requested WAL segment 0000000100006FF4000000C6 has already been removed

[VladislavePolyakov]

description:

it was created by crontab once per week. backup made fro replica.
/usr/bin/pg_probackup-12 backup -B /mnt/BACKUP-postgres-ssh/ -b FULL --instance=nestle --stream

question:

how to guaranty all WAL will accepted after backup?

log from pg_probackup:

2021-01-30 05:27:25 MSK [61564]: ERROR: Problem in receivexlog
2021-01-30 05:27:25 MSK [61564]: ERROR: Interrupted during page reading
2021-01-30 05:27:25 MSK [61564]: ERROR: Data files transferring failed, time elapsed: 1d:5h
2021-01-30 05:27:25 MSK [61564]: WARNING: backup in progress, stop backup

log from PostgreSQL:

2021-01-30 02:27:25.249 UTC,"backup","",61728,"[local]",601325e0.f120,1,"streaming 6FF4/C6060000",2021-01-28 21:00:16 UTC,5/0,0,ERROR,XX000,"requested WAL segment 0000000100006FF4000000C6 has already been removed",,,,,,,,,"pg_probackup-12"

show from pg_probackup:

================================================================================================================================================
 Instance  Version  ID      Recovery Time           Mode   WAL Mode  TLI     Time    Data     WAL  Zratio  Start LSN      Stop LSN       Status
================================================================================================================================================
 nestle    12       QNNX01  ----                    FULL   STREAM    1/0    1d:5h  1756GB       0    2.91  6F1E/59443290  0/0            ERROR

versions :

pg_probackup-12 2.4.8 (PostgreSQL 12.5)
NAME="Ubuntu"
VERSION="18.04.5 LTS (Bionic Beaver)"

[gsmolk]

Hello!
You can use --temp-slot flag.

[VladislavePolyakov]

thanks!
will try to use it
if all ok, on next week will close issue

[gsmolk]

Glad to help.