HyperScale X Q&A

hi guys, i just deploy a hyperscale with :2 x1gbe - bond1- management network2 x 10 gbe - bond2 - data protection network 2 x 10 gbe - bond3 - storage network the customer configured the data protection network on the esxi nodes like - commvault-backup network (kernel adapters) the commserve runnning on the pre-installed (ova) appliance on the Management Network and we deployed a linux access node (ova) on vsphere infra and configure the eth0 on the Management Network, but all backups now run maximum like on 1gbe network. What is best practices  ? deploy access node on the data-protection network ? (im using hotadd) or add secondary adapter to linux access node?  thank you 

Hello Team, Hello,I have good experience on Commvault environment upgrade but I want some advices about HSX environment where I a not very familiarCommCell version is 11.34.47 End of life target version is latest 11.36Current HSX Clusters info :whichCommit.shBranch: QBD[v-5.0.8.2]rpmhedvig-common-5.0.8.2_3820_f13a4ee0db-1.el8.x86_64hedvig-cluster-5.0.8.2_3820_f13a4ee0db-1.el8.x86_64In a standard environment, I update :1 - Commserve Linux PROD2 - Commserve Linux DR3 - Media agents one by one, and I connect to each media agent to perform the update locally not push4 - Access nodes5 - Index cache Servers4 - update clients I thing for HSX environment it is the same order :1 - Commserve Linux  PROD2 - Commserve Linux  DR3 - HSX Cluster 1 3 - HSX Cluster 24 - Access nodes5 - Index cache Servers4 - update clients

Hi, I have deployed a Hyperscale X with 4 N24 nodes and in Command Center there is a toggel to enable File Indexing for VMware backups. If I read the documentation correct, toggle this switch will automatically deploy the Index Store package to one of the Media Agents associated with the Plan. At the same time there is a recommendation to setup a separate MA as Index Server and not to use the main Media Agents (in this case the HS-X nodes) I’m having difficulties to find the best practice and sizing guidelines for this setup. I would want to avoid installing a bunch of separate Media Agents and use the Hyperscale for this if supported. What is supported/recommended and best practice? Regards,/Patrik

Hi! We recently deployed 2 HyperScale X Reference Architecture clusters at a customer, and found out that: Ransomware Protection is not enabled by default Linux Firewall is not enabled by default In my opinion, in current times, it is strange that by default the solution is so insecure.Why are the firewall and ransomware protection not enabled by default? And will that be addressed in a future release?

Does Commvault support 100 GbE, 200GbE and 400 GbE Network for HyperScale X reference arcitecture  and commserver.On a 400G network, what is the maximum backup speed achievable? Can we reach 40GB/sec? Would using multi-streaming or multiplexing across the mount points help in achieving this speed?

Hello allplease can you help with this issue there the A nad PTR records exists ?Thanks Pavel 

Hello fellow CommVault users,I run a three node HyperScale 1.5 media agent on HPE reference hardware and recently one of the drives failed. I’ve replaced the failed drive and have started the recovery process to rebuild the contents of the drive.Question: Is there a command that can be run to show the percent complete on the rebuild process?Thanks in advance for any information.Ken 

Hi, I am currently going through the documentation to plan the update but some points are not clear to me:https://documentation.commvault.com/2023e/expert/migrating_hyperscale_x_version_on_existing_nodes.htmlIs this paper for both Hyperscale (appliance and RA)? The migration process migrates and reboots each node in the cluster from RHEL7 to Rocky8 sequentially. This process may take several hours to complete.Can I continue to perform backups while the cluster is being updated? (I'm thinking of the database log backups)We have a 6-node cluster, does anyone have experience with the duration? ThanksMichael

I am currently looking to automate some of our tedious task we do manually on commvault. One of which is removing the user DBs from being backup via commvault and deconfiguring a server from being backup if there are no active DBs on that server and vice versa. I started looking at this page here for available API that I can use to help with this PowerShell function I am looking to create. Was wondering if this is something doable as some of the API I saw on that page, I didn’t see one to remove user db from being backup if there are configured to be backed up. Any guidance with this would be greatly appreciated.

We recently received a fresh HSX 2.2212 ISO download from commvault supporth here. Whenever we try to mount the ISO and image servers after we have configured them, the installation has been failing. We have had some success install on 3 of 12 servers with the same configuration and ISO, but the remaining are always failing due to timing out or just fails immediately.  We are using Idrac to mount the ISO and running the install.   

