency look like on one of volumes. They mainly are used for backing up Oracle DB's. As you can see, the writes caused most of latencies. While the new feature NVMe for read is good, but, in our case, writes are our concern. I am wondering if the new feature or what can we do to improve the write performance. Currently, the throughput is 512MB/s, and a lot of capacities.

[spice barn]

Those write latencies look very high.. is the workload rman or something else?

[idle token]

Yes, the volume is for storing Oracle rman backups

[spice barn]

What kind of throughput numbers are you seeing for the writes?

[idle token]

The throughput on writing for the same volume in the same time frame. How come 4.4G/sec could be reached under the provisioned throughput of only 512M/sec?

Total throughput:

[spice barn]

The throughput on writing for the same

[idle token]

The latency is very high as shown in the Latency graph. What about my original question that was what can we do to improve the performance and reduce the latency on writing?

[spice barn]

I wouldn't be too concerned with latency if the throughput at the same time corresponds to 4.4GB/sec

[spice barn]

If you want, you can email me (cwaltham at amazon) the file system ID and I'll take a look

[idle token]

@spice barn To your point of bursting over the provisioned throughput for a short period of time, understood that the bursting should last for a short period of time, but in this case, we could see 4.4GB/sec on writting for hours long, and happened in different days, how could we explain that?

[spice barn]

That usually occurs due to storage efficiency saving on the amount of I/O that needs to be written to disk.. but I must say that's a very large delta - maybe the biggest I have seen

[idle token]

I am not sure I understand that this type of writes would be part of I/O's within the provisioned throughput. By the way, I've already sent you my FSxO file system ID via private message, if you can check and find anything.

[spice barn]

Thank you, I will take a look

[random marsh]

Cost for Cloud Manager / Blux XP.
Can somebody please advice on ways to reduce the cost on Cloud Manager or now called Blue XP? We have surprisingly received a large bill on it. What items of this cost is made up of? I know connector instance should be the part of it, what else?

[carmine cloak]

For BlueXP, I'd like to manage an on-prem ONTAP cluster and I'd like to create a unique user account instead of using Admin. What roles and login methods does the account need?

[pure goblet]

login is a little different using a connector vs direct, but either way requires an admin user account
https://docs.netapp.com/us-en/cloud-manager-ontap-onprem/task-discovering-ontap.html

[carmine cloak]

I figured it would need the admin role, but login methods, would I just need ONTAPI and HTTP? I just want to minimize the user login methods.

[hollow warren]

#┊・hybrid-cloud - We want to implement an integration with SIEM/DL/NOC/Ticketing systems and we want your help to determine our priorities for these integrations. Which tools are you or your customers are using??? (e.g, ServiceNow, Splunk, Slack, ELK, Tivoli, AIQUM/CI/etc...)

[junior badge]

What kind of integration are you thinking of implementing? Strictly event/incident notifications, or something else?

[junior badge]

Adding to this for context - I see ServiceNow everywhere I go. Speaking from experience though, it doesn't seem to be implemented in a uniform way across environments. If you're thinking about like an inventory integrator, there can oftentimes be quite a bit of complexity in terms of which SNOW environment does or does not use the CI categories you expect to find, or they may be configured relationally different than you'd expect. Event/Incident notification is more clear-cut however. The main thing to consider is that SNOW doesn't have a universal webhook listener, so you have to provide some generic capabilities which the user then has to define how it will be ingested (i.e., I've run into quite a few cases where a middlewhare like Kong or Mulesoft is in place between the NetApp cloud service and the actual local SNOW service).

[lofty python]

Hi @junior badge , the context for @hollow warren question is around forwarding of notifications from BlueXP and specifically CBS. We are seeing customer requirements for alert/notification forwarding to 3rd party tools like Slack and Splunk. We also had some customers talking about SNOW for automatic ticketing management ( backup alerts from CBS automatically triggering SNOW incidents opened). We are trying to see if we can widen the scope and get a better understanding on how to prioritize 3rd party tools and interfacing methodologies ( Webhook only vs. dedicated plugins or other).

@junior badge do you know of any BlueXP/ CVO/CBS customers or prospects which have a strong requirement for SNOW integration for notifications?

[junior badge]

Unfortunately I don't at the moment - I will keep an eye out however and ping you if this comes up. Being part of the Cloud Insights team I do run into these kinds of integration needs quite a bit though. Granted, it's usually in the context of heterogeneous compute and storage environments - and more focused on inventory/asset management or incidents for capacity/saturation/latency issues than for say a job error or something.

[junior badge]

It certainly wouldn't cover every case, but you could look at leveraging the log ingestion (push X or Y events from CBS service to customer's CI tenant) and webhook capabilities already in Cloud Insights (templates for Teams/Slack/Discord/PagerDuty all already exist). Monitoring of CBS would of course then come with some kind of cost in Cloud Insights (as it would in Splunk or most other platforms you integrate to), but for inclusion of that to the overall estate I don't see how customers wouldn't want that. As an example, the ONTAP group has built a pretty great set of views and predefined monitors in Cloud Insights part of the 'ONTAP Essentials' section.

[lofty python]

thanks @junior badge looks like an interesting direction to implement a quick solution

[junior badge]

quick yes, and probably plenty of overlap between CBS and CI customers - but also good to consider a more general approach since there will be plenty without CI

[stable sage]

Has this been addressed?

[formal forge]

Is Refresh Token Generator working?
I am trying to use Cloud Sync API and this is the first time for using NetApp Cloud Central Services API. So I believe I need to get refresh token at first, but the Refresh Token Generator responses like this.

{"message":"Failed to generate refresh token."}

Not sure if it's related, I am a NetApp employee.
Thanks for your help in advance.

