folder.id=aaaaa-bbbbb folder.type=receiveonly log.pkg=model)
```
…And here’s what I’m seeing on the Qnap for the overall storage pool and the shared folder:
When this happened a bit earlier, restarting the NAS appeared to resolve it - for a short while. Now it’s not making any difference.
Any ideas how to troubleshoot this further?
Thanks!
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [August 27, 2025, 6:40pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/2 "2025-08-27T18:40:55Z")
> [@Moisie](#):
>
> `database or disk is full`
Apparently this is a SQLite error which seems as likely to mean the database is corrupt.
---
### Author:  [@Moisie](https://forum.syncthing.net/u/Moisie)
#### Post date: [August 28, 2025, 7:53am UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/3 "2025-08-28T07:53:09Z")
Ahhh… thankyou!! I will delete the DB and try again!
I was convinced it was some hidden quota setting in the Qnap’s ZFS settings which I couldn’t find - but thanks for steering me yet again!
---
### Author:  [@USRFSledge](https://forum.syncthing.net/u/USRFSledge)
#### Post date: [August 28, 2025, 8:18am UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/4 "2025-08-28T08:18:07Z")
Worrying though, taken that it is a fresh set up doing a first index/sync. Why would the DB get corrupt? Even twice? That seems quite uncommon to me.
My first thought on that same log entry was, that the DB is incorrectly stored on a non-data-share and actually running out of disk space.
---
### Author:  [@Moisie](https://forum.syncthing.net/u/Moisie)
#### Post date: [August 28, 2025, 1:51pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/5 "2025-08-28T13:51:34Z")
> [@USRFSledge](#):
>
> Worrying though, taken that it is a fresh set up doing a first index/sync. Why would the DB get corrupt? Even twice? That seems quite uncommon to me.
Indeed - curious…
> [@USRFSledge](#):
>
> My first thought on that same log entry was, that the DB is incorrectly stored on a non-data-share and actually running out of disk space.
On this system, the syncthing config folder is stored in `/share/ZFS530_DATA/.qpkg/syncthing/` - and running a `df -k` suggests it’s not out of space:
```plaintext
[localadmin@NAS-140 syncthing]$ df -k
Filesystem 1K-blocks Used Available Use% Mounted on
none 442368 360052 82316 81% /
devtmpfs 6089744 4 6089740 0% /dev
tmpfs 65536 1788 63748 3% /tmp
tmpfs 6108432 156 6108276 0% /dev/shm
tmpfs 16384 0 16384 0% /share
/dev/sdi5 7928 47 7473 1% /mnt/boot_config
tmpfs 16384 0 16384 0% /mnt/snapshot/export
/dev/md9 505300 165632 339668 33% /mnt/HDA_ROOT
zpool1 55435834428 250 55435834179 0% /zpool1
zpool256 392200 144 392056 0% /zpool256
zpool1/zfs1 1048576 183910 864666 18% /share/ZFS1_DATA
zpool1/zfs5 104857600 3510 104854090 0% /share/ZFS5_DATA
zpool1/zfs18 68987912192 20586480078 48401432114 30% /share/ZFS18_DATA
zpool1/zfs530 55437407268 1573090 55435834179 0% /share/ZFS530_DATA
cgroup_root 6108432 0 6108432 0% /sys/fs/cgroup
/dev/md13 426984 393432 33552 92% /mnt/ext
tmpfs 32768 27900 4868 85% /samba_third_party
tmpfs 49152 32 49120 0% /share/ZFS1_DATA/.samba/lock/msg.lock
df: /mnt/ext/opt/samba/private/msg.sock: Permission denied
zpool1/zfs18/RecentlySnapshot
55435834424 246 55435834179 0% /share/ZFS18_DATA/ ***** SharedFolder/@Recently-Snapshot
tmpfs 29688 19260 10428 65% /mnt/ext/opt/FileStation5
/dev/loop0 29688 19260 10428 65% /mnt/ext/opt/FileStation5
tmpfs 10240 0 10240 0% /tmp/wfm
zpool1/zfs19 53687091200 366429534 53320661666 1% /share/ZFS19_DATA
zpool1/zfs19/RecentlySnapshot
55435834384 206 55435834179 0% /share/ZFS19_DATA/ ******** UP2/@Recently-Snapshot
```
---
### Author:  [@Moisie](https://forum.syncthing.net/u/Moisie)
#### Post date: [August 28, 2025, 2:04pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/6 "2025-08-28T14:04:54Z")
I’ve just nuked the DB from orbit, and we’ll see what prevails.
