Wow!

Congratulations!

This is interesting. My first reaction is to say that's a symptom of cable trouble or IDE header trouble, and that it's got nothing to do with the builder software.

Mark, can you confirm or deny the concept that the drive's position on the cable could affect the builder? Is there anything, even at the lowest level, that could cause that behavior he observed?

I can understand possibly needing to have only one disk drive plugged in for the builder to work. But having it need to be on a certain header? That just sounds like hardware/header/cable trouble to me.

The cable is not keyed for master/slave, so a single drive should always be at the end of the cable to reduce signal ringing.

If it doesn't work on the end, then there's a cable/connector fault.

Cheers

...and, according to the behavior he cited, it was an intermittent fault, making it look like the position on the cable was some sort of a requirement.

Well, hang on, 360GB is a lot but it's not as much as 2x250GB. Is that a typo for 460GB, or could there be some problem where one of the drives got formatted at 128GB by mistake? 232GB+128GB is suspiciously 360GB...

Peter

Hi,

I haven't had to worry about drive placement on the cables. Formatted alot of Dual 250GB drive Empeg systems. I haven't used the latest builder though.

Yeah, he should see 244GB for each drive on the Empeg Vitals display (push and hold the round knob on the player for a few seconds, turn it till it highlights Vital Signs, push the knob again).

The first line should say Mk2a: 16MB, 244+244GB.

In Emplode (Windows) it should say at the bottom of the display window 465GB.

That would indicate that you have 2 drives with full formatted (pumped) capacity.


****************************************************************
Anyone tried to use a PATA-to-SATA adapter yet with the 250GB or larger 2.5 inch SATA drives? WD has some 750GB & 1TB drives in the 12.5mm high form factor. I would like to have one drive plus the PATA-to-SATA adapter (I have already modified one for proper connectoring).

Have to get the PATA to SATA to work first (if it will work at all), then I'll probably run out of playlist RAM, adjust cache, re-drill the mounting bracket for the 12.5mm height, and so on, and so on. Maybe it can't work and it will put me out of my misery. Seems like a logical upgrade path for the Empeg, now that PATA laptop drives seemed to have topped out at 320GB.

Am I out of my mind Mark? Probably my next Empeg project...

****************************************************************

Thanks,

Ross

Vital Signs does say 244+244GB so OK on the build.

The 360 is what Emplode (Windoze file version 2.0.19.0) displays as the capacity on the bottom right corner.

My appologies for any confusion.

Hijack shows the "raw" capacity as reported by the drive and doesn't care what partitions are set up as I understand it.

Emplode shows the partition capacity and this is probably wrong.

So I would say from what you're telling us, your build is not ok.

run these two commands:
fdisk -l /dev/hda
fdisk -l /dev/hdc

Yep, you are correct, it looks like the first drive has been setup incorrectly.
Couldn't be the second one could it, nah too easy...

empeg:/empeg/bin# fdisk -l /dev/hda

Disk /dev/hda: 255 heads, 63 sectors, 30401 cylinders
Units = cylinders of 16065 * 512 bytes

Device Boot Start End Blocks Id System
/dev/hda1 1 5 40131 5 Extended
/dev/hda2 6 10 40162+ 83 Linux
/dev/hda3 11 13 24097+ 10 OPUS
/dev/hda4 14 16709 134110620 83 Linux
/dev/hda5 1 3 24034+ 83 Linux
/dev/hda6 4 5 16033+ 82 Linux swap

empeg:/empeg/bin# fdisk -l /dev/hdc

Disk /dev/hdc: 255 heads, 63 sectors, 30401 cylinders
Units = cylinders of 16065 * 512 bytes

Device Boot Start End Blocks Id System
/dev/hdc1 1 11 88326 5 Extended
/dev/hdc2 12 16 40162+ 83 Linux
/dev/hdc3 17 21 40162+ 10 OPUS
/dev/hdc4 22 30401 244027350 83 Linux
/dev/hdc5 * 1 3 24034+ 83 Linux
/dev/hdc6 4 11 64228+ 82 Linux swap

Looks like it. You could swap the drives over, install the developer image, and then partition the (now) second disk manually.

Or you could just run the builder again...

Oddly the first drive appears to have been a troubled initial build in that it had 104.89GB unallocated and a rebuild from scratch (starting with bigdisk_v6 etc) on it's own, didn't clear that.

I removed it and manually cleared all partitions and then rebuilt it again and it came up as full size this time.

Re-assembled it as second disk and re-ran car2_c2.01_hijack.upgrade and it's now fine.

