firefly crashes at boot (uNSLUng) :?:

[Anonymous]

hello,

since some days the firefly mediaserver crashed at boot of my uNSLUng box. The log file reads:

cat /var/log/mt-daapd.log

2008-08-26 20:18:00 (00000400): Firefly Version svn-1696: Starting with debuglevel 3

2008-08-26 20:18:00 (00000400): Plugin loaded: ssc-script/svn-1696

2008-08-26 20:18:00 (00000400): Plugin loaded: daap/svn-1696

2008-08-26 20:18:00 (00000400): Plugin loaded: rsp/svn-1696

2008-08-26 20:18:00 (00000400): Starting rendezvous daemon

2008-08-26 20:18:00 (00000400): Starting signal handler

2008-08-26 20:18:00 (00000400): Initializing database

2008-08-26 20:18:57 (00000400): Query: vacuum

2008-08-26 20:18:57 (00000400): Error: disk I/O error

2008-08-26 20:18:57: Aborting

2008-08-26 20:18:57 (00000400): Rendezvous socket closed (daap server crashed?) Aborting.

2008-08-26 20:18:57: Aborting

Some search brought up that the database may be corrupt. I opened it with sqlite, dumped the content but could not find any errors. So I deleted the file (/opt/var/mt-daapd/songs.db). Then firefly started as normal and generated a new database, afterwards it served songs. But after reboot firefly crashed with the same message as shown above.

Yust in case it is a memory issue, here the df output as well

df

Filesystem           1k-blocks      Used Available Use% Mounted on

rootfs                76683552  33178240  42726244  44% /

/dev/sda1                 6528      6348       180  97% /initrd

/dev/sda1             76683552  33178240  42726244  44% /

/dev/sda1             76683552  33178240  42726244  44% /share/flash/data

/dev/sda2               116661      4181    111276   4% /share/flash/conf

Any idea whats wrong here?

Goemon 🙄 ❓

[fizze]

The database file is most likely corrupted. Delete it, and then start firefly again.

You may also try to boot without the HD, and execute an fsck, or e2fsck to repair any issues. But rebuilding the db from scratch is usually a safe bet.

[Anonymous]

mhhh,

does not run stable again again yet. as I said, I had tried to delete the songs.db – the failure occured again after reboot. Anyway, if I start firefly serveral times (5-7) from commandline with “S80mt-daapd” I get the same log every time, than suddenly it rebuilds the database and runs until reboot.

Now an other try: changed database settings to use sqlite3 and deleted again the current songs.db – reboot was successfull. I will keep an eye on this.

So I probably will loose my ratings in the database and the smart playlists I had… any idea how to transfer them in the new sqlite3 database from the old sqlite file?
And last question: any idea out there how to write the song rsting direct into the ID3 tag of my mp3 files?

Goemon

[fizze]

yes, ron has outlined the commands to migrate the DB from sqlite2 to sqlite3 in the forums.

Sqlite2 is buggy with svn-1696 on the slug! sqlite3 isn’t, however.

[Author]

Posts