Storage and Deduplication

Hello, hope this finds you all well! A few days ago I configured backups for an Oracle RAC client and I’ve been noticing a very strange behavior… Every incremental backup gets switched to a full backup and there always this error being displayed: “UpdateIndex failed on MediaAgent [media agent name] - Index files are missing. Job could have earlier run on a different MediaAgent possibly due to a failover.Source: media agent name, Process: UpdateIndex”.It is true that we have a pool of 4 media agents and it can use any of them but we don’t notice this error on any other of our hundreds of backups and they also jump media agents between jobs. A lot of them being other Oracle RACs…What is even stranger is that this error happened even during its first full backup. Could this be happening because this DB is a duplicate of one configured with other Storage Policy that uses other media agents? Even though they use different DDBs? Can it be

We are frequently encountering a problem where data chunks are missing, and this is impacting our ability to restore multiple VMs. Initially, we suspected an issue with the cloud storage provider and requested the support team to investigate. However, upon further analysis, we discovered that a delete command was issued by Commvault, which led to the deletion of the data from cloud storage.As a result, multiple storage policies have been affected, and we are currently unable to restore the data for several virtual machines. Analysis: According to DataCore logs:The SFILE delete request was received at 08:08:49 The Compact File delete request was received at 08:07:55Commvault is expected to send a delete request for the Compact File only after a successful remote copy of the Compact File to recreate

Based on our current Commvault infrastructure: We have one CommServe and a Media Agent at the DC location. A Standby CommServe has recently been configured in Azure, and we successfully tested a planned failover. During the test, we were able to run backup and restore jobs using the Standby host while the Production CommServe (services) was down. Now, we would like to extend our testing to include the Media Agent as well, specifically for a scenario where both the Production CommServe and the Media Agent at DC are unavailable.From the planned failover test, we have confirmed that the Standby CommServe can take over. However, in the event of a complete DC outage (CommServe and Media Agent down together), we need clarity on how to proceed.Our current policy setup is: Primary copies → Local storage (without DDB). Secondary copies → Cloud storage (with Commv

Hello, TeamWe have Unity Storage With Multiple LUNs, acting as mount paths attached to 6 Media agents with about 2.6TB datasize in total. There are multiple disks failure on the Unity storage end and we need to move the active data into another Newly purchased storage. The plan is to have all mount pats or data written to that storage with their  dependencies.The plan is to use commvault for this migration and here are my queries below:Should we use the Move Mount Path option? if yes, can restore happen while the mount paths are being moved? Each LUN is about 10-20TB each and we have multiple of them. Should we use the Storage policy create new copy or auxillary copy to move the data and promote the destination storage to primary.Also, considering data size, how could it take the data to complete migration?When the migration is completed. We would love to see that the data has left the Unity storage. The source storage and desti

Hi,according to BOL Deduplication Building Block Guide,a Media Agent can host up to two (2) DDBs (from on LUNs of different RaidGroup) ALL DDBs of all Storage Pools on a Media Agent should share the same volume / paths (e.g. D:\DDB)My question is now, does this require to create a streched / expanded Volume accross multiple (2) LUNs / Disks ?otherwise I don’t get, how one single directory or volume can be used to place the data to mutiple LUNsand should this be a striped or a concatinated volume layout ?In the past I used D:\DDB and E:\DDB as paths (different volumes on different LUNs).This doesn’t seem to be best practice anymore, if I understand the BOL entry correct.I could understand, if this restriction is valid for all (3 horizontal scaling) DDBs bel

Hi,I have a question regarding move mount path. The one and only possible MA is Linux.Source is a nfs share which supports sparse files. Target is a CIFS share which does NOT support sparse files.“move mount path” is a tool in Commvault.What happens with the files if using move mount path?wikipedia wrote:Sparse files can cause problems when copying: When copying to another file system that does not support sparse files, the target file cannot be created in compact form. Therefore, the undefined areas of the source file are written as null characters in the target file. This results in a significantly larger target file. The target file system must have sufficient capacity. In addition, both the copy or backup program and the target operating system must support sparse files and be able to recognize them as such.How does Commvault handle these files between the different file locations? SW is not too old 11.

