Re: [expert] persistent cdrecord failure
Alex ZIJDENBOS wrote: > > Since I upgraded my box from LM 7.0 to LM 8.0, I have had nothing but > trouble writing CDs (from 100% success to about 10% success - I'm now > drowning in coasters). See below for a sample cdrecord output. I'm > currently using cdreecord from the cdrecord-1.10-1mdk rpm, but I've > also tried other versions - with similar results. I have a Sony CRX > 160E CD/RW hooked up using ide-scsi. Did you not upgrade through 7.2? If so, did that work? The best advice I have is to remap your IDE devices across the two channels in such a fashion that while writing CDs the cdwriter is the only active device on one channel. The worst case would be that the source iso file and the cdwriter are on the same IDE channel. The difficulty appears to be caused by motherboard chipset deficiences, most notably with the VIA chipset. If you have ATA-100 drives, set them up on the two extra IDE channels provided by the Promise, etc, ATA-100 PCI card (or the Promise motherboard chip). This leaves the original secondary IDE motherboard channel available to be dedicated to the cdwriter (ATA 33 is plenty). -- Ron. [au]
Re: [expert] persistent cdrecord failure
Alex ZIJDENBOS wrote: > > Since I upgraded my box from LM 7.0 to LM 8.0, I have had nothing but > trouble writing CDs (from 100% success to about 10% success - I'm now > drowning in coasters). See below for a sample cdrecord output. I'm > currently using cdreecord from the cdrecord-1.10-1mdk rpm, but I've > also tried other versions - with similar results. I have a Sony CRX > 160E CD/RW hooked up using ide-scsi. > > I found a few hints in the expert archives (upgrade mandrake_desk, > modify XF86Config-4, dma enabled on IDE channel), but none of these > work for me. > > If anybody has any advice, solutions, or things to try, please let me > know! > > Regards, > > -- Alex > > -- > Blocks total: 359849 Blocks current: 359849 Blocks remaining: 89943 > Starting to write CD/DVD at speed 4 in dummy mode for single session. > Last chance to quit, starting dummy write in 0 seconds. Operation starts. > Waiting for reader process to fill input buffer ... input buffer ready. > Starting new track at sector: 0 > Track 01: 31 of 527 MB written (fifo 100%). > /usr/bin/cdrecord: Input/output error. write_g1: scsi sendcmd: no error > CDB: 2A 00 00 00 3E 1F 00 00 1F 00 status: 0x2 (CHECK CONDITION) > Sense Bytes: 70 00 04 00 00 00 00 12 00 00 00 00 08 03 00 00 > Sense Key: 0x4 Hardware Error, Segment 0 > Sense Code: 0x08 Qual 0x03 (logical unit communication crc error (ultra-dma/32)) Fru >0x0 > Sense flags: Blk 0 (not valid) > cmd finished after 0.105s timeout 40s > write track data: error after 32569344 bytes > Sense Bytes: 70 00 00 00 00 00 00 12 00 00 00 00 00 00 00 00 00 00 > Writing time: 55.086s > Fixating... WARNING: Some drives don't like fixation in dummy mode. > Fixating time:0.010s > /usr/bin/cdrecord: fifo had 577 puts and 514 gets. > /usr/bin/cdrecord: fifo was 0 times empty and 419 times full, min fill was 73%. > -- Since this is an upgrade from 7.0 to 8.0 check and see if you possibly have some extra statements in /etc/modules.conf file. -- alias block-major-11 scsi_hostadapter <- *** pre-install plip modprobe parport_pc ; echo 7 > /proc/parport/0/irq alias scsi_hostadapter ide-scsi --- For some reason the line, left after the upgrade, "alias block-major-11 scsi_hostadapter" caused a lot of problems. On a clean install of LM 8.0 this line never appeared in the modules.conf file and things work fine. .. This is from a clean 8.0 install, works fine. ** notice there is only one reference to ide-scsi *** pre-install pcmcia_core CARDMGR_OPTS=-f /etc/rc.d/init.d/pcmcia start alias usb-interface usb-uhci alias sound-slot-0 emu10k1 alias scsi_hostadapter ide-scsi alias eth0 8139too -- I did work on one machine for about 2 weeks with cd-writer acting up such as you have ... finally discovered it was a drive gone bad. It just did it at the time of the upgrade. Larry -- Sword'sEdge VoiceMail/Fax: (858) 860-6406 x1587
Re: [expert] persistent cdrecord failure -> "fixed"
Thanks - I know that pure scsi burners are more reliable this way, but I have been using ide-scsi for many years now without any problem, on many distros - until I upgraded to LM 8.0. However, I experimented some more with hdparm, and found that the problem goes away when I turn DMA *off* on /dev/hdc (hdparm -d0 /dev/hdc). It's highly reproducible; as soon as I turn using_dma on again (which is the default config) it fails 9 out of 10 times. I doubt this is the way things are supposed to work, but if it shuts down my coaster factory, I can live with it :-) -- Alex On Fri, Aug 03, 2001 at 11:29:43PM -0500, J. C. Woods wrote: | | Alex, | | Since you ask, and normally this is the kind of advise I am disinclined | to give, my advise would be to think about a pure scsi cd writer. There | are several good and cheap scsi boards out there, such as the Adaptec | 29xx series. Mandrake uses the "aic7xxx" driver, and has yet to fell me | in verision 7.0 thru 8.0. I know money can be an issue but I have never | had any problems burning a cd with my scsi setup in all the Linux | distros I have used, and there have been a few. If the frustration gets | to much with your ide cd writer, at least this is something your might | want to think about. Wish I had more for your situation. Hang in | there. | | drjung | -- | | Art is the illusion of spontaneity...
Re: [expert] persistent cdrecord failure -> "fixed"
On Saturday 04 August 2001 11:43, Alex ZIJDENBOS wrote: > However, I experimented some more with hdparm, and found that the > problem goes away when I turn DMA *off* on /dev/hdc (hdparm -d0 > /dev/hdc). It's highly reproducible; as soon as I turn using_dma on > again (which is the default config) it fails 9 out of 10 times. > > I doubt this is the way things are supposed to work, but if it shuts > down my coaster factory, I can live with it :-) > > -- Alex > I think this is in the documentation (a howto?), so this is precisely how it is supposed to work. mg
Re: [expert] persistent cdrecord failure -> "fixed"
On Sat, Aug 04, 2001 at 02:46:56PM -0400, mike wrote: | On Saturday 04 August 2001 11:43, Alex ZIJDENBOS wrote: | > However, I experimented some more with hdparm, and found that the | > problem goes away when I turn DMA *off* on /dev/hdc (hdparm -d0 | > /dev/hdc). It's highly reproducible; as soon as I turn using_dma on | > again (which is the default config) it fails 9 out of 10 times. | > | > I doubt this is the way things are supposed to work, but if it shuts | > down my coaster factory, I can live with it :-) | > | > -- Alex | | I think this is in the documentation (a howto?), so this is precisely how it | is supposed to work. I haven't been able to find any documentation to this effect - do you recall exactly where you saw this? the only thing I found was a response from Civileme to another (similar) question on this list, basically saying the reverse, i.e., that you need DMA on the burner to successfully burn CDs. -- Alex