---
### Author:  [@Moisie](https://forum.syncthing.net/u/Moisie)
#### Post date: [August 28, 2025, 3:25pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/7 "2025-08-28T15:25:12Z")
Sadly nothing positive:
In case there’s any use in it, I see the following in my `index-v2` folder:
```plaintext
[localadmin@NAS-140 index-v2]$ ls -lah
total 1016M
drwxrwx--- 2 admin administrators 8 2025-08-29 00:02 ./
drwxrwx--- 6 admin administrators 30 2025-08-29 00:02 ../
-rw-r--r-- 1 admin administrators 1.1G 2025-08-29 00:38 folder.0001-mdyzme37.db
-rw-r--r-- 1 admin administrators 32K 2025-08-29 00:49 folder.0001-mdyzme37.db-shm
-rw-r--r-- 1 admin administrators 8.0M 2025-08-29 00:49 folder.0001-mdyzme37.db-wal
-rw-r--r-- 1 admin administrators 36K 2025-08-29 00:57 main.db
-rw-r--r-- 1 admin administrators 32K 2025-08-29 00:59 main.db-shm
-rw-r--r-- 1 admin administrators 4.0M 2025-08-29 00:59 main.db-wal
```
…and the following when I get the database stats (if I’ve found the right incantation):
```plaintext
[localadmin@NAS-140 syncthing]$ sudo ./syncthing debug database-statistics --home=/share/ZFS530_DATA/.qpkg/syncthing
Password:
DATABASE FOLDER ID TABLE SIZE FILL
======== ====== == ===== ==== ====
main.db - folders 4 KiB 1.2 %
main.db - folders_database_name 4 KiB 0.9 %
main.db - kv 4 KiB 23.1 %
main.db - schemamigrations 4 KiB 3.3 %
main.db - sqlite_autoindex_folders_1 4 KiB 0.6 %
main.db - sqlite_autoindex_kv_1 4 KiB 12.9 %
main.db - sqlite_schema 4 KiB 26.8 %
main.db - sqlite_sequence 4 KiB 0.5 %
main.db - sqlite_stat1 4 KiB 3.7 %
folder.0001-mdyzme37.db arrkq-wqckd blocklists 888048 KiB 97.5 %
folder.0001-mdyzme37.db arrkq-wqckd blocks 42728 KiB 99.5 %
folder.0001-mdyzme37.db arrkq-wqckd counts 4 KiB 5.2 %
folder.0001-mdyzme37.db arrkq-wqckd devices 4 KiB 3.7 %
folder.0001-mdyzme37.db arrkq-wqckd fileinfos 23188 KiB 96.7 %
folder.0001-mdyzme37.db arrkq-wqckd files 21932 KiB 96.9 %
folder.0001-mdyzme37.db arrkq-wqckd files_blocklist_hash_only 3464 KiB 89.4 %
folder.0001-mdyzme37.db arrkq-wqckd files_device_name 17996 KiB 80.2 %
folder.0001-mdyzme37.db arrkq-wqckd files_name_only 17624 KiB 80.9 %
folder.0001-mdyzme37.db arrkq-wqckd files_remote_sequence 1248 KiB 87.2 %
folder.0001-mdyzme37.db arrkq-wqckd indexids 4 KiB 1.5 %
folder.0001-mdyzme37.db arrkq-wqckd kv 4 KiB 0.8 %
folder.0001-mdyzme37.db arrkq-wqckd mtimes 4 KiB 0.2 %
folder.0001-mdyzme37.db arrkq-wqckd schemamigrations 4 KiB 3.3 %
folder.0001-mdyzme37.db arrkq-wqckd sqlite_autoindex_blocklists_1 2648 KiB 89.3 %
folder.0001-mdyzme37.db arrkq-wqckd sqlite_autoindex_blocks_1 43468 KiB 88.6 %
folder.0001-mdyzme37.db arrkq-wqckd sqlite_autoindex_devices_1 4 KiB 3.6 %
folder.0001-mdyzme37.db arrkq-wqckd sqlite_autoindex_kv_1 4 KiB 0.6 %
folder.0001-mdyzme37.db arrkq-wqckd sqlite_schema 12 KiB 44.8 %
folder.0001-mdyzme37.db arrkq-wqckd sqlite_sequence 4 KiB 1.0 %
folder.0001-mdyzme37.db arrkq-wqckd sqlite_stat1 4 KiB 10.9 %
folder.0001-mdyzme37.db arrkq-wqckd (total) 1062396 KiB 96.5 %
main.db + children - (total) 1062432 KiB 96.5 %
```
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [August 28, 2025, 3:32pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/8 "2025-08-28T15:32:56Z")
Support bundle?
