I've had a problem with all v3-alphas with them forgetting settings across power cycles. Sometimes shuffle, sometimes repeat, sometimes the current running order. I've assumed this was a v3 glitch but I guess my player may have an individual problem with its powerfail interrupt.

Rob

I've rebuilt my hds and am starting from scratch with my spare empeg running the new alpha 11. I've tried adding music 4 times now and after about 85-100 songs Emplode bombs with errors saying the connection was forcably closed by the remote host. On the front of the Empeg it has an error saying something about the hds timing out. I'm assuming Hijack is displaying the error. I got the same error while trying to delete the music that was on the Player. I'm running Emplode 2.0 as I read somewhere on here that v2.1 was too buggy. Is this a bug, or some setting that is incorrect? The hds check out fine and the stress test during the build was OK. I'm going to try going back to 2.0 release version and loading up the music with that.

JeffreyB

OK,

This is not a bug in alpha 11. It does it on release 2.0 as well. I get alternating "hda: disk error: status timeout" and "hdb: disk error: status timeout" errors.

The log shows:

ide0: reset timed-out, status=0x80
hda: status timeout: status=0x80
end_request: I/O error, dev 03:04 (hda), sector 147686
hda: drive not ready for command
hdb: status timeout: status=0x80
end_request: I/O error, dev 03:44 (hdb), sector 147694
hdb: drive not ready for command
hda: status timeout: status=0x80
hda: drive not ready for command
ide0: reset timed-out, status=0x80
hdb: status timeout: status=0x80
hdb: drive not ready for command
ide0: reset timed-out, status=0x80
hda: status timeout: status=0x80
hda: drive not ready for command

I'm assuming it's a HD or drive cable issue. Thing is, the player was working fine until I rebuilt the drives. I have a new extended HD cable and the IDE header was reflowed 2 years ago. Also the stress test during the rebuild was OK...

JeffreyB

For info: the only times I have seen this message have both been sorted by ensuring good IDE cable connection (by pressing it down hard again

One was not long after an upgrade, but the other appeared to be unconnected to anything, so my assumption is general day to day bumping wiggled the connector slightly loose.

I reseated the IDE cable. All is fine now. I wonder why the stress test didn't show any problems...

Thanks!

JeffreyB

Someone asked this earlier, but I didn't really see a response. How do I use Mark's ethernet upgrader from Windows?

Check out this post, it's on page #2.

Ahh ok.. I read that but when I started seeing chmod and bash, I figured that was a linux tool. I'll give it a try.

they are linux tools, but the cygwin distro creates a linux shell inside of windows even with it's own command line interface window.

OK.. well I got the distro and installed everything. I downloaded the upgrader from Mark's site, but I can't find an upgrader-cygwin.exe anywhere. The file on Mark's site is upgrader.tar and when I open that, all I see is 'upgrader'. What do i do at that point?

Quote:
the cygwin distro creates a linux shell inside of windows

\begin{nitpick}

ok, it doesn't really create a linux shell. Cygwin is a bunch of unix tools ported to windows... many of which also happen to exist happily (and natively) under linux, solaris, HP/UX, AIX, BSD, etc.

\end{nitpick}

-- Gary F.

Well I ran it once and it said BROKEN PIPE at the end - not sure if that means it was successful or not. I accidentally closed it, but i ran it again and here's the paste of the log:

Oh and I'm currently at a7 and the player is in standby when I ran this.

$ ./upgrader_0.8_cygwin.exe 192.168.1.28 car2-developer-v3.00-alpha11.upgrade

./upgrader_0.8_cygwin: Empeg/RioCar Ethernet Upgrade Tool, version 0.8, (c)2003
by Mark Lord
--> requires Hijack kernel on player
--> requires gunzip and ncftpput in $PATH

UPGRADE Requires upgrader version: 2
HWREV Supported hardware versions: 7,9
Placing player into Standy-By for the upgrade: ncftpput -E -c -Y "site button=t
op.L" "192.168.1.28" "/dev/null" </dev/null

INFO This release is suitable for Mark II empeg and Rio Car players (with r
otary control)
WHAT empegcar developer image
2005/07/25 18:54
RELEASE Rio Car player software - developer release v3.00-alpha11
VERSION Rio Car player software - developer release v3.00-alpha11
LOADER /proc/flash_0c000, size=452 bytes
--> pumping: ncftpput -E -c "192.168.1.28" "/proc/flash_0c000"
LOADER /proc/flash_0e000, size=1692 bytes
--> pumping: ncftpput -E -c "192.168.1.28" "/proc/flash_0e000"
KERNEL /proc/empeg_kernel, size=476784 bytes
--> SKIPPED
RAMDISK /proc/flash_b0000, size=327680 bytes
--> pumping: ncftpput -E -c "192.168.1.28" "/proc/flash_b0000"
Broken pipe

failed. dunno why.

I put 3.11 on the empeg today and the playlists were totally screwed up! Crazyness. There wasn't really any rhyme or reason to it either, it was like it was totally random and many songs/playlists weren't even there any more. I'm going back to 2.0 now to see if that fixes things.

Weird. Anyone else experience this issue?

Yes, with earlier alphas. Haven't tried 11 yet.

Took me a couple of synchs to get there... First load of the alpha worked fine, it was only after a couple of synchs that my playlists got scrambled.

yes, I think I had v3 alpha3 on there long ago. I went back to 2.00 and all was back to normal. I'll try deleting the database files and see how that goes.

Actually, can I just synch in emplode and see if it fixes the problem? The playlists were all messed up in emplode too, so I was afraid that if I synched then they would be permanently messed up.

thanks!

I deleted the database files, put on alpha11 and all is well.

Thanks for your help Peter!

Quote:
People experiencing crashes and glitches with alpha11 should probably try increasing their ReserveCache

Well, either because of the above suggestion or just through statistical anomalies, Alpha 11 has been doing much better lately. Only had one crash in the last couple of weeks (a segfault, first time I ever saw that message), and very few cases of odd behaviour besides long pauses when I switch from radio back to player mode, so all-in-all it's behaving itself nicely.