[dusky condor]

looks that latest deployment of blue xp connector caused an issue in mgwd because after deployment of 3.9.25 occm we have hundreds of error messages each second in mgwd log: 0000003f.00232fcc 01337ef2 Sun Jan 01 2023 09:37:59 +00:00 [kern_mgwd:info:3268] 0x82103bc00: 8803e8000009147f: ERR: NDU::mgwd: ValidationPayload::ValidationPayload - error reading showValidationDetails; [ANDU_LOG at src/ValidationPayload.cc:149 in 'ValidationPayload']
0000003f.00232fcd 01337ef2 Sun Jan 01 2023 09:37:59 +00:00 [kern_mgwd:info:3268] 0x82103bc00: 8803e8000009147f: INFO: NDU::mgwd: Error size = 0; [ANDU_LOG at src/tables/cluster_image.cc:1359 in 'setSubsystemField'] and there are more error messages:

and this caused a crash of mgwd: Message: spm.mgwd.process.exit: Management Gateway (mgwd) subsystem with ID 3268 exited as a result of signal normal exit (0). The subsystem will attempt to restart.

i opened a ticket for this:

the log files are now increased from daily 4197914 to 66087998 bytes

[random marsh]

No, we don't know what is this charge for in detailed items

[prime pilot]

Can FabricPool be used on any S3 bucket? We have Vast in our environment and out management wants to move to it 100% but I'd rather keep the front end NetApp for migration reasons.

[placid talon]

https://docs.netapp.com/us-en/flexpod/hybrid-cloud/cloud-fabricpool_fabricpool_requirements.html#platforms

[prime pilot]

Doesn't mention SG. Guessing the answer is we have to move 100% to Vast. Lame.

[frosty spade]

Hey @prime pilot I'd suggest engaging with your NetApp accounts team (or partner) to see if they can look into that for you on your behalf. Docs might say not supported right now, but there might be something on the roadmap that we can't speak about here but accounts team can help with that.

[half summit]

mentions SG here: https://docs.netapp.com/us-en/ontap/fabricpool/requirements-concept.html#general-considerations-and-requirements

[smoky shore]

Hey @random marsh following back up on this one. All of our pricing is published here: https://bluexp.netapp.com/pricing Are there any components you're referring to, specifically?

[dusky condor]

regarding bugs of blue xp: looks, that it reached a new level of existing unsolved bugs or what is the reason that netapp rolling out updates 3 times via docker during christmas /( new year week, which is usually a nogo? {
"current version": "cloudmanagerinfraprod.azurecr.io/cloudmanager_tiering:production-bf01b17-66-S",
"new version": "cloudmanagerinfraprod.azurecr.io/cloudmanager_tiering:production-ad26465-68-S", was updated on 26.12

then occm service was updated on 01.01 -> seriously? {
"current version": "cloudmanagerinfraprod.azurecr.io/docker_occm:production-3.9.24-247-S",
"new version": "cloudmanagerinfraprod.azurecr.io/docker_occm:production-3.9.25-253-S",

then also on 01.01. updates of cloudmanager tiering:
{
"current version": "cloudmanagerinfraprod.azurecr.io/cloudmanager_tiering:production-ad26465-68-S",
"new version": "cloudmanagerinfraprod.azurecr.io/cloudmanager_tiering:production-e6e10f6-69-S",

"current version": "cloudmanagerinfraprod.azurecr.io/cloudmanager_k8s_integ_v2_connector:production-56349c8-33-S",
"new version": "cloudmanagerinfraprod.azurecr.io/cloudmanager_k8s_integ_v2_connector:production-18666bc-34-S",
"same configuration": "false",

update of cloudmanager_k8s_integ_v2_connector

on 03.01 update of cloudmanager_tagging was done and an update of cloudmanager_storage_services

and finally on 08.01 update of cloudmanager_tiering was done (which does nor surprised me because we had a ticket, that cloud manager tiering is viewing bullshit) but then the update on 08.01 fails (invalid configuration) - later update works fine - and then also on 08.01 cloudmanager_compliance on 09.01 cloudmanager_cbs

i am sure some sysadmin is killing because of failing updates during his holidays 😉 since 08.01 we seeing issues with api throttling in blue xp timeline... maybe it have to do with the updates? we are seeing in azure logs that blue xp connector is request every minute 4-5x storage account key - strange - is there no caching or does it have to do with mgwd issue which we seeing fro m begin of the year (see my last posting)

i get more and more the impression, that updates goes through with all force in blue xp, updates having poor quality and every updates comes with new bugs, needs hotfixes....which causing new issues

someone urgently needs to check quality of blue xp updates

[dusky condor]

#┊・hybrid-cloud message ok looks that we are not the only ones who are having issues with the api throttling or the amount of api calls... i am curious about the bill...

[cursive siren]

is this the place for trident discussion?

[desert flume]

https://discordapp.com/channels/855068651522490400/1062048885847117935 There's an Automation forum here, or there is a Cloud one. I think Trident falls under automation.

[primal mirage]

I have customer who has turned on cloud tiering in ANF and not seeing any data tierd down to cloud tier

they are leveraging robocpy to get the data on prem on daily basis

using robocopy does it make all the blocks hot?

and will never be able to tier to cloud tier?

[half summit]

Not sure about ANF but what tiering policy do they use?

[primal mirage]

15 days tier to blob

was set up about 2 months ago

and see 0 data tiered or usage in the blob

[half summit]

So does that mean it will also tier the active filesystem? (like with auto tiering policy)

[primal mirage]

yes

[half summit]

or do you mean the latest 15 daily snapshots?

ok

[primal mirage]

active file system

[half summit]

then yes, robocopy would increase the temperature of the read files in my opinion

[primal mirage]

intresting

i thought tiering would ignore the sequential reads and not make the blocks hot

[half summit]

even if files are already in the cloud tier and they get read by robocopy they would get moved back to the performance tier, because I guess robocopy equals to "random reads"

[primal mirage]

wow

[half summit]

yeah but is robocopy really sequential?

don't take me for granted, I'm reallly not sure

[primal mirage]

i guess not

[half summit]

read here about the tiering and cloud retrieval policies: https://docs.netapp.com/us-en/ontap/fabricpool/tiering-policies-concept.html

not sure though how ANF operates... does it use the cloud tiering in BlueXP?

or can you configure the tiering in Azure?

[primal mirage]

configuration was done in ANF console

this feature is currently in ANF private preview

[half summit]

ah that explains why I cant find anything about it 😅

you could try Cloud Sync to get the data from ANF to an on-prem ONTAP or even any other SMB/NFS server

not sure this will do sequential reads though

[primal mirage]

welli best would be allowing snapmirror back from ANF 🙂

[gentle yacht]

You can discuss Trident in the astra channel in THE PUB section.

[cursive siren]

how do I get invited so I can chat in there?