Total capacity now reported as 465GB :-)

I know I don't really need the build on both drives but it's proved handy this time as the drive with all the MP3's on were on the working drive. I'd only put 60GB on so far but that can take a while to do with a 10meg nic.

thanks for your help
Hodge

PS for confirmation:
empeg:/empeg/bin# fdisk -l /dev/hda

Disk /dev/hda: 255 heads, 63 sectors, 30401 cylinders
Units = cylinders of 16065 * 512 bytes

Device Boot Start End Blocks Id System
/dev/hda1 1 11 88326 5 Extended
/dev/hda2 12 16 40162+ 83 Linux
/dev/hda3 17 21 40162+ 10 OPUS
/dev/hda4 22 30401 244027350 83 Linux
/dev/hda5 * 1 3 24034+ 83 Linux
/dev/hda6 4 11 64228+ 82 Linux swap
empeg:/empeg/bin#
empeg:/empeg/bin# fdisk -l /dev/hdc

Disk /dev/hdc: 255 heads, 63 sectors, 30401 cylinders
Units = cylinders of 16065 * 512 bytes

Device Boot Start End Blocks Id System
/dev/hdc1 1 5 40131 5 Extended
/dev/hdc2 6 10 40162+ 83 Linux
/dev/hdc3 11 13 24097+ 10 OPUS
/dev/hdc4 14 30401 244091610 83 Linux
/dev/hdc5 1 3 24034+ 83 Linux
/dev/hdc6 4 5 16033+ 82 Linux swap

My attempt has been unsuccessful so far, using a WD2500BEAS SATA drive + adaptor.

The attached is the output immediately after running builder_bigdisk_v6.upgrade - are there any obvious errors?

That output all looks good (apart from being mangled by hyperterm).

Next step is to install the empeg software onto the drive.

To clarify:

Copy/Paste from Hyperterminal is horribly broken on XP and later. Use "Capture to text file" instead of copy/paste. Explained in a callout note in the relevant FAQ entry.

Ok I've loaded the software on, then tried an ambitious 60GB upload, and the sync failed with the following:

- Synchronise failed while checking media. Read on socket failed (error 0x8004003d)
- Player stuck in reboot cycle with '0000.-1 sigkill err' in a box
- emplode then crashed

I then manually rebuild the database and synced a single album - success

I am now attempting a 2GB transfer. After about 5 tunes were uploaded the sync appears to have stalled with some error messages on screen, along with some chatter over the serial port - see attached (sorry I didn't have the Capture Text feature running at that time, so its a mangled copy and paste job).

Probably running out of memory, due to the increased usage requirements of managing big drives.

I cannot remember if swap is normally enabled or not during a sync (?), which would help if it was.

Hi,

Is he having a problem with Cache size? I had to adjust mine for large playlists.

I wouldn't think that the SATA interface would use any additional memory. Is the SATA adapter reporting back unexpected information that has to be thrown away or conflicts with configuration files?

I guess I will play with this sometime next week to see how big a drive I can get to run too.

Ross

It is, but only on the primary disk.

Is there any way of increasing swap for the period of the sync?
This is a single drive installation btw (only room on the drive tray for 1 sata drive + adaptor)

No, not the physical device driver, but rather the filesystem layer on top. Bigger disk == bigger filesystem == bigger data structures to keep track of everything.

Cheers

Hi,



Mark Says...
No, not the physical device driver, but rather the filesystem layer on top. Bigger disk == bigger filesystem == bigger data structures to keep track of everything.




That was one of the reasons why I condensed by tags and file information (lots of information in tags) - to accomodate the existing structure.

I ran out of memory alot until I shrunk things down alot and adjusted cache.


Ross

Did you run out of memory during the initial sync (like me) or during normal player use?

Hi,

No, during normal player use. Actually, after a good sync on the first or second item in the play list. I would get a NOMEM xxxx on the display.

Ross

Hi Ross, quick query - what Reserve Cache did you use? I'm running a 250Gb drive, used the v10 big disk builder, and the latest 2.01 with hijack upgrade and jEmplode sees 232.59Gb so that is all good, but before I start to load up the new disk with songs I thought I'd try and get everything right.

Completed:
Build
Telnet - love the fact Mark's latest has telnet daemon built in, so no need for the additional hassle of installing that one

Things I want to do:
Set_max_fids - Can't run it (I get a text file in use error)
Fidsift - is this needed any more?
Set up the cache correctly

Stop the player application first.

Hi,

I used 16, but my file sizes are quite large.

Ross

Mark - yep, I went through it in my head, and that should have been obvious :-) Got it now... (must be getting old)

Ross - will give that a shot, think I used 10 with previous incarnation on v3 alpha but it is obviously a tad different with this.

