# Query regarding incremental-lsn option in PXB

**URL:** https://forums.percona.com/t/query-regarding-incremental-lsn-option-in-pxb/30964
**Category:** Percona XtraBackup
**Tags:** percona
**Created:** [June 11, 2024, 8:43am UTC](https://forums.percona.com/t/query-regarding-incremental-lsn-option-in-pxb/30964 "2024-06-11T08:43:05Z")
**Posts on this page:** 5
**Page:** 1

<div class="post-metadata">

### Author: ![pravata\_dash](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/pravata_dash/32/12098_2.png) [@pravata\_dash](https://forums.percona.com/u/pravata_dash)
#### Post date: [June 11, 2024, 8:43am UTC](https://forums.percona.com/t/query-regarding-incremental-lsn-option-in-pxb/30964/1 "2024-06-11T08:43:05Z")

</div>

Hi Team,

We have a requirement for full and incremental backups. Currently, we are utilizing the --incremental-lsn option with the last to\_lsn value using the xbcloud binary to perform incremental backups. Below is a sample command:

xtrabackup --backup --stream=xbstream --extra-lsndir=/tmp/xtrabackup-lsndir --target-dir=/tmp/xtrabackup --compress --compress-threads=8 --stream=xbstream --parallel=8 --incremental-lsn=xxxxxxxxxx | xbcloud put --storage=google [options]

Now, I have a question:  
Suppose I have taken several incremental backups using incremental-lsn. If, while attempting the next incremental backup with the incremental-lsn value (to\_lsn of the last incremental backup), the backup node fails due to a hardware OR some other issue, resulting in data loss and necessitating a node rebuild.

In this scenario, would a subsequent incremental backup succeed with the same older value of incremental-lsn from the new node that we have rebuild or would it fail due to data inconsistency?

MySQL Version: 8.0.35  
Xtrabackup: 8.0.35

---

<div class="post-metadata">

### Author: ![pravata\_dash](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/pravata_dash/32/12098_2.png) [@pravata\_dash](https://forums.percona.com/u/pravata_dash)
#### Post date: [June 12, 2024, 9:50am UTC](https://forums.percona.com/t/query-regarding-incremental-lsn-option-in-pxb/30964/2 "2024-06-12T09:50:35Z")

</div>

Any update here team?

---

<div class="post-metadata">

### Author: ![matthewb](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/matthewb/32/34_2.png) [@matthewb](https://forums.percona.com/u/matthewb)
#### Post date: [June 12, 2024, 9:59pm UTC](https://forums.percona.com/t/query-regarding-incremental-lsn-option-in-pxb/30964/3 "2024-06-12T21:59:04Z")

</div>

Hello @pravata_dash ,

> [@pravata\_dash](#):
>
> Any update here team?

The forums are a voluntary effort from Percona employees. The forums are not a source of immediate assistance. If you need immediate assistance, you can [contact us](https://percona.com/support) for a support contract.

---

<div class="post-metadata">

### Author: ![matthewb](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/matthewb/32/34_2.png) [@matthewb](https://forums.percona.com/u/matthewb)
#### Post date: [June 12, 2024, 10:01pm UTC](https://forums.percona.com/t/query-regarding-incremental-lsn-option-in-pxb/30964/4 "2024-06-12T22:01:25Z")

</div>

> [@pravata\_dash](#):
>
> would a subsequent incremental backup succeed with the same older value of incremental-lsn from the new node that we have rebuild or would it fail due to data inconsistency?

There is no guarantee that LSN X on node1 is the same as LSN Y on node2. The LSN is (in general) a byte-offset of the REDO log. So, no, you would not be able to restart a failed backup on a new node using the LSN of the old node. LSN is not a GTID; it is InnoDB-specific.

---

<div class="post-metadata">

### Author: ![pravata\_dash](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/pravata_dash/32/12098_2.png) [@pravata\_dash](https://forums.percona.com/u/pravata_dash)
#### Post date: [June 13, 2024, 12:36pm UTC](https://forums.percona.com/t/query-regarding-incremental-lsn-option-in-pxb/30964/5 "2024-06-13T12:36:40Z")

</div>

Thanks a lot @matthewb for the details.
