# PBM - Mongo - Compatibility Matrix Questions

**URL:** <https://forums.percona.com/t/pbm-mongo-compatibility-matrix-questions/36080>\
**Category:** Percona Backup for MongoDB\
**Tags:** mongodb, pbm\
**Created:** [January 10, 2025, 6:42pm UTC](https://forums.percona.com/t/pbm-mongo-compatibility-matrix-questions/36080 "2025-01-10T18:42:02Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![space](https://avatars.discourse-cdn.com/v4/letter/s/8797f3/32.png) [@space](https://forums.percona.com/u/space)\
**Post date:** [January 10, 2025, 6:42pm UTC](https://forums.percona.com/t/pbm-mongo-compatibility-matrix-questions/36080/1 "2025-01-10T18:42:02Z")

</div>

Hey folks,

Using the Percona Distribution for MongoDB (7.x). I have PMM set up to make backups with PBM, I picked the latest version of PBM. This is a Mongo in a primary-secondary-arbiter replicaset installed directly on the host(s) via RPM’s.

Currently thinking about using Logical Backups on a Local Disk using PITR. However the compatibility matrix says this is NOT supported, only S3 is supported for this.

Do you have any other information about this compatibility matrix and why only local backups are not supported for Logical PITR?

Matrix:

> **[Percona Monitoring and Management - MongoDB Backup and Restore support matrix](https://docs.percona.com/percona-monitoring-and-management/get-started/backup/mongodb_limitations.html#restore-logical)**
>
> Creating and restoring MongoDB backups in PMM currently has the following limitations and requirements:

```auto
Restore: Logical
Full or PITR	Storage type (S3 or Local)	Support level
PITR S3 Full
PITR Local No
Full S3 Full
Full Local Full

```

PBM Version:

```auto
# pbm version
Version: 2.8.0
Platform: linux/amd64
GitCommit: 15b51a88dc76f4d61b56c478ced4d4e35d038744
GitBranch: release-2.8.0
BuildTime: 2024-12-12_13:40_UTC
GoVersion: go1.22.8

```

PBM config:

```auto
# pbm config --list
storage:
  type: filesystem
  filesystem:
    path: /mnt/test/mongodb_backups/rs0
pitr:
  enabled: true
  compression: s2
backup:
  oplogSpanMin: 0
  compression: s2
restore: {}

```

Very curious what the limitations are and where the compatibility matrix is sourcing information from, I notice a few cases of the documentation being out of date and hoping this is one of those things.

Appreciate your help, let me know if I can provide any additional information to help.

---

<div class="post-metadata">

**Author:** ![Ivan\_Groenewold](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/ivan_groenewold/32/6299_2.png) [@Ivan\_Groenewold](https://forums.percona.com/u/Ivan_Groenewold)\
**Post date:** [January 15, 2025, 11:53am UTC](https://forums.percona.com/t/pbm-mongo-compatibility-matrix-questions/36080/2 "2025-01-15T11:53:51Z")

</div>

Hi, using a shared storage like S3 or NFS is required due to the way Percona Backup is designed. Each agent needs access to the same backup location.

---

<div class="post-metadata">

**Author:** ![space](https://avatars.discourse-cdn.com/v4/letter/s/8797f3/32.png) [@space](https://forums.percona.com/u/space)\
**Post date:** [January 29, 2025, 8:13pm UTC](https://forums.percona.com/t/pbm-mongo-compatibility-matrix-questions/36080/3 "2025-01-29T20:13:59Z")

</div>

This makes sense, my “Local” store is actually an NFS mount, so data accessible between all mongos. Are you suggesting that this compatibility matrix would in fact imply ‘Full’ support under this condition?

My only concern was just that there might not be hooks to support a pseudo-local NFS recovery since it’s “assumed” the local drive is just a local drive, not an NFS mount.

I appreciate your time and thoughts on this.

---

<div class="post-metadata">

**Author:** ![Ivan\_Groenewold](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/ivan_groenewold/32/6299_2.png) [@Ivan\_Groenewold](https://forums.percona.com/u/Ivan_Groenewold)\
**Post date:** [January 30, 2025, 11:43am UTC](https://forums.percona.com/t/pbm-mongo-compatibility-matrix-questions/36080/4 "2025-01-30T11:43:19Z")

</div>

Perhaps the wording is not the best… but actually it is not possible to use independent local volumes. It always has to be a shared volume type like NFS, where every node “sees” the same files.