Hi Team,One of my customer having seventeen-node HSX 3.x cluster. They have 12 drive tape library and planning to connect 6 drives out 12 with HSX through to perform aux copy.Here is my doubt to outline the solution.Do I need to zone all the MA with tape library or only six MA? If six MA, how do I choose those six MA alone? Can you share your thoughts/inputs. TIA,Mani 

Dear Community,Have a nice day. I got a question from security team and i need to know your thoughts,Can we install FireEye EDR on HSX Nodes as they are linux based servers ? Thanks in advance 

Dear Community,I try to configure airgap interval on HSX cluster from 7AM to 4AM next day to allow aux copy for 3 hours only , when I run command “./cvsds -c 7:00 4:00” I get “Failed to get storage services status”.Kindly need your support.

Hi All,A customer asked me about the size of the Ethernet frames used between 2 HSX clusters during Auxiliary Copies.From what I understand, the customer is concerned that the frames are too small. And that in relation to certain network components, the throughput obtained is not very high if the frames are small.And he asks me to put the question to Commvault ...As I understand it, this is independent of Commvault. But I'd like to check my understanding to the community.Ethernet has a minimum frame size of 64 bytes. It also has a maximum frame size of 1518 bytes. (if jumbo frame is not configured).For Auxiliary Copy jobs between HSX clusters, Commvault uses 128kB data blocks.These data blocks are sent to the Rocky Linux kernel, which splits them into 1500-byte segments (if jumbo frame is not configured) and sends them to the Ethernet interface driver.Can you please confirm my correct understanding? Is there any reason why Ethernet fr

Hi alli’m checkin a configuration on node of HSX and i can see that we have an SSD disk (2.9T) not in use, a NVME disk (6T) with ddb and IC and the last one SSD disk (2,9T) for metadataHedvig.It is normal to have a disk not in use ? maybe it could be useful in case of disk fault ?  thanks in advancevincenzo

Hello teamHave a question about HSX RA disks for Boot and MetadataFor boot i saw reccomendations Create a RAID1 volume, but for MetaData i didn’t see anything, jus 2 dedicated NVME\SSD disksI’m trying find some info see below for Cache (CVFS/Hedvig) as i understand we use EC 4+2, so for this disk we doesnt need a RAID1 - yes or no?for Logs (Index cache/DDB) can we used RAID for this disks or its just dedicated NVME\SSD?  and one more question whats happening if a metadata disks will go offlain or crushed in one or more nodes? 

Hi, We are adding a HSX node to a cluster with 4 nodes with 18TB drives. 18TB drives are no longer available for Dell server so we have to go with 20TB drives, that according to doc will be formatted to 18TB usable volumes. Is the requirement for 768GB RAM based on usable drive size or the actual size?Are we able to run these servers on 512GB RAM as the drives will be 18TB usable to the Storage Pool? /Patrik

Is there a command you can run to check the status of all Hedvig Nodes in a cluster?In gluster it was “gluster peer status”Thanks

Hi,I have a question, where the answer just might be trial and error and worst case scenario is performance loss. But other teams experienced a crash of there ESX systems after modifying this setting. So, I thought let me first check here.From a legal obligation to take measures to save energy (in a data center) we are obliged to set servers to so-called balanced mode where possible.We have a hyperscale environment, 3 Dell PowerEdge R740xd servers.If we enable the setting; Power Factor Correction (PFC) here on all 3 servers, could that possibly cause problems, this also in combination with CommVault? Regards,Danny

can we use hyperscale x node as access node from hyper-v backups.I dont see it in list of access nodes

we are trying to do OS upgrade the HSX from the existing 7.9 to 8.9 to be able to add additional new nodes (HSX ISO 3.2312) we tried to upgrade the OS using the Command center and CLI both showing upgrade completes successfully however the OS version is still 7.9 i used the below reference for the OS upgrade operation. https://documentation.commvault.com/2024/expert/installing_operating_system_updates_on_hyperscale_x_reference_architecture.html

HiMapped ESX Datastores via FC to a HSX cluster - reason behind is using SAN Transport mode.Had to reboot the nodes as rescan-scsi-bus.sh didn’t work and couldn’t see the new devices.However, now that I see the disks, backups are not running properly with “failed to read disk over SAN transport” error.Any hints/tips on this? Couldn’t find much around as this is a storage related topic.Appreciate the help around.ThanksIvan

Hi guys, Is there a way to remove the welcome banner below so its just a blank screen.  

Hello Everybody, During a meeting with customer where two new HSX clusters were deployed they highlighted the fact that Rocky Linux release 8.9 was out of support.Rocky Linux Release and Version Guide - Rocky Linux Wiki  Do we have any plans in future to allow the update of this OS to a higher version or this is something customer should not be worried about?Thanks in advance

Hi,I’m trying to update the OS for my Hyperscale X; however, in the documentation they stated that we should use KVM console instead of using SSH. I opened the KVM console on 1 of my nodes, and I can’t find the log in option. I wonder which button I need to press or which command I need to put there in order to log in?