Hot answers tagged

7

You can always do what you want with an external utility. We had similar request from the customer years ago and had successfully solved the issue for Veeam who can’t limit backup file sizes itself. https://forums.veeam.com/veeam-agent-for-windows-f33/any-way-to-limit-vbk-vib-file-size-t32186.html Other backup software like say Acronis and CommVault aren’t ...


6

You don’t have differential cabling so HBA is switching into UltraWide mode. Get proper cable in between your HBA and SCSI device. https://ask.adaptec.com/app/answers/detail/a_id/11007/~/scsi-standards-%2F-transfer-rates-%2F-bus-width-%2F-cable-length (mixing LVD and SE is what’s limiting your performance)


5

RAID is not for backups-it's for availability. If you lose a disk in a RAID setup, your systems can still run. Without RAID, you would have to buy and prepare new disks, and restore your backup before your business can continue.


5

It’s the same storage you use as a backend for your running virtual machines, say NFSv3/v4, iSCSI and FC. Concepts behind are very similar. You might want to read the Docker/Kubernetes backup overview here: https://www.networkworld.com/article/3514560/how-to-back-up-kubernetes-and-docker.amp.html It won’t be acceptable by ServerFault to recommend any ...


5

rsync cannot have its bandwidth limit parameter changed dynamically. However you could Split the job into several smaller tasks where each will have a specific bw limit based on the time it starts, eg for instance rsync a b c d e target:/dir/ to rsync --bw-limit=bw1 a target:/dir/ ... rsync --bw-limit=bw2 e target:/dir/ Use the --time-limit=MINS option ...


4

Connect your cloudsql instance with proxy > https://cloud.google.com/sql/docs/postgres/sql-proxy then this commands should do the job pg_dump -U postgres -h localhost -p [PORT] [DBNAME] -n public --format=p >publicBackup.sql psql -U postgres -h localhost -p [PORT] [DBNAME] -c "drop schema public cascade;" psql -U postgres -h localhost -p [PORT] [...


4

Use tape for long-term archives and multiple public clouds (multiple tape-in-cloud still as no way anybody can store PBs of "ice cold" data on anything except tape). Hard disk isn't designed to store data for a long periods of time. 3-2-1 backup rule is in power either way. https://www.starwindsoftware.com/blog/3-2-1-backup-strategy-why-your-data-always-...


4

Dumping current state can be done by sending USR2 signal to keepalived parent process: kill -USR2 `cat /var/run/keepalived.pid` See result in /tmp/keepalived.stats. Note: if SELinux is used (CentOS 7), it does not allow writing to this file. You can get around this with this preparation: touch /tmp/keepalived.stats chmod go+w /tmp/keepalived.stats ...


4

You need a backup solution, which can use Linux server as backup repository. Or you can configure CIFS(SMB) on Linux and most of backup solutions can utilise SMB. For example VEEAM B&R can use Linux repositories - https://helpcenter.veeam.com/docs/backup/hyperv/linux_server.html?ver=95u4


4

Check Veeam Backup and Replication. In v9.5 they allow to append incomplete tapes. https://helpcenter.veeam.com/archive/backup/95/vsphere/add_gfs_media_pool_advanced.html The community edition is free for a use of 10 or fewer VMs. Haven't yet upgraded to their new v10.


4

Use Veeam agent-based VM backup. Before Veeam would re-publish their Nutanix AHV backup as a “generic” KVM VM backup.


3

Is downtime acceptable at all? Sneakernet is still a thing, and depending on the physical distance it may provide enormous bandwidth. For machines with limited allowed downtime it might still be an idea to try creating an initial seed to a locally attached drive using robocopy, then send the drive to the second data center and follow-up with a second ...


3

Robocopy is good for moving a lot of data. Command help here - https://ss64.com/nt/robocopy.html


3

I only have access to one Windows Server, but managed to rehearse the 'server won't boot but can only fully read user files from Server Backup via 'Windows Backup tool' on a server' conundrum by downloading an evaluation copy of Server 2019 Essentials from the Microsoft Evaluation Center and spinning up that 6-month trial copy in a Virtual Machine (in my ...


3

In some envs, you simply split the job into two teams - one team runs/guards the servers, the other runs/guards the backups. the other part is, imo even more, important: if the backup is writeable, especially from the system that is being backed up, it is NOT a good backup. systemic issue in many tools like borg. it is best if you can (in a cloud world) ...


