Part 4 – Cyber Recovery in Practice with Cohesity NetBackup 11.X

Storage Lifeycle is a set of instructions or a plan what happens to the data once it’s written to disk or tape. We did have Vault or bpduplicate scripts in the past where stored backup data was replicated elsewhere. With advanced disks, SLP can copy a data within in a storage unit to another storage unit via a set schedule.

So, this somewhat looks like this,

  1. Write the backup (copy 1), for example to MSDP on nbu-media01.
  2. Make more copies automatically afterwards (copy 2, copy 3) to other storage, such as another MSDP pool, tape or cloud.
  3. Give each copy its own retention, so the disk copy expires quickly and the tape copy keeps for months.

NetBackup tracks every copy and keeps retrying until it’s done. You never schedule duplication jobs yourself.

What Optimised duplication is

It’s a copy between two deduplication pools that only sends the blocks the target doesn’t already have. A normal duplication rehydrates the backup back to full size and copies all of it. Optimised duplication compares fingerprints, so after the first copy a full backup might move only a few percent of its size. It’s the standard way to get a second copy to another site cheaply, within one NetBackup domain. A.I.R. is the same idea between two separate domains. This is the same concept of Pure disk product we had in v6.

A job gets triggered by master server ( nbjm ) and it opens the Backup/Restore manager ( bpbrm) to move data from client. Client process bpkar runs of client preparing data to be send to media server. De-dupe plugin compares what it holds in the de-dupe engine and separates files into small segments and throws away all duplicates.

Source – Cohesity.com

Types of Storage Life Cycle Policies

There are three types of SLP’s

  • Backups – Used as Backup Target ie: Copy 1
  • Duplication – Used for Backup Target ie : Copy 2 with extended retention
  • Replication – Used for Offsite Replication of a snapshot
  • Import – Used as part of Auto Image Replication. 

Creating Storage Life Cycle Policies

Backup operation in an SLP

Use the Backup operation in a storage lifecycle policy to create a backup.

Select Storage > Storage lifecycle policies > Give the SLP a name

Click Add provide the following information and Click Create

  • Operation – Backup
  • Destination Storage – stu-msdp-media01
  • Retention – 2 weeks
  • Retention Type – Fixed

Our Storage Life Cycle Policy for Backup is now ready

The SLP can now be attached to policies

Go to policy

Click Save and this now means our backups are being targeted to this SLP

Duplication operation in an SLP

This is used to create a copy of a backup

To create a duplication job for the backup policy go to the slp-win-silver SLP and click Add Child

Here we are setting a copy from (media01 → media02)

  • Backup to stu-msdp-media02, retention 2 weeks.
  • Child Duplication to stu-msdp-media02, retention 3 weeks.

Click Create

We can create as many duplicate copies as possible, so lets try the third copy to tape

  • Backup to stu-msdp-media02, retention 2 weeks.
  • Child Duplication to stu-msdp-media02, retention 3 weeks.
  • Child Duplication to stu-tape, retention 1 yr.

Configuring Media servers access to MSDP pool.

Each MSDP storage server has its own login, the MSDP username and password we set in the wizard. By default, only the media server that hosts it holds that login, so only that server can talk to its deduplication engine.

So, for this SLP to work now, we need the Media servers section on a storage server is the list of other media servers that get that login too so Adding nbu-media01 to nbu-media02’s storage server means “media01 may read from and write to media02’s dedup pool” and likewise the other way round too.Without it, the SLP has no media server that can see both pools and you will get 156 error in activity monitor.

Go to Storage → Disk storage → Storage servers tab and click nbu-media02.nbu.local to open its details.

Scroll to the Media servers section → Add nbu-media01.nbu.local and click Add, then Save.

Then do it the other way round: open nbu-media01.nbu.local → Media servers → Add nbu-media02.nbu.local.

This gives the other media server credentials for that storage server, so it can read from one pool and write to the other. Optimised duplication needs a media server that can see both.

Testing SLP jobs

Kick off the daily-win-clients and run the check via nbstlutil report

The activity monitor will show us the state of the SLP backup job

A quick hack to speed up SLP jobs is to change the timers for the lab.

Go to Host properties → nbu-primary → SLP parameters:

  • Job submission interval: 1 minute (default 5)
  • Force interval for small job: 5 minutes (default 30)
  • Extended image retry interval: 10 minutes (default 2 hours)

References

Maximum snapshot limit retention type for SLP operations — NetBackup 11.1 | Cohesity Documentation

(Visited 3 times, 1 visits today)

By Ash Thomas

Ash Thomas is a seasoned IT professional with extensive experience as a technical expert, complemented by a keen interest in blockchain technology.