[sage nest]

#┊・astra🔒 is locked. While we work on clearing up some of the lingering doubts, please use the #1062049057675169882 forum channel (with the Astra tag)

[cursive siren]

ok, I think I've got what I need for now so I'll just wait for it to open up

[sage nest]

We're in the middle of a bit of a reorganization of channels, so definitely hop in and ask in the new area if you need to.

[civic kernel]

Hello, I am not sure if this is the right channel to ask, but I have a question about ACC. We recently set up an on prem install of ACC as part of a POC, and we are having an issue with 22.11. Even though we set telemetry to false, the telemetry pods were still deployed, and are failing to start. We are getting a 404 when trying to connect to ACC, and we are wondering if the two things are related. Is there any documentation regarding what endpoints we need to whitelist on our firewall for telemetry? I haven't been able to find anything.

[worthy cipher]

Hello, What would be the best presentation (in Field Portal) for a customer who is not yet familiar with all of NetApp's cloud capabilities?
An overview of NetApp cloud solutions?

[neon badge]

search "Blue XP Cloud Portfolio". and I found 3. one per hyperscaler.

[idle token]

Are there anyway to show disk information on FSxONTAP? I could not find any, and understand it might be due to the fact that it is a AWS service. But, I am asking anyway to see if any tricky's there.

[true garnet]

no trick here ... unlike CVO, FSx for ONTAP does not offer the proper credentials to see the cluster settings (i.e. disks, aggregates, other SVMs, etc.).

so to your point, AWS managed service, so you can't see the inner workings 🙂

[idle token]

Right, there are some discrepancy on the capacity as shown on FSxO as 44TB and in Billing Dashboard as 50TB. From AWS FSxO console, it only shows me volume information. So, the figures may be different from FSxO or from AWS perspective on the disk level. What can I do to find out where is the discrepancy coming from?

[true garnet]

Are you saying that in the billing dashboard the provisioned ONTAP SSD storage is showing 50TiB (~54,975GB), but the CloudWatch metrics on the file system are only showing 44TB provisioned? As you are billed on provisioned capacity and a few other components as you see below:

[idle token]

@true garnet Understood the charge is based on the provisioned capacity.
"$0.125 per GB-Month of provisioned ONTAP Single-AZ SSD storage" from billing dashboard shows "51,200 GB-MO". However, as shown below, the summarized from "df -A -g" command on FSx ONTAP showed 45,604GB. How could I explain the discrepancy?

Also, since you sounds familiar with FSxO, I have one more question, if you can help me out:
What are these top 3 respective charges for from the list of billing dashboard the same as shown in your screenshot? Are these for reading from, writing to and the capacity itself of S3 that are all generated by or related to Tiering policy?

1. read requests to ONTAP standard capacity pool storage -
2. write requests to ONTAP standard capacity pool storage -
3. ONTAP standard capacity pool storage -

[radiant lantern]

Is it possible that the provisioned capacity is what is provisioned and the df capacity is lower due to efficiency savings? I know there was a change in the model for Keystone, don’t know it for FSx

[dusky condor]

Seriously? we have found two VMs which are deployed automatically without any notification from Blue XP Connector.... it's CloudInsights and the Compliance VM in Azure... Why Blue XP is deploying without confirmation and only with ONE CLICK absolutly hidden VMs? I wll discuss internal now, if and when we remove the rights from the Connector VM to Deploy automatically VMs... this is a NOGO!

[spice barn]

No, there is not. FSx is a managed service and this information is not exposed to customers

[spice barn]

Please open a case with AWS support and we can walk you through this!

[idle token]

We don't have Contract with AWS, and each call costs. I am hoping if somebody here has the same experience or have some ideas. I am going to wait for @true garnet, he sounds like have some experiences on that

[desert flume]

How do you have a FSxN instance but not have AWS support with it?

[idle token]

Probably depends on support model, here the cost is based on each call.

[desert flume]

Ah. As it's third party I have no idea...

I assumed it just came with FSxN bundled...

@spice barn @forest violet ?

[supple shale]

My understanding is all FSx support flows through AWS first. The exception being there's an option to "upgrade" by purchasing an enhanced license for backup via NetApp which provides a BlueXP Serial Number which can be used to open an NTAP case. But I've been wrong before

[forest violet]

Every AWS account comes with Basic Support which allows you to ask questions on re:Post (repost.aws) and get support there at no additional cost. You can upgrade to Developer Support for $29/mth (or 3% of monthly AWS spend, whichever is greater) and get the ability to open cases. Opening cases does not come at an additional cost. The only additional cost would be incurred if you are using AWS Managed Services or a 3rd party MSP on top of AWS.

[forest violet]

In regards to capacity consumption, @idle token , this doc explains the breakdown: https://docs.aws.amazon.com/fsx/latest/ONTAPGuide/choose-ssd-capacity.html#how-ssd-is-used But overall, what you're likely seeing is primarily due to WAFL reserve. In order for WAFL to be efficient, it requires 10% of space to be available at all times. WAFL reserve does not appear in the usable capacity of an aggregate. When you provision an FSx for ONTAP cluster, you're dictating how much raw capacity is needed for the performance tier. The documentation includes some examples of how to calculate the usable capacity and work backwards to the raw capacity that is required.

[idle token]

I don't have the answer for your question. But, we are looking into Blue XP cost too. I am wondering how could you find out these hidden VM's? what steps you need to go thourhg, if you can please share?

[idle token]

This is exactly what I am looking for without spending money on calls. Thank you! @forest violet

[dusky condor]

look for ressourcegroups with the name: cloudinsights, compliance.... it's a good question why blue xp is generating extra ressource groups. if you are using subscriptions with 100 other ressourcegroups

[idle token]

@forest violet I have one more question, if you can help me out:
What are these top 3 respective charges for from the list of FSxONTAP billing dashboard the same as shown in your screenshot? Are these for reading from, writing to and the capacity itself of S3 that are all generated by or related to Tiering policy?

1. read requests to ONTAP standard capacity pool storage -
2. write requests to ONTAP standard capacity pool storage -
3. ONTAP standard capacity pool storage -

[forest violet]

@idle token - Correct. Those would be incurred if-and-when you configure a resident volume with any tiering policy other than NONE.

[idle token]

Yes, you are right, there is "cloudinsights" here as well. How would you know "cloudinsights" being charged and it was generated by Blue XP?

