Help with disk I/O error

[Anonymous]

I’m running 1498 on a D-Link DNS-323 NAS with a 50,000+ song library and everything has been running fine for months until this week when I added another 15 GB of music.

I suspect that when Firefly runs this query:

Query: delete from songs where id not in (select id from updated)

it’s timing out and that this may be due to the limited horsepower on the 323 and a ‘low’ timeout value in sql-lite or the Firefly app.

I’ve had no problems copying, adding deleting files outside of Firefly so I don’t believe the issue is disk related.

Anyone have any ideas on this? How,where do I increase the timeout value for this operation.

I’m a bit of a linux noob and I’m running pre-compiled libraries for the 323 so I’d like to avoid upgrading if possible.

Any help is appreciated! 😀

2008-03-30 11:22:49 (00000400): Firefly Version svn-1498: Starting with debuglevel 2
2008-03-30 11:22:49 (00000400): Starting rendezvous daemon
2008-03-30 11:22:49 (00000400): Starting signal handler
2008-03-30 11:22:49 (00000400): Initializing database
2008-03-30 11:25:22 (00000400): Starting web server from /mnt/HD_a2/lnx_bin/admin-root on port 3689
2008-03-30 11:25:22 (00000400): Registering rendezvous names
2008-03-30 11:25:27 (00000400): Serving 51592 songs. Startup complete in 153 seconds
2008-03-30 11:25:27 (00000400): Rescanning database
2008-03-30 12:13:02 (00000400): Starting playlist scan
2008-03-30 12:13:16 (00000400): Query: delete from songs where id not in (select id from updated)
2008-03-30 12:13:16 (00000400): Error: disk I/O error
2008-03-30 12:13:16: Aborting
2008-03-30 12:13:16 (00000400): Rendezvous socket closed (daap server crashed?) Aborting.
2008-03-30 12:13:16: Aborting

[mas]

Such an error indicates a hardware problem with the harddrive.

Enable smart, search for unrecoverable errors and run a general HD check with some specialized tools.

[Anonymous]

Hey, Thanks for the reply.

I actually solved the issue by moving the songs.db to another directory and doing a full rescan.

I suspect that a power outage we suffered a couple weeks ago may have caused corruption to songs.db.

[Author]

Posts