Is there any reason i couldn't put a MkIIa motherboard into a MkII case, keeping the original MkII screen and harnesses? Since i've got 70GB on my MKII, the drives are ALWAYS spun up and it's overheating regularly on trips of half an hour or more. It needs more RAM. And until the extra RAM hack is worked out, i need a quick fix.

Pretty sure i know the answer to this, but just wanna make sure before i do it.

Okay. Obvious question here but couldn't you just move the drives over instead?

Okay. Obvious question here but couldn't you just move the drives over instead?

I believe he also wants the display board assembly from the 2A because it's got the newer type of rotary encoder.

What you want to do, Loren, is do-able I think. The display boards are meant to be interchangeable. My only problem is... I don't think the extra RAM is going to make a difference in the temperature. There's not going to be a huge difference in spin-up times. It will only be slight, not enough to make a thermal difference.

I think it might be time for the big step. Go see if WeirdStuff still has that big styrofoam crate full of tiny fans.

No. He wants to keep the old display and stuff, for whatever reason. And it's not going to affect the spinup, but, rather, keep the drive from spinning all the time by increasing the amount of cache available.

Here's why i think it will make a difference: The same drive configuration, two 40 gb drives, has been in the WRX since i installed the empeg, no change there. I did get overheating, but once i drilled vent holes into the pocket above the empeg, no more overheating. It's been fine for months that way with daily use. Then last week i uploaded another 30 gigs or so that i'd been retagging and now the drives are literally ALWAYS spun up according to the disk activity icon. So after half an hour it gets overheated. The drives being spun down at all seemed to help.

And as for why i don't just use the MkIIa... the way i have my sled installed, i can't even fit a MkIIa into it. It's bolted in from the side and is one of the coated black sleds. Despite the size of the empegs supposedly being exactly the same, i can tell you from experience they definitely are not. My MkII slides in fine with plenty of room, but the MkIIa won't go pas the first sled nipple... which can't be bent out due to the mounting bracket in the WRX. Should be an easy swap so i'll give it a try before i venture into fan world. Unless i have one... *me goes digging in my junk computer box*

Thanks!

To be honest, I am kind of curious why the drives are not spinning down. Do you have that disabled through hijack? Have you uploaded different filetypes (flac)? Adding more music should not change drive behavior like that.

Wish i knew. I haven't disabled it, and i'm still running good old VBR MP3s. I just figured the database was taking up all the RAM. I'm still running 2.0Final

Hmmm... Something sounds a bit strange. Try manually fsck-ing the drives. Also, try going into your config.ini and verify your settings. See what kind of memory reserve you have, any applications that are loaded that may be logging or keeping the drives spinning. I would almost recommend backing up your existing config.ini and putting a clean one in its place (not loading any additional software or memory tweaks) just to see if it makes a difference. Also try inserting a forced spindown time to see if it changes things.

Doing all of the above now...

Definitely nothing weird in the config.ini:

[hijack]
khttpd_root_index=/drive0/opt/khttpd/charcoalgray99/index.html
[output]
notify=1
[hush]
hushpercent=20
[display]
play_count=1
visual_names=1
caching=1
[volumecap]
enabled=0
[kenwood]
disabled=0
[ramp]
disabled=0
[menu]
quit=0
sort_playlist=0
[tuner]
region=us
[controls]
stalk_side=none
[Options]
Name=emp3g
[User Info]
Email=loren@*******.com
Phone=415-*******
Name=Loren Cox
[Network]
;@HOME DHCP=0
;@WORK DHCP=1
Gateway=192.168.1.1
Netmask=255.255.255.0
IPAddress=192.168.1.20
DHCP=0
[sense]
muteaction=1
mute=1
[power]
off_timeout=60000
[aux]
source=
title=
artist=
[synchronise]
seq_num=88
[wendy]
flag_amount=0
[custom]
wendy=
[searches]
amount=0
[VisualFavourites_AM]
count=0
[VisualFavourites_AUX]
count=0
[VisualFavourites_DSP]
count=0
[VisualFavourites_FM]
count=0
[shuffle]
deduplicate=0

Running fsck...

K.... seems that i've found something. I plugged er up to the serial and i'm getting this message after reboot:

Vcb: 0x4058a000
spindown_chunk_cache_runner.cpp: 294:Filling up space (-12)
spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)
spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)
spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)
spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)
spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)
.....ad nauseum....

any ideas?

a quick search shows Roger saying these aren't anything important. Running fsck now.

a quick search shows Roger saying these aren't anything important.

Yeah, except your numbers are negative. In other words, the cache is telling you something you already know: the database is taking up so much memory that there isn't enough cache left ever to spin down.

Out of interest, how big is your database file (ls -l /empeg/var)?

Peter

I'll check as soon as this fsck is over... could be a while... or is it okay to interupt an fsck?

edit: okay, fsck done, with this result on the first drive:

/dev/hda4: 18643/304448 files (23.3% non-contiguous), 30281113/38965657 blocks

and ls -l /empeg/var results in:

lrwxrwxrwx    1 0        220            13 Apr  1  2003 /empeg/var -> ../drive0/var

ls -al /drive0/var

empeg:/empeg/bin# ls -al /drive0/var
ls: /drive0/var: No such file or directory

looks like it needs to be rebuilt eh?

Huh???