[idle token]

Thank you for confirming!

[hollow warren]

BXP and the Connector is free. What cost you are the different capabilities (e.g., CVO, CBS, etc.). What are you using?

What do you have on Canvas and in Digital Wallet (search all sub tabs)?

[forest violet]

FYI... RedHat OpenShift on AWS (ROSA) support for Amazon FSx for NetApp ONTAP.... https://aws.amazon.com/blogs/storage/scaling-container-workloads-with-shared-storage-for-red-hat-openshift-service-on-aws/

[lapis oyster]

Oh I might have to play about with that and Astra

[forest violet]

FYI. If you're looking for a replacement to Docker Desktop due to licensing concerns, AWS has built a new OSS container client/dev tool called Finch that has most of the same functionality and works on x86 and ARM (Mac with Apple Silicon) systems: https://github.com/runfinch/finch

https://aws.amazon.com/blogs/opensource/introducing-finch-an-open-source-client-for-container-development/

[lethal tulip]

The growth of Spot's business leading to the acquisition by NetApp has been critical to our success.

Read more about it here and look for even stronger partnership motions in 2023!

https://aws.amazon.com/partners/success/spot-by-netapp/

[strong lichen]

There’s also podman

[smoky shore]

https://govdatadownload.netapp.com/2022/12/amazon-fsx-for-netapp-ontap-granted-fedramp-and-dod-srg-authorization/

I don’t think we’ve got any Federal folks in here but still cool to see nonetheless!

FSx getting FedRAMP’ed is a big deal!

[shrewd gazelle]

@smoky shore agree, it is a huge deal and open many doors.

[chrome scarab]

I've gotten errors that re a bit confusing and different per filer or way I try to connect. I have several 8200s, and most of them are running 9.11 or 9.12 (always at least one problem child: 9.8) When I have gotten the on prem to talk to either Cloud Ontap or FSx via BlueXP, then I get errors I think about licensing. Are any of the cloud connections into AWS covered in what seems to be a pretty vanilla set of licenses? I did get the S3 license for each of my filers, management approved since they weren't additional cost. All I want is to to either tier to S3 for cold blocks on one filer to free up space, or do my DR replication to the cloud to free up space that way.

[green ginkgo]

Hi there! the no cost S3 license is to allow ONTAP to be a S3 server. To tier data using Fabricpools you also need a capacity based license for it. https://docs.netapp.com/us-en/bluexp-tiering/task-tiering-onprem-aws.html#quick-start

[lucid remnant]

Not quite: FabricPool tiering to NetApp S3 storage (both ONTAP S3 and SGWS) require no license. You only need the capacity-based FabricPool license when tiering to non-NetApp S3 (AWS, Azure, or other S3 vendors)

[lucid remnant]

The S3 license that you got makes ONTAP your S3 server . What you want is FabricPool, i.e. S3 client. Whether or not you need a license for that depends on where you want to tier your data to

[green ginkgo]

Indeed

[lucid remnant]

My tip: find an S3 provider that uses StorageGRID, that way you only pay only once (for the used storage), and not for storage and FabricPool license. Or, if you have a few NetApps: build your own StorageGRID. Third option, if you happen to have a little-used NetApp somewhere (a no longer used NearStore maybe), use the S3 license on that and tier there.

[candid slate]

I've got a team that insists on using AWS for a product, then pulling it down locally to sgrid to beat it up afterwards.
Does cloudmirror just do SGrid to AWS? Or can it do both ways?

[smoky shore]

is Object storage a pre-req?

Would file be a option?

[candid slate]

Object is a prereq

Tl;Dr is cloud native app where people upload stuff than they want to pull it down for centralized processing and retaining.

We're trying to convince them to do it all internal

[shrewd gazelle]

trying to convince with performance and latency, etc...

[candid slate]

Yeah the goal is allow people to upload stuff and then pull down for long term on sgrid

[shrewd gazelle]

is sounds like one way...
"StorageGRID supports replication endpoints that represent S3 buckets. These buckets might be hosted on Amazon Web Services, the same or a remote StorageGRID deployment, or another service."

[smoky shore]

https://arstechnica-com.cdn.ampproject.org/c/s/arstechnica.com/information-technology/2023/07/ars-on-aws-01/amp/

Solid read!

[desert flume]

Cool and all, but no NetApp storage there. (Unless I missed it)

So boo :p

[green ginkgo]

My personal website started off on shared hosting with arstechnica back in like 1999

[smoky shore]

ICYMI here's my session with @silk axle at AWS Storage Day today in Seattle, celebrating the two-year anniversary and going over the latest and greatest updates!

https://www.twitch.tv/videos/1894274060?t=06h53m55s

[smoky shore]

https://community.netapp.com/t5/Tech-ONTAP-Blogs/FSx-for-ONTAP-supplemental-NFS-datastores-Technical-capabilities-and/ba-p/447124

A year since VMware announced support for VMC + FSx ONTAP, and some major enhancements have taken place!

[smoky shore]

https://youtu.be/xfwBC9WlJ-M?si=fWyYL9MoNMGzcWYY

Great video by @hollow warren showing how to use BlueXP with FSx ONTAP to build out complex architectures

[left orchid]

Hello guys, anyone with experience with Altavault? I have an equipment in a "emergency mode" ... im not able to boot it

[green ginkgo]

Any idea why it’s in emergency mode?

[left orchid]

No idea to be honest

[green ginkgo]

Psus, disk shelves, disks all look ok?

Any orange lights?

[left orchid]

Yes, but it seems it's not able to find some filesystems

[green ginkgo]

Paste me some error messages?

I’ll have a dig and see what I can find

AFK - @ me with your response, I’ll be back in a couple of hours but potentially someone else will see it and help

[left orchid]

Thanks

[cloud basin]

Question around Blue XP backup restore pricing . On the website (cloud.netapp.com) and presentations under pricing there is a price of 0.05$/GB/Month with a mention that endpoint storage cost is not included but when going to the TCO calculator page most simulation are below 0.05 (0.018$/GB/month with 1PB , 0.023 for 100TB ) including the storage cost , These are huge differences. So what is the right pricing and what is really included and not included?

