Self-Hosted Q&A

When looking at the Fallen Behind Secondary Copies report, it shows 2 different numbers for “Total App Size to be Copied” and “Total Media Size to be Copied”. what is the difference between these 2? Also, what’s actually remaining to be copied? Is it the total of these 2 numbers?  

I’ve spent most of today trying to work out why aux copies to a new media agent keep failing connecting to the SIDB on the destination MA.It looks like it is Windows Firewall related as the moment I turn that off things work.On the new MA which is Windows 2022 it looks like Commvault has only put in 2 firewall rules very similar to the ones in this post.On the existing Windows MA’s I have lots of automatically added exceptions for tons of individual Commvault processes.My MAs and Commcell are all in the same VLAN so other than Windows Firewall on each host there is no firewalling between them.I’m struggling to find the documentation on what should be in Windows Firewall and I keep coming across “Network Topologies” but the documentation that isn’t exactly clear either about what they actually do in this situation and whether I s

Hi allI’m busy doing some MongoDB testing in our lab. I have a 3 node Mongo cluster. The backups work as required and I ran a restore succesfully.I have 2 issues during a restore process. The first is that the Mongo services won’t restart post restore. I’ll spend more time investigating that, as it’s easy to manually bring the instance up.When I try to run a point in time restore, I get the below error where the oplog won’t replay. From the looks of things, it looks like a permission issue. The account we’re using has the permissions below:   Error Code: [7:255]Description: Failed to replay the oplog dump [/opt/commvault/iDataAgent/jobResults/CV_JobResults/2/0/3940/local/oplog_rs0_1677651150.bson] on

I created a custom Package to install the commvault agent on a window machine. I’m planning on using ActiveBatch to run the job. When I run the custom package from the window machine manually and it install. However, when I attempt to us the ActiveBatch job it’s not installing. Below is the script used for the AB Job. $servers = "${server}"foreach($s in $servers.split(','))    {        $servername = $s + ".aptproduction.com"        New-Item "\\$servername\c$\temp\commvault\" -Type Directory         Copy-Item \\backup02s\mdt\Applications\Commvault\CommvaultAgentStl.exe \\$servername\c$\temp\commvault\        icm $servername {Start-Process -filepath c:\temp\commvault\commvaultagentstl.exe -argumentlist "/silent /install /silent" -Wait}        Remove-Item \\$servername\c$\temp\commvault\ -Recurse -Force            }  Below is the command we use to start the activeBatch job. neither of those work. 

Redhat 8 VM server on VMWare VCenter 7 with UEFI secure boot enabledHi everybody :) !I would like to do the enrollment as described here:https://documentation.commvault.com/fujitsu/v11/expert/118661_enrolling_commvault_keys_with_uefi_mok_machine_owned_key_list.htmlbut to do that I first need to do the import in the OS (RHEL 8 in this case):mokutil --import “commvaults secure boot key”I searched for the commvault keys but couldn’t find them.Any hint , download location, would be great!Thank you in advance whoever knows how to help me here, brgdsstefan

We have our Commserve and Media agents running Windows 2019 Server Standard Edition.We have MySql 2014 and would like to go to MySql 2019We would like to upgrade commvault 11.26.27 to 2022E(LTS) versionWhat are our best options? ThanksLarry

During the installation of Commvault CommServe is it possible to:A: Use an instance of SQL NOT installed on the same server as CommServe?B: Assuming YES to above, use a clustered Always On Deployment of SQL?C: Is CommServe supported on a shared SQL instance and not an instance dedicated to CommServe?D: Do any of the above have implications on the backup of the system database?Many thanks for responses!  

Hi,Need to know from your experience if Commvault license has been expired are client backup will not run or will failed when run? Our license is about to expire soon. We cannot ensure if our renewal will close out before the expiration. thank you.