Slowly adding tracks on - am up to 'ACDC' this evening - currently doing it in chunks bigger than a couple of gig has had issues. Will see how it goes.

Cheers

Rory

Hi,

I loaded chunks of 10GB sometimes and it was a little slower.

Mark, is V3 alpha a good base for the larger drives? I thought that we required 2.01 Software.


BTW, I think I recently found a problem with the file system getting too large. I recently loaded an additional 10GB to one drive set which loaded fine - completed normally. That made around 455 GB on the system drive set. When I tried to load on another 2 GB, it didn't complete. While debugging through the serial port using Hyperterminal, I noticed that it can't find the Tags.

(! tags.cpp : 61:Failed to open tags (0xc0041002).

Emplode just croaks, after it starts to rebuild the database, remounts the partitions, stalls and dies with a core dump for debug.

It takes forever to build the database upon power boot. Normal time when powered on from front button panel.


From the FAQ...

1) I performed fscks of both drives, took forever, per the FAQ. No Change.

2) I tried to restore the partition from all of the alternate locations (24577 was best), took forever, per the FAQ. No Change

3) Re-installed Emplode. No Change.

4) Tried 3 other Empegs that hadn't had the additional 10GB installed - they all work fine.

5) I tried to use USB (pronounced Ugh), instead of Ethernet, same thing.

6) Drive cable is good, other drives worked.

I think I broke the database. I don't think that Emplode is hosed, I think it is a player thing.

Funny thing is after it completes building the database, it runs fine. It plays fine, still takes forever to build the database if it is powered off.

I want to remove some and add some more with this system. With this system, I can't even get past initialization in Emplode to play with it. I could clone the drive set from another of my RIOs, but, I am curious as to what happened.

I will probably dump the drives contents and start a new database image anyway.

It's a Mk2a, 16MB RAM, 244+244GB, Player Software car2_v2.01, Builder_Bigdisk_v10.


Serial Port output string:


empeg-car bootstrap v1.02 20001106 (hugo@empeg.com)
If there is anyone present who wants to upgrade the flash, let them speak now,
or forever hold their peace...it seems not. Let fly the Penguins of Linux!

e000 v1.04
Copying kernel...
Calling linux kernel...
Uncompressing Linux..................................... done, booting the kerne
l.
Linux version 2.2.17-rmk5-np17-empeg55-hijack-v505 (hijack@rtr.ca) (gcc version
2.95.3 20010315 (release)) #1 Wed Dec 31 11:10:05 EST 2008
Processor: Intel StrongARM-1100 revision 11
Checking for extra DRAM:
c1000000: wrote ffffffff, read e28cc001
NetWinder Floating Point Emulator V0.94.1 (c) 1998 Corel Computer Corp.
empeg-car player (hardware revision 9, serial number 40103852) 16MB DRAM
Command line: mem=16m
Calibrating delay loop... 207.67 BogoMIPS
Memory: 15000k/16M available (996k code, 20k reserved, 364k data, 4k init)
Dentry hash table entries: 2048 (order 2, 16k)
Buffer cache hash table entries: 16384 (order 4, 64k)
Page cache hash table entries: 4096 (order 2, 16k)
POSIX conformance testing by UNIFIX
Linux NET4.0 for Linux 2.2
Based upon Swansea University Computer Society NET3.039
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
TCP: Hash tables configured (ehash 16384 bhash 16384)
IrDA (tm) Protocols for Linux-2.2 (Dag Brattli)
Starting kswapd v 1.5
SA1100 serial driver version 4.27 with no serial options enabled
ttyS00 at 0xf8010000 (irq = 15) is a SA1100 UART
ttyS01 at 0xf8050000 (irq = 17) is a SA1100 UART
ttyS02 at 0xf8030000 (irq = 16) is a SA1100 UART
Signature is 636f6972 'rioc'
Found custom animation at offset 0x9c388
Tuner: loopback=0, ID=-1
show_message("Hijack v505 by Mark Lord")
empeg display initialised.
empeg dsp audio initialised
empeg dsp mixer initialised
empeg dsp initialised
empeg audio-in initialised, CS4231A revision a0
empeg remote control/panel button initialised.
empeg usb initialised, PDIUSBD12 id 1012
empeg state support initialised 0089/88c1 (save to d0005d00).
empeg RDS driver initialised
empeg power-pic driver initialised (first boot)
RAM disk driver initialized: 16 RAM disks of 4096K size
empeg single channel IDE
Probing primary interface...
ide_data_test: wrote 0x0000 read 0x8080
ide_data_test: wrote 0xffff read 0x8080
ide_data_test: wrote 0xaaaa read 0x8080
ide_data_test: wrote 0x5555 read 0x8080
ide_data_test: wrote 0x0000 read 0x8080
ide_data_test: wrote 0xffff read 0x8080
ide_data_test: wrote 0xaaaa read 0x8080
ide_data_test: wrote 0x5555 read 0x8080
ide_data_test: wrote 0x0000 read 0x8080
ide_data_test: wrote 0xffff read 0x8080
ide_data_test: wrote 0xaaaa read 0x8080
ide_data_test: wrote 0x5555 read 0x8080
ide_data_test: wrote 0x0000 read 0x8080
ide_data_test: wrote 0xffff read 0x8080
ide_data_test: wrote 0xaaaa read 0x8080
ide_data_test: wrote 0x5555 read 0x8080
ide_data_test: wrote 0x0000 read 0xdbfb
ide_data_test: wrote 0xffff read 0xdbfb
ide_data_test: wrote 0xaaaa read 0xdbfb
ide_data_test: wrote 0x5555 read 0xdbfb
hda: WDC WD2500BEVE-00WZT0, ATA DISK drive
hdb: WDC WD2500BEVE-00WZT0, ATA DISK drive
ide0 at 0x000-0x007,0x038 on irq 6
hda: WDC WD2500BEVE-00WZT0, 238475MB w/8192kB Cache, CHS=30401/255/63, LBA48
hdb: WDC WD2500BEVE-00WZT0, 238475MB w/8192kB Cache, CHS=30401/255/63, LBA48
empeg-flash driver initialized
smc chip id/revision 0x3349
smc9194.c:v0.12 03/06/96 by Erik Stahlman (erik@vt.edu)

