Server 2.4.12 to 13 update broke something :(

Server
1

Hi folks.

Been running 2.4.12 in docker (uroni/urbackup-server:latest) in OpenMediaVault 5 (debian something under the hood). I use Portainer to start and stop a few dockers & my backups have been running well for months. Last week I tried to update the urbackup server by stopping the container, pulling an image, and starting the container. After the restart, the web page for the server wouldn’t load completely, then came up with the red box saying the database was being upgraded. This went on for a day and never completed. I checked the log and found lots of this repeated:

2020-10-17 16:11:27: WARNING: SQLite: os_unix.c:36830: (19) mmap(/var/urbackup/backup_server_link_journal.db-shm) -  errorcode: 5386
2020-10-17 16:11:27: WARNING: SQLite: disk I/O error in "PRAGMA synchronous=NORMAL" errorcode: 5386
2020-10-17 16:11:27: ERROR: Error preparing Query [PRAGMA synchronous=NORMAL]: disk I/O error
2020-10-17 16:11:27: WARNING: SQLite: os_unix.c:36830: (19) mmap(/var/urbackup/backup_server_link_journal.db-shm) -  errorcode: 5386
2020-10-17 16:11:27: WARNING: SQLite: disk I/O error in "PRAGMA cache_size = -2048" errorcode: 5386
2020-10-17 16:11:27: ERROR: Error preparing Query [PRAGMA cache_size = -2048]: disk I/O error. Retrying in 1s...

I’ve been bumbling around searching here, but not really finding anything that is the same. Anyone have any suggestions on what I could try next? More info I could share?

2

urbackup.zip (2.8 MB)

I managed to get the log file from the container. I think the 14th was when I did the update to the server… :frowning:

3

I think it is mmap error 19 ENODEV which probably means that the -shm file is on file system which does not support mmaped files. Move it to a file system that supports this and it should be fixed…

See e.g. threads like this https://forum.openmediavault.org/index.php?thread/32959-mergerfs-sqlite3-plex-homessistant-disk-i-o-error/

4

Nice! The solution at the link did the trick, the server started without the errors this time. Thank you.

I wonder why it happened? I hadn’t purposely changed anything in months. There were a few updates that occured, but it wasn’t until this update from 2.4.12 to 13 that it happened.

Wondering, watching a bit, I noticed the server identified my clients as ‘new backupclient: computername’. I wonder if it’s starting from scratch leaving some abandoned backup files from past lives…