Hi,Please help me with the following questions:Let’s say I deploy Hyperscale using the reference architecture on Hitachi hardware. I need to set long term retention backups to anything but tape.Can I connect an external storage as secondary disk/cloud library to the environment? Also, what type would it be (FC luns, iSCSI luns, NAS shares/exports (NFS, CIFS), S3 buckets)? If any external storage is permitted as disk/cloud library, would it be connected to all nodes on the Hyperscale cluster, it being the one who controls writes between nodes avoiding conflicts? It being an appliance, does it have any operating system hardening besides the security features included in the software (which run with or without the appliance)? For example, intrusion prevention and detection (like avoiding root logins, avoiding running OS commands or tampering with autostart config, etc.) When upgrading the software, does it work just the same as a non

Hi,We have some physical servers which use the File Level Backups. I wonder if I just want to restore some certain files in those physical servers from tapes; what is the best practice for it? Thank you.

HiI need  a people team can done a sql bak restore to a disk but these people don’t have the “browse and restore option” dialog box when the launch the restore task which rights are necessary to have it  ?

Hello Community  We are looking to  restoring an older version of SQL DB to a more recent version of SQL ... IS that possible :  Source SQL 2000 WIndows 2003 / destination SQL 2003 - WIndows 2008  source db currnetly backup via Commvault v11.24.69

Hi Everyone, I’m trying to build a backup map via commserve rest-api. Does anyone what the relational data structure of the commserve of the client groups, clients, apps, storage policy, schedules and subclients is like? I want to create a report with all subclients configuration. Thanks,Regards,Juan Forero 

Hello Here is a snip: I cant seem to download maintenance releases or even copy maintenance releases.  We are on v11.20.128 (I know behind on releases)Please adviseI see the following under the “Download Software” log- IsRootFilesPatchingNeeded() - Updates folder [C:\Program Files\Commvault\ContentStore\SW\CVMedia\11.0.0\SP20_2779330_R798\Windows\BinaryPayload\LooseUpdates\CU142\Updates] does not exist, try patching Maintenance Release files only7372  1c88  03/08 21:28:51 9278 CreateTempFolder() - Created folder at [C:\ProgramData\Commvault Systems\Galaxy\Temp\DownloadPatching]7372  1c88  03/08 21:28:51 9278 PatchRootFiles() - Source root file [C:\Program Files\Commvault\ContentStore\SW\CVMedia\11.0.0\SP20_2779330_R798\Windows\BinaryPayload\LooseUpdates\CU142\BinaryPayload\Setup.exe.zi

I have a client that is kicking off DB2 backups via cron job, there is an issue with the backup and I killed it but it keeps starting again.  is this normal?What am I missing? 

Our current deployment has 4 physical MediaAgents on prem to backup to a flash array, we have an Aux Copy from the flash array to Azure which has 4 MediaAgents in Azure.  The current Aux Copy job shows Source MediaAgent as the 4 OnPremMAs and the destination as the 4 AzureMAs.  We have recently deployed an on prem S3 with an additional 4 physical MediaAgents and have added it as a Synchronous Copy similar to the Azure Aux Copy.  The on prem S3 Aux Copy job shows one physical MediaAgent as the source and destination MediaAgent.  One of the “new” MediaAgents was actually part of the line for the OnPremMAs at one point but was replace by newer hardware.  This one MediaAgent is the one that is showing as the Source and Destination MediaAgent.  This OldMA shows up on the Mount Paths for the flash array with Read/Write Access.The S3 has been configured with 4 Mount Paths, one for each new physical MediaAgents and each has been shared with each of the new

Good day all!  sharing this proactively in case you encounter this issue.  Any questions, post below!IssueA synthetic full started after incremental may get stuck in job controller and cannot be killed.Evidence of the issue can be observed on the Commserve in Jobmanager.log.CommServe\JobManager.logServant [---- SCHEDULED SYNTHETIC FULL REQUEST ----], taskid [x] Clnt[Name] AppType[x] BkpSet[Name] SubClnt[default] BkpLevel[Synthetic Full] (old job [x])JobSvr Obj Skipping reference time check for SynthFull job.JobSvr Obj Skipping reference time check for SynthFull job.JobSvr Obj Another backup [x] started running before this synthetic full could start to run, failing this synth full job.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\####### createSynthFullBackupFromIncr doJobInitializationChecks failed