---
### Author:  [@Moisie](https://forum.syncthing.net/u/Moisie)
#### Post date: [August 28, 2025, 3:36pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/9 "2025-08-28T15:36:11Z")
[Support bundle sent privately]
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [August 28, 2025, 5:40pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/10 "2025-08-28T17:40:38Z")
Support bundle doesn’t really give more clues than it just suddenly starts happening. Googling the error indicates that it can also have to do with various temporary storage locations being full. I don’t see `/tmp` being full in your paste above, but perhaps there are additional quotas, memory limits, container mounts, something happening when Syncthing is started?
---
### Author:  [@Moisie](https://forum.syncthing.net/u/Moisie)
#### Post date: [August 28, 2025, 7:04pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/11 "2025-08-28T19:04:21Z")
Thanks Jakob - I do appreciate your looking into it.
I saw references to `/tmp` too - but unless it’s a very spiky fill-and-then-empty occurrence, I’m seeing much the same level of usage during the problem as noted earlier (it’s also almost-identical to this Qnap’s companion, which is working quite happily thankyou):
> [@Moisie](#):
>
> ```plaintext
> [localadmin@NAS-140 syncthing]$ df -k
> Filesystem 1K-blocks Used Available Use% Mounted on
> ...
> tmpfs 65536 1788 63748 3% /tmp
> ...
>
> ```
I suspect I’m not going to be able to bring much more to bear on this, and as it’s clearly not a widespread issue, I’ll probably roll back to v1 and live with it there. I’m hopeful this Qnap will be replaced at some point soon anyway, at which point it will all become academic anyway.
---
### Author:  [@gadget](https://forum.syncthing.net/u/gadget)
#### Post date: [August 30, 2025, 1:43am UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/12 "2025-08-30T01:43:45Z")
> [@Moisie](#):
>
> ```plaintext
> [localadmin@NAS-140 syncthing]$ df -k
> Filesystem 1K-blocks Used Available Use% Mounted on
> none 442368 360052 82316 81% /
> devtmpfs 6089744 4 6089740 0% /dev
> tmpfs 65536 1788 63748 3% /tmp
>
> ```
Although `/tmp` usage was low at the time of the output above, the capacity is only 65MB, which is unusually low.
When `/tmp` is a ramdisk (indicted by the “tmpfs” filesystem) instead of on-disk, it’s often a percentage of system RAM, commonly 50%.
What model QNAP is it and what does the `free` command say?
---
### Author:  [@Moisie](https://forum.syncthing.net/u/Moisie)
#### Post date: [August 31, 2025, 8:49pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/13 "2025-08-31T20:49:15Z")
Hi @gadget - thanks for your input.
> [@gadget](#):
>
> Although `/tmp` usage was low at the time of the output above, the capacity is only 65MB, which is unusually low.
I compared this with this NAS’s sibling, and the capacity on `/tmp` was the same; I didn’t personally set either system up from scratch, but I doubt anything unusual would have been configured here - it’s probably just the Qnap default.
> [@gadget](#):
>
> What model QNAP is it
It’s a QNAP TS-873A, with 12GB RAM installed.
> [@gadget](#):
>
> what does the `free` command say?
