I use v3a11 regularly (but not every day -- my car rarely moves from its spot outside the house, and I'm not using my mlord dock at the moment), and it reboots occasionally. I can live with it.

For what it's worth, I don't use the tuner (I have one, but I only use it as a Ford Focus stalk adapter), and I listen to MP3 files only. I also don't bother with the visuals.

I've updated the relevant FAQ entry a bit.

i have found one error... but the check is not complete

but is there a way to exclude the bad sectors from beeing used?

I think that with modern IDE drives, the act of exercising that sector will automatically mark it as bad, you don't need to do anything. Mark, correct me if I'm wrong?

Huh? If it hits any error, it will bail out completely.
So what makes you think it "found one error" ?

Log, please.

I believe that you have to write to that sector to cause it to be re-mapped.

it has not returned to the command line yet.. but it was telling something about input/output error

empeg:/empeg/bin# cat /dev/hd* > /dev/null
hdb: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
hdb: read_intr: error=0x40 { UncorrectableError }, LBAsect=96695, sector=96695
end_request: I/O error, dev 03:40 (hdb), sector 96695
hdb: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
hdb: read_intr: error=0x40 { UncorrectableError }, LBAsect=100349, sector=100349
end_request: I/O error, dev 03:40 (hdb), sector 100349
hdb: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
hdb: read_intr: error=0x40 { UncorrectableError }, LBAsect=115173, sector=115173
end_request: I/O error, dev 03:40 (hdb), sector 115173
cat: /dev/hdc: Input/output error
hdb: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
hdb: read_intr: error=0x40 { UncorrectableError }, LBAsect=65264558, sector=65055713
end_request: I/O error, dev 03:44 (hdb), sector 65055713
cat: /dev/hdc4: Input/output error
empeg:/empeg/bin#

i thought it could not hurt to format (build) the drives

i had to do it 2 times before it began the stress test, which i remeber means that the build went ok

during one of the builds ig got these errors on the secound drive:

Writing inode tables: done
Writing superblocks and filesystem accounting information: done

real 4m0.304s
user 0m2.120s
sys 0m14.420s
hdb: read_intr: status=0x59
hdb: read_intr: error=0x40
end_request: I/O error, dev 03:44 (hdb), sector 119816194
EXT2-fs error (device ide0(3,68)): read_block_bitmap: Cannot read block bitmap - block_group = 7313, block_bitmap = 59908097
EXT2-fs error (device ide0(3,68)): ext2_check_blocks_bitmap: Wrong free blocks count in super block, stored = 77942805, counted = 77934623

btw... will both drives be stress tested?

just did the build 3 times in a row with no problems, just waiting for cat /dev/hd* > /dev/null to tell if the errors are still there.. will take a very looooooong time :-)

Ahh.. that explains why you think it "keeps on going".

Just for fun, try typing echo cat /dev/hd* and you'll see what command you really issued there.

For future advice-givers, the correct syntax is:
cat /dev/hd? > /dev/null
Which will finish in half the time with no errors, or less in the presence of errors.

But regardless, in this case there is something wrong.
So now use this command to repair the drive:

cat /dev/zero > /dev/hdc

And then re-run the disk-builder stuff.

Cheers

big thanks...

btw.. you said that you could repair a broken handle?

do you ned a pic of it first ?

I can probably repair the broken handle, but won't know for sure until I try it.

A picture (here) might be fun.

well one pic will be here later

another question, was there ever made a guide for the auto eq thing? i guees that i have to use the mic input on the sled, and some spec. files....

but by guide i mean for dummies.. sometinh like this:

1. do this
2. then do this
...

This is about as good as it got. Worked for me, although personally I tweaked a little to my like first.
http://empegbbs.com/ubbthreads.php?ubb=showflat&Board=1&Number=231894

There's also this manual way:
http://empegwiki.org/index.php?title=Empeg_Sound_-_parametetric_eq

good guide.... but where do i get the auto-eq files? and what order do i have to set them?

i may be blind but did not sse that

I think the files are these ones:

http://persiconi.com/wrx/downloads/empegsound/soundfiles/parametric/

thanks for the files


how long should cat /dev/zero > /dev/hdc take for a 80 gb drive?

Most likely on the order of 80 * 1000 * seconds, or from 10-22 hours.

-ml

after some 11-12 hours i got this:

empeg:/empeg/bin# cat /dev/hd? > /dev/null
hdb: read_intr: status=0x59
hdb: read_intr: error=0x40
show_message("36e0.-1 hdb: err: read_intr")
end_request: I/O error, dev 03:40 (hdb), sector 115464112
cat: /dev/hdc: Input/output error
empeg:/empeg/bin#

keep in mind that i did the repair as mark told me to do...

Mmm.. "read_intr: status=0x59 error=0x40" really means "bad sector". That shouldn't happen here if you successfully did the cat /dev/zero > /dev/hdc command beforehand.

So repeat both steps again. If it still fails, something else (connectors) may be bad.

Or get the smartctl binary for the empeg and run it -- it will read out the error logs from the drive to tell us what *really* happened.

Cheers

while i do another cat /dev/zero > /dev/hdc

i wondered.. the auto.eq thing.. is that only in ver 3?

Yes

i guees that doing cat /dev/zero > /dev/hdc

it's normal to get no space left on drive error

By doing this you will remove everything. You are zeroing every single sector on the drive. So all data will be gone. It will also erase any partitions set up previously.

i know...

is there a way to do it for both drives? some of the partions ar mounted read only

To do it for both drives, just use /dev/hda and /dev/hdc for the second parameter.

It doesn't care that the partitions are mounted read-only, because it writes over the entire drive, including the partition table.

What it does is cause the disk to remap the bad sectors (the drive will only remap when writing).

Running out of space is expected: /dev/zero is an infinite-length stream of zeroes, and your disk has finite space.

(slightly drunk; being unnecessarily concise)