Hello Commvault Community!In one environment, we noticed that strange entries appeared in the logs regarding block pruning as part of deduplication: SIDBPrune.log - from August 5, 202511512 5468  08/05 03:00:45 ### 1-3 OpenIdx:3330 Cannot lock chunk [2499152].11512 5468  08/05 03:00:45 ### 1-3 PruneChunk:1391 Cannot get child information. Folder [E:\LOCAL_STG\Folder_06.17.2016_11.59\CV_MAGNETIC\V_393205\CHUNK_2499152], Error [0x80070003:{CQiDirectory::Open(106)/SearchPath=[E:\LOCAL_ABC\Folder_06.17.2016_11.59\CV_MAGNETIC\V_393205\CHUNK_2499152\*]} + {FXFile::CVFindFirstFile(1821)/W32.3.(The system cannot find the path specified. (ERROR_PATH_NOT_FOUND.3))-FindFirstFileW failed, Path=[\\?\E:\LOCAL_ABC\Folder_06.17.2016_11.59\CV_MAGNETIC\V_393205\CHUNK_2499152\*]}]11512 4f14  08/05 03:00:45 ### 1-3 PruneChunk:1391 Cannot get child information. Folder [E:\LOCAL_ABC\Fo

Hi guys,We are currently on the proccess of configuring WORM S3 Object Locking + Commvault 11.36 + Deduplication.The requirements is to enable WORM on the following JOBS:4 weekly Fulls24 monthy Fulls5 Yearly FullsThe thing is that we are not able (we dont know if we are doing something wrong) on applying Extended Retentions on Deduplicated S3 Object Lock Storage pools. (If we disable deduplication is posible to configure Extended Retentions).That means we have to have 3 different buckets 1 for each retention.If we sum the periodically sealing of DDB (yearly or less..) that means HUGE space needed with this configuration. Do you know if it’s possible to apply extended retention on deduplicated S3 object Lock configured Storage Pools?Thanks!!

Hello, I need to verify if commvault support auxiliary copy from cloud storages (HCP) to Tapes ? Couldn’t find any official document regarding this. 

I need your advice.We have a plan to close a Data Center. All the backup infrastructure will be decommissioned but We have around 10K tapes written thru the years and we need to guarantee data can be recovered from tape if needed.What are the options we have?Is better to import the whole Tape library into an another CommCell server?Is better to import only the needed tape?If all backup infrastructure is decommissioned and we receive a restore request how we can identify the needed tape Id to restore the data from.Does Commvault provides this recovery service?Please help me to know what the best option is 

Hello team,I am trying to browse and restore the data from snapshot backup and from E drive of the client.I am getting an error “Failed to process mounted VM’s dynamic disk”, i can browse and view the items from other drives. Upon checking we found E drive is configured as Dynamic. Kindly confirm, is Dynamic disk configurations are not supported snapshot backup/restore?Is there any other ways to restore the data in such scenarios.  

I wanted to verify what happens to additional copies’ backups when a Plan is deleted? I would assume that it would be similar to a Storage Policy in that when a Storage Policy is deleted, all backup jobs under all the copies are aged and the Storage Policy and all copies are deleted.I have doubt due to a customer’s environment where a Plan no longer exists (I would assume it was deleted) but the corresponding Storage Policy still exists in the Java Console, albeit with Primary Copy being one of the original Secondary Copies to the cloud, and the original Primary Copy to disk deleted. All the jobs within the remaining copies have aged at this point; the Storage Policy itself still exists, but the Plan is gone. If all copies are deleted when the Plan is deleted, I’m unsure how this is possible.They’re trying to retire a MediaAgent from their environment and have created a new Plan to use with the new MediaAgent as they have a 1:1 relationship with their business unit’s d

Hi,We have a deduplication Database that it doesn’t seem to deduplicate at all and is ocupying more than we guess it should be.that DDB only backups VMs ans the deduplication ratings showed in the DDB are horrible and it caused that the library is full.   Best regards.

A storage pool was sealed about a month ago. All jobs for that DDB appear to be gone, and the old folders are no longer seen under the physical paths for each partition, yet the backing storage has yet to be freed up to pre-sealed levels. Meaning - when do those old chunks get pruned from the libraries?Space reclamation fails because the DDB paths aren’t around anymore obviously. There appears to be over 150TB locked up in the S3 library used by the DDB.thanks

We are implementing a new StoreOnce system as primary copy and test performance (and other topics) between FC-catalyst stores and FC-Virtual Tape Libraries A Catalyst Store on StoreOnce can be configured either as high-bandwidth or low-bandwidthhigh-bandwidth: Backup server (media agent in case of CommVault) will send backupstreams unmodified to StoreOnce and then StoreOnce will do dedupe etc. before finally storing the data on its disks. low-bandwidth: already the backup server (media agent) will do de-dupe and only sends new things to the StoreOnce.VTL is similar to high bandwidth (all goes from MA over FC-SAN to StoreOnce and only there the data reduction takes place). As we have plenty of ressources on FC-SAN and the StoreOnce is made for doing all the datareduction we’d like to get StoreOnce connected/configured in high-bandwidth for Catalyst, especially as important servers are doing LAN-free backups (have their own