[smoky shore]

@hollow warren might be the best to help here

[dreamy delta]

I can definitely help with this. Our BlueXP Backup Service has volume based pricing automatically applied in the TCO calculator (1-99 TB has a set discount, 100-x has higher discount, y-z has even higher). Now in order to take advantage of those discounts you have to either 1) purchase BYOL or 2) do a private offer through the cloud marketplace of your choice. If you’re backing up to StorageGRID that should be a separate convo for pricing that I can also help with). The back end storage, whether AWS, Azure, or GCP is a separate bill from that provider as we can’t dictate or calculate what you may be paying them. There are benefits of purchasing through marketplace if you have a committed spend (MACC, EDP, etc). Happy to discuss with you in more detail about best choice for your situation but boarding a flight right now. Please feel free to tag, DM, or email me at [e-mail] to get this worked out at your earliest convenience @cloud basin.

[smoky shore]

For those not keeping tabs, Microsoft has refreshed their popular Well-Architected Framework guide. If you're a heavy Azure user, there's something in here for everyone. From cost modeling, to workloads and solutions, and everything in-between.

[forest violet]

I've been going through a lot of the recorded re:Invent sessions from last week. There are enough on YouTube that you could likely watch 3 or 4 sessions per day and not have to repeat any until re:Invent 2024 starts next year. Going to post a few here as I go through them that I think some here may find interesting.

If you have devs using CloudFormation (or the Serverless Application Framework - aka SAM) to deploy and manage Lambda functions on AWS, this is a great talk on setting guardrails for devs and using tools like cfn-lint and cfn-guard to speed up deployment cycles and provide faster feedback. Also covers how to detect security vulnerabilities and ensure code verifiability before and during deployments. cfn-guard can also be used for Terraform. There's also a tflint Terraform linter that works similarly to cfn-lint. Similar concepts across both CloudFormation and Terraform.

NetApp has been making good use of Terraform and recently launched the ONTAP Terraform Provider. Very likely that IaC "de-linters" and guardrails are things of interest.

https://youtu.be/qlz15v-gHFI?si=tFfDrlBl9xXMvfWa

[forest violet]

The unsung heroes of the cloud: https://youtu.be/_Q8FyqI4a9M?si=DmTS3pGKPoo3eGUg

[atomic mulch]

anyone know if a client is wanting a serverless environment and uses azure ad and oauth only if we can support on premise file with a device like an a150 or something?

[desert flume]

It wouldn't be serverless...

[desert flume]

There is local user auth available.

[green ginkgo]

https://www.unisuper.com.au/contact-us/outage-update .. yikes

6 day and counting outage at a retirement fund due to a Google Cloud error.

everyone was skeptical it was actually GCP's fault and not a clueless admin, but it seems it actually was GCP

[acoustic grotto]

Is there a CVD/NVA or for FlexPod w/ No-FI?

Is that “technically” FlexPod express?

[smoky shore]

@dusky wolf Is this something you could help with?

[dusky wolf]

Hi @acoustic grotto , yes - you are correct. No FI is an express configuration. Let me find you a link to the most current validation...

@acoustic grotto here is the link for the FlexPod Express validated designs: https://www.netapp.com/data-storage/flexpod/validated-designs/

@acoustic grotto and the link to the FlexPod Express technical specifications: https://docs.netapp.com/us-en/flexpod/fp-def/fp-express-tech-spec_overview.html

[smoky shore]

Thanks J-Edge!

[lucid remnant]

FlexPod Express also uses FIs. They're just not a separate device but rather in the IO module slot of the chassis

ah wait apparently there's also a config with standalone servers.

FlexPod Express ≠ UCS Express ... that has confused me

[ancient trail]

Howdy, would anyone know what is going on with NetApp insofar as their certification of cinder drivers for RH-OSO v18? I see Manila Drivers for ONTAP but nothing for block (yet)

Pretty Curious about storagegrid as well as ontap for block as well as what protocols to expect support for in 2025 (for Red Hat OpenStack Services for OpenShift)

[lucid remnant]

OpenShift has a very good integration via Trident. NFS, SMB, iSCSI, FCP and NVMe/TCP. As for OpenStack, I have no idea.

[outer garnet]

At least three years ago, NetApp had Cinder drivers.

[faint cosmos]

AWS, Azure, or GCP?

[lapis oyster]

Any, all, plus others. Don't limit yourself 😆

[faint cosmos]

I've been playing with the free tier of Oracle's cloud. Doesn't seem as mature as some of the others, but the ARM instances are speedy and pretty cheap

https://www.oracle.com/cloud/compute/arm/

[dreamy delta]

There are a couple of short videos on GFC (Global File Cache) in the #877954017312006185 that I'd recommend checking out for a couple of use cases/needs. Between return to office mandates and cloud migrations w latency issues w file shares, we are seeing a huge increase in interest for GFC. At just $3k an edge, it is a very economical way for ONTAP customers to improve performance for remote/branch offices, facilitate collaborative file sharing, and consolidate their data footprint without compromising on performance. The #877954017312006185 channel doesn't allow comments, so I thought I would share what I am seeing here even though GFC is not necessarily a cloud solution.

[worthy mason]

Anyone facing Azure VM issue, looks like azure outage. I am getting below error.

{"code":"DeploymentFailed","message":"At least one resource deployment operation failed. Please list deployment operations for details. Please see https://aka.ms/DeployOperations for usage details.","details":[{"code":"ArtifactVersionNotFound","message":"No version found in the artifact repository that satisfies the requested version '' for VM extension with publisher 'Microsoft.WindowsAzure.GuestAgent' and type 'CRPProd'."}]}

This error is with Windows VM when you are creating or starting a vm.

[smoky shore]

@worthy mason what region is that? I can check

Are you deploying manually or via template/tool?

[worthy mason]

looks like global issue

https://docs.microsoft.com/en-us/answers/questions/588347/windows-vms-won39t-start-because-the-guestagent-ex.html?page=1&pageSize=10&sort=oldest

[smoky shore]

If anyone has any questions about any of our cloud portfolio, I'll be around all day!

[final widget]

German content, Google Translator is your friend: https://www.heise.de/news/Die-Zukunft-der-Cloud-Public-Top-Private-Flop-6226387.html#nav_sonder_clouds__1

Future of the cloud: public cloud: top! private cloud: fail!

