HyperScale X Q&A

Hi All. I have a HyperScale 1.5 setup, with HBA cards installed on the nodes to be able to do VSA SAN backups.I have configured the multipath.conf to use round-robin, so the HBA card will utilize both ports on the card I have noticed that even though round-robin have been configured, it seems that only on port is utilized and that the port being used is then being maxed out.Do I miss somewhere in the HyperScale configuration where I can set the multipath setting or does anyone have any suggestions on how to utilize both ports on the HBA in a HyperScale setup? Regards-Anders

This is my 6th or 7th RA deployment and first time I have run into this issue. I ran through the process up setting up the 25G bonded networks (one for DP and one for storage) got the system into the commserve and then went to have it update to the latest version. cvupgradeos.py ran fine for hedvig updating to 4.7.13, but when I went to upgrade from 11.24 which was pre-loaded on the nodes the first server succeeded, but the second hung. It didn’t give me any feed back. I opened a case and have been waiting to get any feedback.  I am unable to ssh between nodes now and occasionally can using the DP network. I verified that there is nothing wrong with the network stack. Ping, netstat all is good. The services are up and listening. Firewalls are disabled. I ran a nc -vvv (node name) 22 and it connects but hangs. Connection is not refused, nor terminated at the other end. Commvault on node one shows updated to the latest version but will not commu

Hello Team,One of our requirement is that the current commvault is running hyperscale environment and Commserve also running on hyperscale. Now the hardware want to refresh it, so we are planning to go new hyperscale with new commserve or will migrate existing commserve with all backup data and existing policies. Please advise how to approach this migration. Thanks in advance.   

Hello there, I have a customer for which we deployed HyperScale X clusters (11.32) for which we used the CommServe Registration network along the storage network. But now, they want to separate backups on a dedicated network. So now I’d like to add that management network, but I can’t find a documentation page on how to achieve it. I think this was accomplished in some config files on the nodes, but can’t remember which and how.Anyone has a hint on how to achieve this?Many thanks in advance.

Dear Support,We have a requirement to offload VMware backup processing to HyperScaleX server, we do not have SAN in the environment. The storage server presents the disk to ESXi hosts over the ISCSI method. Can we use the same disk presented to HyperScale X over ISCSI method and perform SAN mode backup? Not sure this is supported, if supported using which network label (Data Protection or CS Registration) the HyperScaleX can see the disks. Please help.

So these deployments have been a real hassle won’t get into previous details but the current one hsxsetup wizard will not finish forming a storage pool. Main difference with this one is we used a known service account and password to authenticate with the Commserve instead of the code method in Command Center(account has Master permissions and more on the Commserve) Logs say the account is not an issue and according to Commvault the cluster is formed(3 nodes) and the nodes are registered in the Commserve as MA’s(they have duplicate dedup engines but Commvault tells me this is normal. You can see this from the Commserve you can also see an offline storage pool in the library with no delineated volume information and an error message on the properties saying the setup has not completed. The Commvault Engineer has been loathe to recommend running a cleanup as he feels it would leave things behind they would have to walk me though manually. This case has been with Development for more t

So these deployments have been a real crap show won’t get into previous details but the current one hsxsetup wizard will not finish forming a storage pool. Main difference with this one is we used a known service account and password to authenticate with the Commserve instead of the code method in Command Center(account has Master permissions and more on the Commserve) Logs say the account is not an issue and according to Commvault the cluster is formed(3 nodes) and the nodes are registered in the Commserve as MA’s(they have duplicate dedup engines but Commvault tells me this is normal. You can see this from the Commserve you can also see an offline storage pool in the library with no delineated volume information and an error message on the properties saying the setup has not completed. The Commvault Engineer has been loathe to recommend running a cleanup as he feels it would leave things behind they would have to walk me though manually. This case has been with Development for mor

Hi All,We are in the process of rolling out HyperScale X 4 disk nodes for a customer as part of a global standard for them.  Hardware has been procured in line with the published Reference Architecture for the Dell EMC R650 nodes.  Checking documentation today I note that the Dell R650 has been removed from the 4-disk node models:https://documentation.commvault.com/v11/essential/132722_design_specifications_for_hyperscale_x_reference_architecture.htmland:https://documentation.commvault.com/v11/expert/132722_design_specifications_for_hyperscale_x_reference_architecture.html

our production and DR having hyperscale 1.5 , and we are refreshing production hyperscale 1.5 to hyperscale x and planning to add old production hyperscale 1.5 to DR to increase the capacity any best and easy way to do it ?  

Hi All, We are building a Hyperscale-X environment for a customer. However, I received a question from my customer.The reference architecture servers that are present are nearing the end of their life cycle. The processors on these servers belong to the 2nd generation. Is it possible for me to include this server in my HSX environment if the next-generation reference architecture is released and I need to add one node to my existing platform? Best Regards

wanted to add comments to this thread - we too ran into this issue … HSX image install failure , after MANY hours you get this in the install console --Inappropriate ioctl for device

i checked the links to configure Bonded Topology With 1 GbE Management Network for HyperScale X Reference Architecture but i can see the bond network type still selected as commserv registration ,  can someone explain if our data backup network is non routable network how is going to register it with commserv, or commserv registration will happen via management netwrok only? 

