barman_xlog large size.

[Marcin Wolak]

Hi 

recently i got problem with barman probably. I don't know why but folder /var/postgres/9.5/main/baran_xlog is very large +100GB and it filled up whole space on postgres partition. it is now on 100% and i can't run database. 

Can Anyone help with this or maybe tell me if this folder content can be deleted?

thanks

[Rafał Karpiński]

Yes, its barman problem. It seems like you dont have enough space on your barman server.

Second question. No, You cant delete any single file in pg_xlog. When you delete wal file you will destroy your database system. 

--
--
You received this message because you are subscribed to the "Barman for PostgreSQL" group.
To post to this group, send email to pgba...@googlegroups.com
To unsubscribe from this group, send email to
pgbarman+u...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/pgbarman?hl=en?hl=en-GB

---
You received this message because you are subscribed to the Google Groups "Barman, Backup and Recovery Manager for PostgreSQL" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pgbarman+u...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/pgbarman/489bfc12-3ede-4638-9246-1bda9a1cff74n%40googlegroups.com.

[Luca Ferrari]

DON'T DELETE the folder content.

On Fri, Aug 19, 2022 at 8:33 AM Marcin Wolak <marcin...@gmail.com> wrote:
>
> Hi
> recently i got problem with barman probably. I don't know why but folder /var/postgres/9.5/main/baran_xlog is very large +100GB and it filled up whole space on postgres partition. it is now on 100% and i can't run database.

Thjis means postgres is no more able to archive wals, and probably
your barman setup is not working. You can dig the postgresql log to
see what error is coming up when archive_command is run. If you fix
it, the system will archive and deletes wals by its own.
If you are in a rush, and you think you can survive without having a
backup, set archive_command to '/bin/true' and it will delete wals by
its own, cleaning up disk space.

[Alnis Morics]

IMHO Barman always uses a replication slot, so there must be an active
Barman slot on your DB server with xlogs that Barman has not yet received.

If the DB is a virtual machine, maybe you can increase a bit the storage
volume where /var/postgres/9.5/main/baran_xlog is located, and then look
at Barman as to why it has stopped receiving the xlogs, maybe it's full,
too, or there's a network issue.

If you increase the DB storage just a bit, it will make it possible to
start Postgres, but you need to solve the Barman problem very soon, or
else the space on the DB server will be filled again. If you can afford
to loose backups, you can also drop the slot.

[Jeff Janes]

On Fri, Aug 19, 2022 at 2:33 AM Marcin Wolak <marcin...@gmail.com> wrote:

Hi 

recently i got problem with barman probably. I don't know why but folder /var/postgres/9.5/main/baran_xlog is very large +100GB and it filled up whole space on postgres partition. it is now on 100% and i can't run database.

Is this directory on your database server, or on your backup server?  Is it full of 16MB files?  How far back do the timestamps date?

Unless you have done something quite odd, I don't think either PostgreSQL or barman should be using a directory with that name.

Does the parent directory also have a pg_xlog folder?  Is one sym-linked to the other?  

Are you for some reason archiving to that folder?  (It is weird to use archive to a folder located inside the data directory).  What is your archive_command?

Cheers,

Jeff

[Michał Kępczyński]

Wojtek, list backup

barman list-backup xyz

and delete domem grom the list barman delete xyz 20170129T010005. In the future you have to use retention policy in the config file od cluster backup

pt., 19 sie 2022, 08:33 użytkownik Marcin Wolak <marcin...@gmail.com> napisał:

--

[Michał Kępczyński]

Sorry but my phone know better what I wants to write. I mean delete some from the list. 

[Luca Ferrari]

On Mon, Aug 22, 2022 at 3:05 AM Jeff Janes <jeff....@gmail.com> wrote:
> Is this directory on your database server, or on your backup server? Is it full of 16MB files? How far back do the timestamps date?

It is on the postgres partition, since he mentioned it is filling up
the database partition.
Either he mispelled the postgresql xlog directory, or he is doing a
backup on the same partition as the database is in.

> Are you for some reason archiving to that folder? (It is weird to use archive to a folder located inside the data directory). What is your archive_command?

Definetely there is the need for more info, like archive_command or,
as I suggested, the failures in the logs.
However, three days have passed, so I guess he survived in some way!

Luca

[Marcin Wolak]

I found problem. 

The problem was not in barman. But on main database. Suddenly he started to generate large WALL size.  Barman work was fine. After full vacuum wall size dropped to normal.

My setup is MAINdb --> BARMAN --> DEVdb1, DEVdb2

They sync everyday.