3

You will most like have to create a new instance from the snapshot, validate the new instance is healthy and delete your second VM instance. The gcloud sdk has a flag for that - gcloud compute instances create with --source-snapshot=SNAPSHOT flag. I am not sure about your case, but it seems the newly release GCP machine images might be a good alternative ...


2

A small addition: You can actually use SFTP in case you are using ssh-keys to access the System. See the MikroTik wiki: https://wiki.mikrotik.com/wiki/Using_SSH_for_system_backup Though not well documented, the Mikrotik supports sftp using the same ssh key system Would be something like that (if you have a ssh-agent that handles the key): sftp myuser@...


2

You can use a script written by fracai on the Freenas forums. The script is hosted on github, it's called zfs-rollup. You only need the rollup.py file. Links at the end. I've used it for years and it works very well. You specify a few parameters on the comandline, 1) the prefix (filter) for the snapshot names it should look to delete, 2) the Dataset you ...


2

I did a quick calculation which puts your current infrastructure at approx US$940/month, but that doesn't include bandwidth. Let's say US$1000. I'll assume that your volume sizes are actual data. That's a fair bit of data, and that volume of data really makes a difference to your options as AWS bandwidth is so expensive. If you have less data the ...


2

You don't specify what Office 365 subscription/license this customer has, but if it includes Sharepoint Online, Microsoft has a free mihration tool that can migrate shared folders to Sharepoint Online Document Libraries. It needs to be manually run, but you can probably figure out how to automate it. https://docs.microsoft.com/en-us/sharepointmigration/...


2

When you are using dd you are copying data on the block level. The partition does not know anything about the filesystem or what is on it. If you copy a 10GB file on the server, and delete it right away, you will only remove the references to it in the filesystem. The data itself will still be there. So, when you copy the partition afterwards with dd, you ...


2

Sync programs like aws sync or rsync are easy to use, but metadata heavy. Large number of files causes a ridiculous number of metadata I/Os. Your tar command customizes the sync to what you care about. Only checking mtime metadata, or traversing trees with dates in the name, will be more efficient. But yes, custom means you must ensure it is correct. Block ...


2

Restore from backup media to a test system. Spot check by opening a few files, and check number of files and size is reasonable compared to the source volumes. Restore applications, and leave the now in sync with production system to users. It’s a restore system not a backup system. -- Raymond Blum, Google SRE


2

Here's how to do it: On Backblaze B2, create an application key that cannot delete files: b2 create-key --bucket MyBucket MyKeyName listBuckets,listFiles,readFiles,writeFiles Setup the backup so that it uses that key and does not attempt to delete old backup files. For example, in duplicity, do not use remove-older-than, remove-all-but-n-full, or remove-...


2

The only way to do this is using a 3 line shell script loop to explicitly change the mountpoint of all the file systems using zfs set mountpoint=...


2

I would start in the other end of the problem: Use Ansible or some other tool to programmatically define a working system, and then you have the ability to roll out new systems from a playbook that can be version controlled in something like gitlab or github. Linode has some guides to get you started in their environment: https://www.linode.com/docs/...


2

TLDR: Install a backup agent within the VM. Backing up the disk image from outside the VM certainly looks tempting, doesn't it? Especially if you have several VMs, then you can just tell the host machine to back up all the disk images and you're done, without needing to individually configure each VM to back itself up. The problem with that approach, ...


1

Are the network shares hosted on one of the VM's? If so, you're creating a difficult dependency if the host crashes and needs to be restored to bare metal. Purchase a suitably sized external USB drive and connect it to the host. Backup the host to this external USB drive.


1

Not the way you are doing it. SeBackupPrivilege is disabled by default and must be present and enabled in an access token presented to the remote system. I also don't believe this would work by specifying credentials for a local account on the target system, so if you want to do this remotely it would need to be a domain account. An example of how this ...


1

Unfortunately it looks like VS160 and DLT4000 formats exist on two different tape types, and it doesn't look like they are supported by the same tape drives. The DLT4000 drive goes with DLT IV tapes, whereas the DLT-VS160 drive goes with VS1 tapes. See https://en.wikipedia.org/wiki/Digital_Linear_Tape for more. If data destruction is your goal, you could ...


Only top voted, non community-wiki answers of a minimum length are eligible