[dawn rover]

Yessss Dark Theme for Cloud Insights

[desert flume]

@grand lotus let's move this here

10:37 AM] DrFreud: Does Cloud Backup service or snapmirror cloud offer an index?
[10:37 AM] Paul Stejskal: Index?
[10:38 AM] DrFreud: Index of the stuff backed up. So you can retrieve what you want
[10:38 AM] DrFreud: catalog etc
[10:38 AM] Paul Stejskal: I'm probably not best to answer on this as I don't know 100% off top of head, but I think it is still regular SnapMirror under the covers.
[10:39 AM] Paul Stejskal: So...no
[10:39 AM] DrFreud: so what exactly is the benefit of cloud backup service
[10:39 AM] Paul Stejskal: That's a good question. I'll see if I can dig that up. Give me a minute.
[10:39 AM] DrFreud: Thanks Paul!
[10:44 AM] Paul Stejskal: Looks like I was wrong about how it works. It backs up to any S3 repository using Cloud Manager the ONTAP snapshots.
[10:44 AM] DrFreud: right but the catalog is the question.
[10:47 AM] DrFreud: Looks like there is a catalog
[10:47 AM] Paul Stejskal: Yeah I'm seeing that as well.
[10:49 AM] Paul Stejskal: There are 2 levels of restores, entire volume and single file. Single file has a browsable catalog that lists the files in a specific snapshot and you can choose which one to restore.
[10:50 AM] Paul Stejskal: This seems like a cheaper way to do backups to a S3 tier than say having another data center with another set of filers.
[10:50 AM] Paul Stejskal: If something happens from the source cluster, it can be restored still.

@grand lotus honestly it seems like this is a good way to manage your backups. I'd recommend it. I'm watching the internal TOI on it now and it's probably one of the best backup products I've seen in a while!

[smoky shore]

Cloud Backup Service does include a catalog that is indexed, correct.

https://cloud.netapp.com/cloud-backup

Free 30-day trial is available to check it out (unlimited)

[sturdy oak]

Hi NetApp may I know what is the requirement to get a POC started for CVO for Azure?

[desert flume]

https://cloud.netapp.com/ontap-cloud <-click start free trial here

[dreamy delta]

@sturdy oak the first step is getting Cloud Manager (our SaaS control plane for almost all things NetApp Cloud) Here are specifics for Azure to get the connector installed and such. Happy to point you towards a NetApp internal specific resource if needed! Network requirements:
https://docs.netapp.com/us-en/occm/reference_networking_cloud_manager.html#endpoints-to-manage-resources-in-azure

Step 1 - Creating a Cloud Central account
https://docs.netapp.com/us-en/occm/task_setting_up_cloud_central_accounts.html

Step 2 – Deploying cloud manager connector in Azure
https://docs.netapp.com/us-en/occm/task_creating_connectors_azure.html

• To deploy a Connector, you need to provide Cloud Manager with a login that has the required permissions to create the Connector VM in Azure.
You have two options:

1. Sign in with your Microsoft account when prompted. This account must have specific Azure permissions. This is the default option.
   Follow the steps below to get started.
2. Provide details about an Azure AD service principal. This service principal also requires specific permissions.
   Follow the steps below to get started.

Step 3 – Discovering your on-premises cluster from the cloud manager (if interested)
https://docs.netapp.com/us-en/occm/task_discovering_ontap.html#checking-for-on-premises-clusters-that-have-not-been-added-to-cloud-manager

[dreamy delta]

@grand lotus we have a TCO Calculator that can help you gauge overall cost of the solution online. https://cloud.netapp.com/cloud-backup-service-tco-calculator?hsCtaTracking=4f8508f7-f8cc-49b1-9b51-a32ab26ae995|b436c2fc-4084-4ce3-b822-d5be3d0cc1e0 Cloud backup Service is by far the most efficient way to backup ONTAP data (because it was built to do just that ;-)) but is also quite economical. We just went live with support for Glacier as a destination for archival backup use cases too which is a fraction of the cost of S3 IA!

[smoky shore]

Deploy one from Cloud Mgr and select Freemium for a fully-featured experience up to 500GB!

[dreamy delta]

Yes I meant to finish that thought! netapp.tv has some awesome reference demos available to stream for free and follow along with too

[sturdy oak]

Thank you!!

[smoky shore]

https://techcommunity.microsoft.com/t5/azure-architecture-blog/disaster-recovery-of-aks-workloads-with-astra-control-service/ba-p/2948089

[smoky shore]

https://github.com/NetApp-Automation/fsxn-openshift

If there are any ROSA users, the RedHat team has been hard at work integrating it with FSx ONTAP

[crystal jacinth]

Do you have any information if the admin account is a must to add on-prem clusters to cloud manager? Or is a seperate account usable which has less permissions?
https://docs.netapp.com/us-en/occm/task_discovering_ontap.html#checking-for-on-premises-clusters-that-have-not-been-added-to-cloud-manager

[dreamy delta]

https://docs.netapp.com/us-en/occm/concept_cloud_central_accounts.html#members

In Cloud Manager, a workspace isolates any number of working environments from other working environments. Workspace Admins can’t access the working environments in a workspace unless the Account Admin associates the admin with that workspace.

However, the admin account will be required to configure the Cloud Manager Connector if not already done so. That role must have the required backend permissions applicable to the cloud provider of choice or access to the on-prems of course. The acct admin can then create the workspaces and associate users, those users can create and connect to working environments. Please let me know if you need more info about that!

[crystal jacinth]

Sorry I might have been not specific enough. Do I have to use the on-prem cluster admin´s credentials to add that cluster as an on-prem cluster within cloud manager? Or can I create another user with less permissions on the on-prem cluster?

[dreamy delta]

@crystal jacinth You will need admin level access to the on-prem cluster for adding them.

[hot lagoon]

Hello everyone. We are running CVO 9.9.1P3, and I'm doing some volume moves from a GP2+FabricPool aggregate to a brand new GP3+FabricPool aggregate, and I'm wondering is there is a way to tell how CVO will handle a move? All of our volumes have their tiering policy set to all, but the moves that I've tried so far don't appear to be just moving meta from the performance tier...

[desert flume]

Yes.

It should be showing in qos statistics workload resource disk/cpu show

[hot lagoon]

Thanks Paul. It looks like it's working correctly. Disk% is 0, and the CPU is busy.

[desert flume]

Cool.

