NAS library dissappears

Hi There

Good looking software, waiting for my HiFiberry :slight_smile:

I seem to be having a problem on my Raspberry Pi. I have set up a NAS connection and the scanning happens (takes a long time). But on reboot the music library NASS does not appear on browse.

So far the only way around it is to restart the MPD DB update.

Am I doin something wrong?

Are there any diagnostics I can run to check?



I have the same problem
I have a nas with a lot of music … and after reboot its library is gone and I have to do a manual update


Philip Van Cleven

Please reboot, ssh to your rpi and do a sudo mount Is your NAS present?


The nas is mounted at reboot (automatically) and there is a green symbol above the nas settings

Mount command shows the nas is mounted …

A refresh command starts the library index ing without the need to mount or re-add the nas. The nas is available 24 by 7 and is a dlink 320


an update
I rechecked the library file and now it is there (even after a few restarts) … I did use the option to let the library check if nothing changed on the NAS …
I also changed to a fixed wlan address …
and I let the library update for about 16 hours (a rather large collection of CDs I bought in the 80tish and 90tish) … maybe because I let it update the library completely without any interruption it wrote the library file to the SD card instead of keeping it in temp?


I am having a similar issue more or less… As soon as I press the button webInterface-menu-library-updateDatabase, the entire database is deleted and the library is being rebuild. Maybe this is a feature, not sure. But half way rebuilding the database, the indexing stops. And only a small part of my NAS is being indexed. Although the tab on the bottom left keeps showing ‘updating’.

In Raspify, the process of building a database was working correctly and was able to see the number of entries growing each time I refreshed the web interface.

Is it possible to find a log somewhere on the r-pi, to find some information on the updating process?

Is it possible that only part of my NAS is both read/write and most of it is read-only? (although, again, this hasnt been a problem before…)

Kind regards,

As a side note, automatic update feature is switch off in the web interface

You can try disabling ffmpeg

put this

decoder { plugin "ffmpeg" enabled "no" }

in /etc/mpd.conf

restart mpd with /etc/init.d/mpd restart

and let me know . I’ll plan to include a swith for that in next version

Thank you for your time and suggestion. however, It didnt work…

When I run the command:

tail /var/log/mpd/mpd.log

after a long list (its a 32MB log-file) of filenames being correctly added to the database, there was the following error the first time i checked the log, on the last line:
src/directory.c:93: directory_get_name: Assertion `(slash == ((void *)0)) == directory_is_root(directory->parent)’ failed.

After another try, the rebuilding database process showed the following error:
[parse.c:1078] error: Giving up searching valid MPEG header after 65536 bytes of junk.

Which linked me to the following thread:

Still havent found a solution, ill keep looking.


Maybe it’s a good idea to include some tail error logging in the website (mpd, nginx, smb mounting, etc) layout. This would definitely makes helping eachother out on the fora a lot easier, and speeds up the process of smashing out the last couple of bugs;)

Old topic but I have the same problem.

Every now and than, my complete library is wiped after what volumio claims was a bad connection to the NAS. I have to completely rebuild the library again. It seems that this happens when I have not turned on the pi for a longer time.

Anyone who had similar issues, waiting a couple hours each time on boot to be able to play music makes this player completely unusable to me :frowning:

I am using the latest version 2.246 and i also face the same issue. Every now and than my music library gets wiped out. Also while updating the database if volumio is shutdown/restarted, it losses all the progress and in next restart it show same old indexed status. its really frustrating that there is no fix so far for this issue and it takes good 7+Hrs to scan a library of approx 60K every time to rebuild the music database index.

same here for 2.246 fresh install, it seems nothing is persisted, wifi pass lost, cifs mount lost … on reboot.

where are the data persisted ?

looks everything is readonly on this version:
/var/lib/dpkg/info/tzdata.postinst: 29: /var/lib/dpkg/info/tzdata.postinst: cannot create /etc/timezone: Structure needs cleaning

image issue ?

[ 1351.200722] EXT4-fs error (device mmcblk0p3): ext4_mb_generate_buddy:758: group 1, block bitmap and bg descriptor inconsistent: 23949 vs 19459 free clusters
[ 1392.892003] EXT4-fs error (device mmcblk0p3): ext4_find_dest_de:1840: inode #24419: block 2323: comm tzdata.postinst: bad entry in directory: rec_len is smaller than minimal - offset=0(0), inode=0, rec_len=0, name_len=0
[ 1528.392176] EXT4-fs error (device mmcblk0p3): __ext4_new_inode:1066: comm node: failed to insert inode 48833: doubly allocated?

=> image issue, took another sdcard and it now works fine