Hi All, We’ve recently created a new Library, storage policies and new Media Agent servers. The Media agents have separate NIC’s for management, backup and replication traffic. My question is how do i know or where do i configure that Aux copy process will use the replication interfaces on the media agants.I would really appreciate any advice.ThanksTerry

Dears, Initially, the auxiliary copy was running with good throughput speeds of 4000 to 5000 GB/hr. However, due to an issue with the tape drives, we recently replaced them. We deconfigured the old tape library and reconfigured the new one.Since the new configuration, the auxiliary copy throughput has significantly dropped.Note:We have six LTO-6 tape drives with LTO-6 cartridges. A secondary copy has been configured, and data paths are enabled for all six drives with the associated Media Agent.Kindly advise on how to mitigate this issue. Regards,Mani D 

Hi everyone,We're planning to reduce our retention policy from 90 to 60 days on a Commvault library and would like to estimate the potential space savings this change would provide.We’re particularly interested in two scenarios: Adjusting the retention period on the entire library Adjusting it only for specific subclients Has anyone done this before or know the best way to extract this information (e.g., using built-in reports, custom queries, or logs)? Any help, scripts, or suggestions would be appreciated.I have seen some inconclusive posts in the community page, so I thought I’d ask.We are on v11.32, and do not have the “Data Retention Report” which could provably help. Also was not able to come up with a way to get this information from the “Data Retention Forecast and Compliance Report.”I guess a way would be adding up the first/oldest 30 days with of backups, but not sure whether this accounts

Hi,I’ve been trying to restore DR backup from DataDomain appliance.First, I’ve created a dedicated StorageUnit for DR Backups and added it in Commvault using DDBoost Client then I’ve put some DR backup jobs on it next I’ve retrieved whole mount path from DataDomain to a local disk on a destination computer added bMERestoreAsFS on destination computer Downloaded Media Explorer and cataloged mount path locationIn Media Explorer I can see all my DR backups just fine, but when I choose to restore, then the restore fails.

Hello everyone, I'm here to ask for your help. I have a QNAP with about 400 shares (the NAS is used at corporate level) so I necessarily have to have a daily incremental backup. Since Commvault has not developed an agent for QNAP, I am forced to use a Linux or Windows VM as a bridge to perform backups. I have already consulted the official documentation but I have not had any success. Is it possible to get some suggestions from you?Thank you all!

Horizontal scaling QueryIf a DDB is marked full because a certain threshold is met, a new DDB is scaled out.The newly scaled out DDB will be use only by new subclients.  So am I right in the assumption that the DDB which was marked full will still be use by the existing subclients and the existing subclients will potentially continue to add more primary records in the DDB.  Thereby potentially increasing QI time on the DDB as more primary blocks are added? 

Hey,we are using Intellisnap Snapshots with NetAppQuestion is, is there a possibility to modify the snapshot Name with a prefix or suffix like “daily” “weekly” or something else? I only found the general Naming convention:https://documentation.commvault.com/2024e/essential/naming_conventions_for_snapshots_created_by_netapp_array.htmlNo other options. Thx & BRMarcel

If you enable horizontal scaling for a DDB.  This is straight-forward for primary copy.  Once it reaches a certain threshold it will scale to a new DDBand new subclients will use the new DDB.  Now for aux copy/secondary synchronous copy ( we dont have subclients to associate here ), how does it work? What happens to the previous DDB and to the newly scaled DDB? Does aux copy continue on the new DDB automatically and leave the old DDB for copied jobs already? Anyone who can explain in simplistic term?

According to the sizing grids:https://documentation.commvault.com/11.20/hardware_specifications_for_deduplication_mode.html  I might need two media agents per site for my storage load.  This provides  for up to 1000 TB storage and, assuming just the 1st column - 1 DDB location per node.  So two DDB locations. From what I’ve read: Using Windows NTFS (simply, no other M$ features at the moment), we cannot share the mount paths (SAN...) as NTFS isn’t built that way.  Thus, we can’t load balance be

Dear Team, We have recently built a Commvault infrastructure with the following configuration: CommServe: Virtual machine running Windows Server 2025 Media Agent: Physical server with locally attached disks, running Oracle Linux 8.8 Mount Paths Configuration for DDB and Disk Library:  /dev/mapper/ol-opt_commvault_dbb xfs 600G 22G 579G 4% /opt/commvault/dbb /dev/mapper/ol-data_commvault_disklib xfs 11T 4.9T 6.2T 44% /data/commvault/disklib  We would like to know how to enable immutability (WORM) on the disk library, mount path, or at the storage policy level. Additionally, we have enabled ransomware protection, but it still shows as disabled in the Media Agent properties. Below is the current SELinux status of the Media Agent server:[