[sour stag]

Hi Guys,

I need some advice if this will work.. On prem i have ONTAP and VMware (NFS datastore)

Can we snapmirror to CVO in Azure ongoing and when we need to perform DR tests.. Spin up AVM (Azure VMware) not VMC ( VMware Cloud) and mount the NFS datastores into AVM..?

We also have other SVM's doing inguest ISCSI luns i assume i can also mirror those to CVO and then mount those luns in guest to the AVM hosts ?

[desert flume]

Yes. ONTAP is the same on AFF, FAS, CVO, or Select.

[smoky shore]

@sour stag Yes. Even better, you have options. If you need more horsepower you can look into Azure NetApp Files (it's an A700 big boi under the covers) if you just want managed storage. Then you can use Cloud Sync to move data there. It will also maintain ongoing syncs and encrypt in-flight.

Shameless plug: Check out my ANF videos if you need an introduction and breakdown of the architecture...

Part 1 - Intro: https://youtu.be/NnXJ9DFACZ0
Part 2 - Design & Architecture: https://youtu.be/6p0joMfJmgY
Part 3 - Snapshots: https://youtu.be/_ctpEtNkciQ
Part 4 - Putting it into Practice (Demo): (coming soon)

[sour stag]

We have been talking to our local NetApp guys here in Aus and it seems GCP is the only cloud provider who allows you at attach a NFS volume from CVO into VMware in Azure or AWS.. So you need to use something like Cloud Sync.. We would of probably liked to keep it to CVO as we can control the SVM and its block level replication. How does Cloud Sync handle VMware NFS Datastores ?

[desert flume]

I think it's more for keeping files in sync as VM datastores are more locked.

It might work, but I don't remember if it does or you shou.d

You could probably just SnapMirror or something like that just as easy or easier.

[sour stag]

but you cannot snapmirror to ANF right..

which is why we need CVO and then we comeback to the issue of CVO not allowed to NFS mount into VMware Azure..

for this customer we have a fairly generous RTO/RPO.. so we could possibly use something like Veeam/Snapcenter for VMware to take a nice quiesced snapshot on NFS datastores and then if Cloud Sync could pull the data from that snapshot and stick it in ANF and then we present that to VMware on Azure that kinda works. But im not sure how we get back

[desert flume]

I mean, ANF is just A700s on the backend. I can check.

[lapis oyster]

It’s in preview in VMware Cloud in AWS and FSxN to do ONTAP NFS data stores. https://blogs.vmware.com/cloud/2021/12/01/vmware-cloud-on-aws-going-big-reinvent2021/

You can then SnapMirror in/out of FSxN as required

[desert flume]

I guess SM is no go.

Maybe you could lodge a special request?

[golden kestrel]

ANF/AVS Datastore support is coming, you can use HCX for replication or other 3rd party solutions - https://azure.microsoft.com/en-us/updates/azure-netapp-files-datastores-for-azure-vmware-solution-is-coming-soon/

[sour stag]

Our thinking for sticking with NetApp and snapmirror for the replication technology is we don't need to keep VMware spun up 24x7. So on-prem they have NetApp and VMware.. We would snapmirror the required datastores to a permanent CVO instance in Azure always on. Then if they needed to invoke DR/DR-test we would use ansible and spin up VMware on Azure and build out that environment all automated and then mount in the NetApp NFS datastores from CVO.. import hosts and boom your up running in DR on Azure.. Once you are done you shut down hosts and snapmirror resync back to on-prem and you back, Destroy VMware on Azure environment as its not used outside DR. VMware in Azure is super expensive and doing this way its only around $1800 (Aus) a day for VMware..

We have quite a lot of "Traditional" customers who do not have any NetApp in their environments and they have large VMware installations with other storage vendors. By showing them you could do this and automate 95% of it would be pretty appealing and make it a lot easier to get NetApp in the door. Also the cost savings by not having VMware on 24x7 when its not used would be significant..

[smoky shore]

Is ExpressRoute factored into that cost?

[verbal furnace]

Hello, we recently spun up CVO in Azure for NFS sharing and I am wondering what settings are recommended for mounting NFSv4.1 on RHEL hosts. The default command given in the CVO portal was "mount <ip>:/<export> <dir>" while when I used Azure Netapp Files, the mount command had more options specified: "mount -t nfs -o rw,hard,rsize=65536,wsize=65536,sec=sys,vers=4.1,tcp <ip>:/<export> <dir>". Are there similar settings that should be used for the CVO command or is the default CVO mount command fine? I know those settings depend on the workload and I'm not as familair with the NFS protocol but was curious if there were better options to use that apply to most use cases than just the default.

[desert flume]

You're right, depends on the workload.

😄

There is the NFS TR. I think there is a KB. 1 minute

https://www.netapp.com/pdf.html?item=/media/10720-tr-4067.pdf

This isn't a CVO specific question.

Like here is a KB on mount settings for databases: https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/What_are_the_NFS_mount_options_for_databases_on_NetApp

[verbal furnace]

Thank you, I came across those but will take a deeper look at them.

[smoky shore]

On GCP I subscribed to Netapp Cloud Volume with a billing account 1 and created some volumes. Now I changed that billing account to a reseller, which cannot subscribe to Netapp Cloud Volume anymore

@torn kernel It’s a different account, so it has its own CVS and Volumes. You can re-enable CVS on the new account and replicate the volumes over, or instead of a whole new billing account, just creator a new user and delegate them access to your existing account with the volumes

[torn kernel]

Thank you for your response Nick.

The thing is, my current project is using a reseller account, which I cannot subscribe to marketplace service (Cloud Volume)

So how do I manage the created volume?

[smoky shore]

@torn kernel Are you saying that you can't get to CVS in the Marketplace when you're on that billing account?

Can you show me a screenshot? That shouldn't be a thing

[torn kernel]

here you go

[smoky shore]

Oh that. Yea they don't want Managed Service Provider co-opting the service and reselling it. You'll have to use a non-reseller account. Apologies for the confusion earlier.

[torn kernel]

Yeah I thought so. However the volumes were already created before. So I have no idea how to manage the existing volumes.

[smoky shore]

@torn kernel were they created under your personal billing account? You should just be able to log in with that account with an associated user and access them.

[torn kernel]

The problem is we have a whole production project to manage and charged via reseller billing account. We can't just swap billing account every now and then.