SMC9194: SMC91C94(r:9) at 0x4008000 IRQ:7 INTF:TP MEM:6144b MAC 00:02:d7:28:0f:0
c
Partition check:
hda: hda1 < hda5 hda6 > hda2 hda3 hda4
hdb: hdb1 < hdb5 hdb6 > hdb2 hdb3 hdb4
RAMDISK: ext2 filesystem found at block 0
RAMDISK: Loading 320 blocks [1 disk] into ram disk... done.
VFS: Mounted root (ext2 filesystem).
empeg-pump v0.03 (19980601)
Press Ctrl-A to enter pump...VFS: Mounted root (ext2 filesystem) readonly.
change_root: old root has d_count=1
Trying to unmount old root ... okay
Freeing unused kernel memory: 4k initempeg init 0.8
I see this is a developer image!
Mounting proc
Mounting first music partition
Tried to mount /dev/hda4 as reiserfs but got error 19
Mounting second music partition
Remounting first music partition read-only
Remounting second music partition read-only
Press 'q' now to go into development mode. You Have Zero Seconds To Comply...
Starting player
Timezone: MST7MDT
Hijack: intercepting config.ini

hijack: removed menu entry: "Hard Disk Detection"
hijack: removed menu entry: "Serial Port Assignment"
kftpd: listening on port 21
khttpd: listening on port 80
Using non-standard cache size 118 (bonus 0Mb, adjustment 16)
player.cpp : 385:empeg-car 2.01 2004/07/06.
! tags.cpp : 61:Failed to open tags (0xc0041002).
Dead temp.sensor, status=0x00
Prolux 4 empeg car - 2.1434 Jul 5 2004
Vcb: 0x4076d000


............ I performed a CTRL-C to Exit ................

Dead temp.sensor, status=0x00
Restored terminal settings
Remounting first music partition read-only
Remounting second music partition read-only
Abnormal player termination
Player received SIGINT, user interruption
Switching to shell-player loop
Starting bash.
empeg:/empeg/bin#


It really just looks like the tags got hosed. Any ideas where to go from here?

Ross

Ross, this was "before your time" so to speak... but do you remember when the original empeg came out with a 2 GB hard drive, unless you wanted to spend the extra money and get the massive 4 GB model? I still remember the discussions about whether there was any reason to get the bigger hard drive. Who would need four whole gigabytes of MP3s? When I finally got my first empeg, I knew I was set for life because by then it came with a truly gigantic 10 GB drive. By today's standards I am still a piker with just 160 GB.

Now here you are, running more than 100 times the capacity of the largest empeg that was available when emplode was conceived and first written. I guess it wouldn't be too surprising to discover that you may have finally managed to outstrip the limits of the program.

11 years later, who woulda thought...

tanstaafl.

Good question. If you'd really exceeded the maximum database size, the player wouldn't be able to rebuild its own database on power-on. The fact that that rebuild works, does give you a potential way to get the player back:

(The "rwm" command takes a while to run.) The player will then laboriously rebuild the database again, but this time should save it to disk afterwards and then be OK from then on.

Peter