After upgrade system has no more database space, reboot corrupts last days of data **solved**

[robr57]

I rebooted to do upgrade to 6.1 as I could not access SMB and noticed that this was fixed some week ago.

After reboot, everything is lost. No database anymore, and only errors in the log.

oled (22.12.2024 12:19:00): process started.
test5m (22.12.2024 13:20:01): weathernet daemon stalled, will be restarted now.
wnetd (22.12.2024 13:20:01): process started.
test5m (22.12.2024 13:20:01): logger stalled, will be restarted now.
wnetd (22.12.2024 13:20:01): send data thread started (pid:1697 tid:1699)
system (22.12.2024 13:20:01): stop of logging process initiated.
setup (22.12.2024 13:29:10): use UTC stamp 2024-12-22 12:29:10 (1 tries)
dbrepair (22.12.2024 13:29:19): database size is .
dbrepair (22.12.2024 13:29:19): test integrity of database.
dbrepair (22.12.2024 13:29:19): database integrity is fine, but rebuild is forced.
dbrepair (22.12.2024 13:29:19): not enough free space to repair database.
dbase (22.12.2024 13:29:21): process started.
dbase (22.12.2024 13:29:21): database sync done (0 tables, 0/0 writes).
system (22.12.2024 13:29:24): start of logging process initiated.
logger (22.12.2024 13:29:24): databases connected. Update interval 3600 seconds.
logger (22.12.2024 13:29:24): data logger started.
logger (22.12.2024 13:29:26): connect station 0 (Meteostick Vantage via USB serial).
logger (22.12.2024 13:29:26): connect station 1 (AirLink via SW Driver).
logger (22.12.2024 13:29:26): connect station 9 (System Data via Plug-in).
logger (22.12.2024 13:29:26): waiting for sensors (wind0, rain0, th0, thb0) to provide data.
logger (22.12.2024 13:29:26): services paused.
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), transmitter 1 registered as ISS
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), transmitter 2 registered as Soil
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), transmitter 3 registered as Wind
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), Meteostick Version 2.5.2
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), debug off
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), verbose off
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), output raw (8 bytes)
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), output raw (10 bytes)
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), filter on
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), rx threshold set to -90db
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), listening to transmitter 1 2 3
logger (22.12.2024 13:29:27): station 0 (Meteostick Vantage), bandwidth narrow
logger (22.12.2024 13:29:28): station 0 (Meteostick Vantage), chip authentication ok: 14
logger (22.12.2024 13:29:28): station 0 (Meteostick Vantage), frequency band 868MHz (EU)
dbase (22.12.2024 13:30:00): cannot create table (after 0 tries in 2ms): database or disk is full (13)
dbase (22.12.2024 13:30:00): cannot commit (after 0 tries in 1ms): cannot commit - no transaction is active (1)
dbase (22.12.2024 13:30:00): cannot rollback (after 0 tries in 1ms): cannot rollback - no transaction is active (1)
dbase (22.12.2024 13:30:00): cannot prepare select on update table 'wind1dir_hour' (after 0 tries in 1ms): no such table: wind1dir_hour
dbase (22.12.2024 13:30:00): rollback on insert forced (2) (hour): wind1dir
dbase (22.12.2024 13:30:01): cannot create table (after 0 tries in 1ms): database or disk is full (13)
dbase (22.12.2024 13:30:01): cannot commit (after 0 tries in 1ms): cannot commit - no transaction is active (1)
dbase (22.12.2024 13:30:01): cannot rollback (after 0 tries in 1ms): cannot rollback - no transaction is active (1)
dbase (22.12.2024 13:30:01): cannot prepare select on update table 'wind1dir_hour' (after 0 tries in 1ms): n

The last messages only repeat.

I tried to delete the database, do a rebuild etc. No luck.

In Luci I do not see any database space enabled.

What to do next ?

Mounted file systems
Filesystem Mount Point Available Used
rootfs / 4.93 MB / 9.13 MB 46% (4.20 MB)
/dev/root /rom 0.00 B / 5.50 MB 100% (5.50 MB)
tmpfs /tmp 18.48 MB / 25.00 MB 26% (6.52 MB)
/dev/mtdblock5 /overlay 4.93 MB / 9.13 MB 46% (4.20 MB)
overlayfs:/overlay / 4.93 MB / 9.13 MB 46% (4.20 MB)
tmpfs /dev 0.00 B / 512.00 KB 100% (512.00 KB)
tmpfs /tmp/mnt 0.00 B / 512.00 KB 100% (512.00 KB)
tmpfs /tmp/mnt 0.00 B / 512.00 KB 100% (512.00 KB)

Mount Points
Mount Points define at which point a memory device will be attached to the filesystem
Enabled Device Mount Point Filesystem Options Root Check
UUID: d9ba-6b1d /mnt/sda1 ? defaults no no
UUID: 68cda153-70d6-4870-bbcb-1bdd40dbc174 /mnt/sdb ? defaults no no

As you can see both are NOT Enabled.

[robr57]

To add, I do not know what kind of Filesystems I can select for /sda1 and /sdb . Can I just enable or do I have to specify a specific filesystem ?

[robr57]

UPDATE:
In Luci I have Enabled the 2 mount points /mnt/sda1 and /mnt/sdb , then I did a power cycle, and most seems to be recovered.

The Opslag now shows again : Opslag ATP: Internal 865 MB total, 424 MB free (51% gebruikt, 319 days)

I only lost 1 day of data it seems.

Still checking if everything works again.

[robr57]

Issue can be closed.
Still not clear what happend, but only lost data of approx 24 hours before upgrade. Strang thing is that external backup on USB stick also was corrupted. This backup should have been only 6 hours old.

[robr57]

New error after controlled Power Down and reboot.

Again 2 days of data lost !

Also the automatic daily backup has failed 2 days in a row. What is going on with this new release ?

I submitted a new issue in addition to this one
viewtopic.php?t=18495

As soon as the system looks to be running stable, it seems that after couple of hours either a memory problem occurs or the internal USB is dropped.

[admin]

Please don't fiddle with LUCI. If you do something there it can/will have side effects which can do anything and I will not support systems manipulated that way.

Can you please share log files?

Did you notice the "disk full" message? What did you do on this? What was the root cause for disk full?

[admin]

UPDATE:
In Luci I have Enabled the 2 mount points /mnt/sda1 and /mnt/sdb , then I did a power cycle, and most seems to be recovered.

The Opslag now shows again : Opslag ATP: Internal 865 MB total, 424 MB free (51% gebruikt, 319 days)

I only lost 1 day of data it seems.

Still checking if everything works again.

Setting mount points yourself will have severe side effects. Please do not do such stupid things. How can you believe mounting things here and there will fix anything, without understanding how Meteobridge makes use of the partitions?

I think you ruined the setup completely... going back to factory reset is the way to go. Otherwise you will get nowhere.

[robr57]

THe issue was a fault USB chip. It has been fixed as a service. Outstanding support from Boris.

Thanks you.

Issue can be closed.