I’ve been using Commvault through the java console for years and backups are what I guess you’d call traditional GFS so even though we’re using global dedupe and synthetic fulls we’re basically running a weekly synthetic full and aux copying all backups from disk to tape each week.The change rate is really low so most weeks we’re basically coping the exact same multi-TB backup to tape as the week before.I’m wondering what people see as the pros and cons of things like server plans where it looks like it might push out the synthetic full to once every 30 days.Or maybe even stick with “manual” schedule policies but look at pushing synthetic fulls out to run a little less often.Right now I’m just thinking about this because I’m building up a new MA so it’s made me read/explore a little more - basically nothing is broken so I could just leave it as it is but I’m quite conscious that if I span up a new commcell and use the web console we probably wouldn’t end

Summary The following thread describes the potential exposure to the Apache Log4j vulnerability and steps to update Commvault software.It has been confirmed that a small subset of Commvault agents are impacted.  Update as of 1st February: Maintenance Release to bring Log4j version to 2.17.1 across Commvault software platform has been released. This release includes the upgrade of components that previously used Log4j 1.x.  Update as of 20th December: Maintenance Release including relevant hotfixes now available for Commvault software, see section “Maintenance Releases”. Please note customers who have already applied hotfixes for Log4j 2.16, do not need to install.For customers with Commvault Hyperscale X and Distributed storage, please see section new Community article here: 

Dear Members,What will be the best practice to create DB policies and schedules for MS SQL Backups. As per our standard policy Primary Copy 7 days retention period, AUX copy to DR 8 weeks and then tape backups for 1 year and 10 years. So how should we create schedules for Daily DB backups, Transaction logs backups every 4 hours. Should we keep separate storage policies for DB and logs backups or they can be on the same.Any addiditional information will be highly appreciated. Regards,AQ

Hi All,We are in the process of decommissioning the on-prem media agent server. After decommission we won’t have to backup any clients and the new server which we are going to build is for restores. The new server needs to be build in Azure.Currently we have on prem backups to disk library for 30 days retention and AUX Copy backup which runs every 60 mins to AWS Deep Archive. In order to prepare the new server for restores till date, I have done the following, but got stuck with some questions, Please assist me in completing this task.Built a new Windows Server in Azure Installed CommVault Media Agent Server with the same build as the on-prem server 11.20.85Ques : Should we move DDB of AWS to new Media Agent Server?https://documentation.commvault.com/2022e/expert/11078_move_ddb

Hi, The customer wanted to release vm socket license for esxi host. In normal condition to do so you have to make full backup of VM on other esxi host, right?What if VM and subclient does not exist anymore and retention for last backup is infinity? He can’t delete the backup. The customer was using VM backup report to check last backups on esxi. The customer had to make backup of all VM’s in environment (all subclients) to solve the case. Is that normal? Regards,Lukasz

Hello,After changing the DR Backup UNC path on CommServ to a new network shared folder, the new DR backup jobs complete without issues.Anyway on the Event Viewer there are a lot of warnings with Event code 34:59, relative to the old folders to be deleted, that cannot be accessed anymore.The messages are like: CommServeDR: Path [<OLD_PATH>\SET_NNNN] cannot be accessed. It may not exist. Otherwise, please check Share and Security settings.I believe that it’s normal because I’ve also changed the user to access the new share.These warnings are repeated every day, after the successful DR backup job.Is there a way to avoid CommServ to try to delete all the old directories forever? Best RegardsLucio

Hello Everyone,I checked the documentation but was not able to find anything related to my query.Does someone know if we can protect Microsoft PKI - Public Key Infrastructure ? I know we can cover the filesystem backups of this servers, but the point is to truncate the logs of the application somehow.Thanks. 

Latest v11.28.44.I’m backing up some Windows File iDA agents and some UNC locations to a disk library on a server that is to be retired.The old server had 12 individual volumes each presented as a mount path.We’re using Global DDB and there is about 50TB of physical storage consumed on the old library.The new server is running Server 2022 and ideally I want the disk library under a nice simple “Backups” folder on a single ReFS 64KB cluster size volume on the new MA.Whilst the volumes on the new MA are BitLocker encrypted we’d also like to use Commvault encryption.So what I think I can do is either move the existing mount points to the new hardware but I end up with 12 mount paths under the “Backups” folder or setup a new library with a single mount path and aux copy the data to it and then switch it to primary.Two questions.Which of these is the better option if I want to enable encryption on all the backup storage and wha