Okay... that is weird... Oh, did you unmount the drives to fsck them? Try first doing "rom" and "ro". Reboot. Go back to shell. "ls -al /drive0/var"...

Yeah, just realized i gotta finish the fsck first =] brb...

So you know, what I (and probably Roger also) am looking for is something that looks abnormal with one of your database files. Something that would distort the player application and make the database image in memory so large that it could not possibly spin down the drives.

For reference, on my player (with lots and lots of FIDS), my files are:

q

Restored terminal settings

Remounting first music partition read-only

Remounting second music partition read-only

Player exited normally: 0

Switching to shell-player loop

Starting bash.

empeg:/empeg/bin#

empeg:/empeg/bin# ls -al /drive0/var

total 4255

drwxr-xr-x    2 root     root         1024 Feb 12 10:58 .

drwxr-xr-x    6 root     root         1024 Feb  4 10:45 ..

-rw-------    1 root     root         1594 Dec 21 14:16 config.ini

-rw-rw-r--    1 root     root      4201145 Feb 12 10:57 database

-rw-rw-r--    1 root     root       131836 Feb 12 10:58 playlists

-rw-rw-r--    1 root     root          376 Feb 12 10:57 tags

empeg:/empeg/bin#

Granted, yours will not be the exact same size, but if they look significantly bigger, or the directory structure looks different, you may want to try to manually rebuild the databases.

Hm...... a database of 70gigs of music is filling the MKII ram, but Paul's database of 160gig isn't filling the MKIIa ram? So that few meg ram difference DOES matter? Or are Paul's drives spun up all the time? And is his player overheating because of it?

And to reply to an earlier post in the thread re: flac and alternative file types: Would it keep the drives spun up to play FLAC or WAV? I'm assuming yes because it's a bit hard to cache an 80 meg song in 16 meg ram.

Hm...... a database of 70gigs of music is filling the MKII ram, but Paul's database of 160gig isn't filling the MKIIa ram?

Well, that is the suspected cause. Still unproven. Just a theory.

So that few meg ram difference DOES matter?

The memory does help quite a bit.

Or are Paul's drives spun up all the time? And is his player overheating because of it?

Not that I know of, on both counts.

And to reply to an earlier post in the thread re: flac and alternative file types: Would it keep the drives spun up to play FLAC or WAV? I'm assuming yes because it's a bit hard to cache an 80 meg song in 16 meg ram.

Exactly. Since you need more data per second, you cannot effectively cache that kind of file as easily. Or perhaps, you can cache it, but you burn through it so quickly that it really doesn't help.

empeg:/empeg/bin# ls -lah /drive0/var

total 3.1M

drwxr-xr-x    3 0        0            1.0k Jul  2 04:41 .

drwxr-xr-x    6 0        0            1.0k May 11  2035 ..

-rw-r--r--    1 0        0             810 Jul  2 04:41 config.ini

-rw-r--r--    1 0        0            3.0M Jul  2 04:41 database

-rwxr-xr-x    1 0        0             25k Mar 10  2002 date

-rwxr-xr-x    1 0        0            4.4k Mar 10  2002 pcmplay

-rw-r--r--    1 0        0             75k Jul  2 04:41 playlists

-rwxr-xr-x    1 0        0            3.7k Mar 10  2002 setvol

-rw-r--r--    1 0        0             417 Jul  2 04:41 tags

drwxr-xr-x    2 0        0            2.0k Mar 10  2002 tts

-rwxr-xr-x    1 0        0            2.5k Mar 10  2002 ttsclock

wonder if that old tts stuff is doing it. I can just nuke this whole directory right? It'll all get rebuilt?

okay... Nothing immediately catches my attention there...

In your config.ini, in the display section, you have caching=1. I am not familiar with that option, and I am looking it up. What does this do?

Rebuilding the database after deleting all the TTS stuff.

no idea about the caching=1 thing... don't recall adding that one.

I believe that just enables the disk activity icon... I think...

I found it. Basically the option that turns on the display of disk activity...

One thought (back to the original subject). If your player is running that hot (granted, with the drives spinning all the time) you may still want to install a cooling fan. Just to be safe.

okay... databases have been rebuilt, and both drives fscked.

Rebooted and still getting this:

Vcb: 0x4058a000

  spindown_chunk_cache_runner.cpp: 294:Filling up space (-12)

  spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)

  spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)

  spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)

  spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)

  spindown_chunk_cache_runner.cpp: 294:Filling up space (-15)

Any other ideas?? I'm going to run a ram check...

Which kernel are you running? Default? Hijack? Custom?

hijack 363 v2final

Error after ram test:

tplayer(13): memory violation at pc=0x00000000, lr=0x020caf3c (bad address=0x000

00000, code 2)

pc : [<00000000>]    lr : [<020caf3c>]

sp : 00000000  ip : bf1ffac0  fp : 00000000

r10: 00000000  r9 : 00000000  r8 : 00000000

r7 : 00000000  r6 : 00000000  r5 : 00000000  r4 : 00000000

r3 : 00000000  r2 : 00000000  r1 : 00000001  r0 : 00000001

Flags: nzCv  IRQs on  FIQs on  Mode USER_32  Segment user

Control: D036517D  Table: D036517D  DAC: 00000015

that doesn't look good

Perhaps a re-flash of the kernel? Other than that, I am at a loss.