Hello,I have an issue with HyperScale x Installation - I’m installing HyperScale x with 3 nodesThe customer planned to connect all data protection and storage pool to 25g portsbut they doesn’t have all the equipmentCan I connect 2 nodes with 25g port and the last node connect with 10g portsand in month just change the cables and the SW configuration (after I will put the third node in maintenance)there is no documentation that address this situation - but it supposed to be OK, right? Is there any issue with doing that?     Best regards,

Hello, i have a project that a client will deploy HSX RA on Hpe proLient DL360 in the environment. i have never implemented HSX Before.and i have checked the documentation for  the RA installation and network configuration. https://documentation.commvault.com/v11/essential/154258_setting_up_installation_media_for_hyperscale_x_reference_architecture.htmlmy question is is there’s a guild explains step by step the installation starting from configuring the servers, and the HSX imaging ?my blind step is how the image will be installed?dhcp is not enabled.documentation says that u’ll use a windows machine that can connect to the nodes and download the iso from the store then mount the image and start the installation.

Just few questions ,we are having hyperscale x nodes , we have standard storage policy - primary copies - are the backup data in this are WORM by default ? can it be modified ? also what about immutability , do we need to perform Enable Retention Lock Workflow. to achieve immutability ??if we have storage policy encryption , do we need to do Software Encryption on HyperScale Storage

Hi, I have a customer who is undergoing a cloud transformation.They currently have a 6-node HyperScale X cluster with approximately 40 % used it with long-term retention. The goal is to migrate all long-term retention to Metallic and reduce the number of nodes from 6 to 3 in order to use the evacuated nodes and create another cluster in another site.Is this easily achievable?Thank you.

Good day!I would like to know how the Hyperscale X cluster is selected for the VMware backup VSA access node and how agent-based backup can point to the HSX cluster in the policy/plan.This is the first time, planning for an HSX deployment. Will there be a virtual name automatically created which can be selected in the Backup Plan/policy? 

Hello All,Environment:I have a requirement to backup vSphere workload. The Target Storage device is Hyperscale X. This is the first time using Hyperscle X. It is a three-node cluster (Reference Architecture). Commserve will be installed as external. Question:Do I need to install VSA on Hyperscale X so that it can act as a proxy access node for vSphere backup? The transport mode is NBD. ORDoes the HSX have an inbuilt VSA it will automatically act as a VSA proxy for backup transport as NBD? VSA agent installation is not required for HSX.

Good day Community,I have a customer running HyperScale X on Dell servers. They received a notification from Dell to update the firmware of the SSD Drive containing the DDB and Index Cache due to a security vulnerability.If we bring down Commvault, and Hedvig using the usual procedure as documented on BOL, are there any risks on doing such upgrade and is there anything else I should do prior to the upgrade? https://www.dell.com/support/kbdoc/en-pa/000134013/dell-poweredge-update-the-firmware-of-single-system-components-remotely-using-the-idrac Thanks very much in advance.

Hi When I am restarting one of Hyperscale X node as per below doc url but after restart node the all disks on that node showing offile till I select “refresh” option from action menu, please check attached screen shots, is there is wrong or it is normal behaviour    Stopping and Starting a HyperScale X Appliance Node (commvault.com)  Thanks

we have production 3 nodes and 3 nodes for DR , to aviod bandwidth issue we set up dr nodes in production site and created the cluster , done the first full copies , now we want to move it to the new site with different ip address  is it doable 

Hi, Just want to ask if there are specific use cases or features that Hyperscale X can provide without Commvault Backup and Recovery or Disaster Recovery software? Or would I need CV backup and recovery to enjoy the features of the Hyperscale X reference architecture (i.e. would I have backup and recovery capabilities using the hyperscale X reference architecture on my servers alone without the CV Backup and Recovery license) 

Hi! Just want to ask how the reference architecture for hyperscale X would be licensed. After acquiring a per node license, do we still need to purchase Backup and Recovery or Disaster Recovery licenses to enjoy backup and DR? Would appreciate any input on this. Thank you! 

2 related questions -If a drive dies on one of the nodes in my RA setup, and that capacity drive is no longer available but a larger one is, can I replace the drive with the larger capacity one and the system operate with perhaps using a fraction of the capacity?  e.g. a 8TB drive dies, could I replace it with a 12TB drive (& perhaps the system just use 8TB of that space)? I want to add an additional node, but the new nodes are only available with larger capacity drives.  Can I do that, and if Yes then will it use a subset of the capacity so the usable is the same as the existing nodes?Thanks.

Hello all!I would like to get insights on what should be our best deployment scenario.Consider a 3 node HSX RA where each node has 2 FC ports (2 physical ports).Consider a FC SANConsider a FC storage system (providing VMware datastores)Consider a FC Tape Library Scenario 1Each node connected to the FC SAN via the two physical ports and zoned via one port to the FC storage and the other port to the Tape LibrarySo, each node has one path to tape library and one path to storage. This way every node (which is an MA) can write backups to tape AND also run VSA SAN transport mode backups from ESXi hosts Scenario 2One or two nodes connected to the FC SAN and zoned via the two FC ports to FC StorageThe remainder node(s) connected to the FC SAN and zoned via the two FC ports to Tape LibrarySo, each node has two ports, either to Storage or Tape. This way I get path failover to storage and tape. Bu