Probably irrelevant as I’ve now removed v2.0.3 and installed v1.30.0 instead - running without issue so far - but `free` currently reports:
```plaintext
[localadmin@NAS-140 ~]$ free
total used free shared buffers cached
Mem: 12216864 9312576 2904288 415044 35864 963320
-/+ buffers/cache: 8313392 3903472
Swap: 63736824 0 63736824
```
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [August 31, 2025, 9:03pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/14 "2025-08-31T21:03:26Z")
This may or may not be useful, but since the error is via SQLite and appears resource related… I’ve seen cases where problems like thread limits, file descriptor limits, etc end up reported as general ”ENOSPC” type errors that look like out of disk space.
---
### Author:  [@rebelman](https://forum.syncthing.net/u/rebelman)
#### Post date: [September 7, 2025, 10:37am UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/15 "2025-09-07T10:37:31Z")
Hi there. Same problem here. I have a TS-873A with 64 GB RAM, 50 TB free space, enough inodes left. Same error (disk or database full). 4 other folders between the two machines synced without issues before (the other Device is a Win PC with current SyncTrayzor). To be more detailed for the QNAP: Its the Syncthing Version 2.06 from MyQnap, in this case the root version. Any hope to have this be handled?
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [September 7, 2025, 11:04am UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/16 "2025-09-07T11:04:40Z")
I’ve narrowed this down to likely be caused by a tiny `/tmp` (64 MB?) where SQLite wants to store temp files for large queries. We can work around that in various ways, none of them super palatable, but we’ll find something. In the meantime, you can try launching with the `SQLITE_TMPDIR` env var set to an _existing, writeable_ directory with more space in it…
>
>
> \### What happened?
>
> syncthing tells me "database or disk is full"
> the disk can't… be full:
> pwd
> /share/CACHEDEV2\_DATA/.qpkg/SyncThing/index-v2
> df
> Filesystem Size Used Available Use% Mounted on
> tmpfs 64.0M 35.4M 28.6M 55% /tmp
> /dev/mapper/cachedev2
> 12.8T 1.6T 11.2T 13% /share/CACHEDEV2\_DATA
>
> \
> \
>
> Folder ID | m9xvj-celqf
> \-- | --
> /share/Backups/LENOVO-Chris
> database or disk is full
> 507’744 57’062 ~605 GiB
> 15’818 30’676 ~52.2 GiB
> 518’312 items, ~553 GiB
> Receive Only
>
>
> \
> \
>
> it changes always between Preparing to Sync and Waiting to Sync.
>
> \[support-bundle-4FKKX5H-2025-09-05T160349.zip\](https://github.com/user-attachments/files/22173676/support-bundle-4FKKX5H-2025-09-05T160349.zip)Folder ID m9xvj-celqf
> Folder Path /share/Backups/LENOVO-Chris
> Error database or disk is full
> Global State 507’744 57’062 ~605 GiB
> Local State
> 15’818 30’676 ~52.2 GiB
> Out of Sync Items \[518’312 items, ~553 GiB\](https://192.168.99.122:8384/)
> Folder Type Receive Only
>
>
>
> \### Syncthing version
>
> 2.0.7
>
> \### Platform & operating system
>
> Linux (64-bit Intel/AMD)
>
> \### Browser version
>
> \_No response\_
>
> \### Relevant log output
>
> \`\`\`shell
> see added support-bundle file
> \`\`\`
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [September 7, 2025, 11:41am UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/17 "2025-09-07T11:41:58Z")
Likely fix:
>
>
> On Unixes, avoid the /tmp which is likely to become the chosen default.
---
### Author:  [@rebelman](https://forum.syncthing.net/u/rebelman)
#### Post date: [September 7, 2025, 12:57pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/18 "2025-09-07T12:57:44Z")
Thank you very much for your efforts! Unfortuanatly I dont know how to change this env variable on the QNAP, but I saw the progress in the change logs and look forward 🙂 A change of the backend is possibly a very big thing. Thanks.
---
### Author:  [@rebelman](https://forum.syncthing.net/u/rebelman)
#### Post date: [September 9, 2025, 2:23pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/19 "2025-09-09T14:23:23Z")
The commit is through … it worked immediatly. Thanks again
---
### Author:  [@system](https://forum.syncthing.net/u/system)
#### Post date: [October 9, 2025, 2:24pm UTC](https://forum.syncthing.net/t/database-or-disk-is-full-might-be-syncthing-might-be-qnap/24930/20 "2025-10-09T14:24:08Z")
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.