Is there a way to migrate the existing volume to a different project?

[smoky shore]

Not if they’re different billing accounts, as far as I know. You could use our Cloud Sync tool to move the data between… let me find out from PM/eng tomorrow and get back to you. They might have a way

[torn kernel]

Thanks Nick. Hope to hear from you soon.

As far as I know, the volumes are tied to a single GCP project.

[thorny cradle]

Hello Everyone I hope this is the correct place to ask a question, I have a client using NetApp Virtual Desktop Service to Access an Application called ProContractor. whenever they launch any hosted apps they get an error stating: This Computer Can't Connect to the remote computer. Try Connecting again. If the problem continues contact the owner of the remote computer or your network administrator.

I am the administrator and not sure what to do lol the tech that did the migration to the cloud is on vacation

[desert flume]

Oh no! Is that the Spot produt?

You'd probably be best working directly with support.

[thorny cradle]

Thank you Paul!

[desert flume]

Yw

[torn kernel]

Hi Nick, do you have any updates for this?

[sour stag]

Can i map an existing LUN through NetApp cloud manager ? I can setup a mirror from onprem to CVO using the CM GUI, i can break the mirror using the GUI.. i cannot find how i map a LUN in the volume i just mirrored using the CM GUI ?

[smoky shore]

Are you able to see the lun within CVO?

[sour stag]

yes

everything is fine with system manager GUI and SSH

i just thought you should be able to do simple things like map LUN or create NFS/CIFS share from a mirrored/broken volume you have sent up to CVO with Cloud Manager GUI.. you can create a LUN or CIFS share/NFS if you create a new volume..

These are the only options you have of a broken/mirrored volume from onprem to CVO using Cloud Manager..

there really should be an option in there to map LUN

[unreal obsidian]

Hello Everyone,

I signed up for the NetApp SaaS Backup 30-day trial yesterday but have not received an email for the activation code. The website stated that an email should have been sent within 30 minutes. Does anyone know the contact information to receive help finding the code?

[half summit]

Do you have a NetApp login for the support site?

If yes, please read this: https://mysupport.netapp.com/info/communications/ECMLP2881821.html

Also this: https://mysupport.netapp.com/info/communications/ECMLP2881902.html

How did you sign up for the trial? On which website?

[unreal obsidian]

Yes, I have a support Login. Thanks for sharing the links! I really appreciate it.

[unreal obsidian]

I used the following link: https://www.netapp.com/forms/san-health-check-inquiry/saas-backup-ms-365-30-day-trial/

[half summit]

Well that's unfortunate

@sage nest Can you maybe bring that up to the right people?
The FAQ here (https://fieldportal.netapp.com/content/1734871) states that: The free trial request form was deactivated on November 1, 2021.

Since there was no public EOA announce for non-customers...

[sage nest]

@distant kelp is this ^ something you're able to look at?

[distant kelp]

The docs folk are aware and are removing the link. I will check in with them.

There is an EOA on the support site, but you’re right, I think comms was only sent to existing customers.

[sage nest]

🙂 Thanks Phoebe!

[blazing palm]

hello, i am having trouble with altavault which do not have support. replication data not decrease it looks like it stuck. when i apply cloud setting i am getting "Error connecting to the cloud: Failed to put file: /var/tmp/conn_check_file".

did anyone saw error like this in altvault by the way storagegrid show connected

[desert flume]

Hey. Did you get a case opened @blazing palm ?

[blazing palm]

No i do not have support for this product :(

[desert flume]

Ok let me see if my contact can reach out.

[desert flume]

He's pretty tied up. Sorry. 😦

[hollow warren]

You should move away from AVA to Cloud Backup, it’s soon reach EOA/EOS

[blazing palm]

I know that but before that i need to fix the error to continue replication

[frosty spade]

Do you have access to the support site to view KB articles? You would need at least 1 system with active entitlement.

[frosty spade]

Reason I ask, is there is 1 article that might be relevant to you - https://kb.netapp.com/Advice_and_Troubleshooting/Data_Protection_and_Security/AltaVault/AltaVault_GUI_Replicated_Data_section%2C"Metadata_replication_is_in_progress"

[blazing palm]

Thanks for kb i am have access support site, i will try then let you know, thanks again:)

[blazing palm]

I did everyhing in the kb but still getting error about file busy :(

[frosty spade]

Sorry to hear it didn't help. Normally our Support folks could take a deep dive into what's going on behind the scenes but in this case that option is not available. There may be some other folks in the channel whom might have some more advice.

[blazing palm]

Thanks for the good wishes :)

[cold gate]

Hi, trying to get the free trial for NetApp Backup to Cloud (for VMs in this case) running and when I get to the step to connect to the SnapCenter Plug-in for vSphere I get a 404 page not found. As far as I can tell nothing is even trying to connect into our local environment. I have noticed that options are different between cloudmanager.netapp.com and the instance of Cloud Manager I set up in Azure as in our Azure instance doesn't have several options like VM or Kubernetes listed under Backup and Restore. Any suggestions? Thanks!

[desert flume]

I'd open a ticket on the cloud.netapp.com chat or with Support. That's probably a backend issue that can only be debugged on the backend.

[smoky shore]

I agree with Paul here. But please keep us posted, as I’d like to know the resolution you end up getting. That’s a first for me too.

[lapis narwhal]

My internal security vulnerability scanner reported that my CVO connector has an HTTP proxy. Is this true?

nevermind. KB article says its no longer needed after 3.9.12 and can be removed.

[strong vigil]

I'm aiming for the NTAP Cloud Champion Architect. 🤓

[sour stag]

has anyone setup Global File Cache and use it in real life ? Also does it have the ability to connect back to onPrem as the master fileserver/SVM or do you need a cloud version like CVO ?

[lapis oyster]

Yes we can use on-prem/hybrid as well as the cloud offerings like CVO/FSxN/ANF etc. I can connect you with the right people on the team if you want to discuss it further?

We also have Cloud Volumes Edge Cache (CVEC) https://cloud.netapp.com/cloud-volumes-edge-cache

[smoky shore]

https://www.linkedin.com/posts/george-fridman-81616712_github-gfridproduction-ready-eks-service-activity-6954391003659128832-LFZ6

[sour stag]

That is pretty slick