[no subject]
subscribe To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
Re: Device pcm causes hard lock
Hi, i have the similar problems with a Sony FX601. (ACPI only BIOS). After upgrading to FreeBSD 5.0-current it worked bye, Rob Hughes wrote: I decided I wanted sound on my Toshiba 5005, so I added device pcm to the kernel config. Now, at boot, I receive the message pcm0: Intel 82801BA (ICH2) at device 31.5 on pci0 pcm0: unable to map IO port space device_probe_and_attach: pcm0 returned 6 Very quickly after that, I get a hard lock, no break to debugger, nothing. Ideas? Thanks Rob -- -- -- E-mail: [EMAIL PROTECTED] JAWA Management Software GmbH Liebenauer Hauptstr. 200 A-8041 GRAZ Tel: ++43-(0)316-403274-12 Fax: ++43-(0)316-403274-10 GSM: ++43-(0)676-93 96 698 homepage: http://www.jawa.at - privat --- E-mail: [EMAIL PROTECTED] homepage: http://www.inode.at/mbretter -- To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
Changes to tar (was Re: strange SSH / tar problem)
Actually, someone pointed out this is a tar problem. I forgot that RELENG_4 has a whole new version of tar. tar (GNU tar) 1.13.25 vs 1.11.2. Old version of tar both forms work fine. granite# tar -czf t.tgz /tmp/root granite# tar -czf - /tmp/root t2.tgz granite# tar -tzf t2.tgz root/ root/ps.out root/m.sort root/q.out root/nestat.out root/m.list root/s.dead root/s1 root/s2 root/exp.list root/dead.txt root/dsl.out root/code-red.txt root/mail.access.times root/own.txt root/ms.rot root/dead.sort root/files.txt root/mail.list root/f.out root/m.dort root/du.out root/list.txt root/list.txt2 root/dead.list root/deadmail/ granite# ls -l *.tgz -rw-r--r-- 1 root wheel 1424563 Jul 27 09:44 t.tgz -rw-r--r-- 1 root wheel 1424563 Jul 27 09:45 t2.tgz granite# new version of tar news% tar -czf - test-files t2.tgz news% tar -czf t.tgz test-files news% tar -tzf t2.tgz test-files/ test-files/ps.out test-files/netstat.out test-files/mount.out test-files/ipfw.out test-files/if.out test-files/t.zip gzip: stdin: decompression OK, trailing garbage ignored tar: Child returned status 2 tar: Error exit delayed from previous errors news% ls -l *.tgz -rw-r--r-- 1 mdtancsa wheel 4320 Jul 27 09:46 t.tgz -rw-r--r-- 1 mdtancsa wheel 10240 Jul 27 09:46 t2.tgz news% At 04:13 PM 7/26/2002 -0400, Mike Tancsa wrote: On a number of my servers I use tar via ssh to do frequent backups of files to a central backup server. The backup server is running an older ssh, but the same problem happens with the most recent versions. SSH-1.99-OpenSSH_2.9 FreeBSD localisations 20020307 If I do a tar over ssh the resulting file has an error at the end. e.g. source is running SSH-1.99-OpenSSH_3.4p1 FreeBSD-20020702 destination (20020307) tar -cpzf - /etc /usr/local/etc | /usr/bin/ssh [EMAIL PROTECTED] cat - /path-to-files/userid/backup.`date +%y%m%d`.tgz When I go to test the file, I get tar -tzf backup.020726.tgz usr/local/etc/apache.bak/httpd.conf.default gzip: stdin: decompression OK, trailing garbage ignored usr/local/etc/apache.bak/access.conf.default usr/local/etc/apache.bak/access.conf tar: child returned status 2 It seems to work OK, but I was a bit surprised about the warning message From the source server md5 * MD5 (t.zip) = 868cd7e432dc2ad7049a2c61a2403194 MD5 (t2.zip) = 90279347e332762c0fc43798b4ae4d57 MD5 (webalizer.conf) = dd948a77ba729a471f7ade18088b3680 tar -cpzf - /tmp/test-files | /usr/bin/ssh [EMAIL PROTECTED] cat - /path-to-files/userid/testb.`date +%y%m%d`.tgz and the target server tar -xzf testb.020726.tgz gzip: stdin: decompression OK, trailing garbage ignored tar: child returned status 2 cd test-files/ backupserver# md5 * MD5 (t.zip) = 868cd7e432dc2ad7049a2c61a2403194 MD5 (t2.zip) = 90279347e332762c0fc43798b4ae4d57 MD5 (webalizer.conf) = dd948a77ba729a471f7ade18088b3680 The sshd_config and ssh_config on both machines are stock. If I got from a stable box as of today to another stable box as of today, the same error happens. ---Mike Mike Tancsa, tel +1 519 651 3400 Sentex Communications,[EMAIL PROTECTED] Providing Internet since 1994www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message Mike Tancsa, tel +1 519 651 3400 Sentex Communications,[EMAIL PROTECTED] Providing Internet since 1994www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
Re: Device pcm causes hard lock
On Sat, 2002-07-27 at 02:41, Michael Bretterklieber wrote: Hi, i have the similar problems with a Sony FX601. (ACPI only BIOS). After upgrading to FreeBSD 5.0-current it worked bye, Rob Hughes wrote: I decided I wanted sound on my Toshiba 5005, so I added device pcm to the kernel config. Now, at boot, I receive the message pcm0: Intel 82801BA (ICH2) at device 31.5 on pci0 pcm0: unable to map IO port space device_probe_and_attach: pcm0 returned 6 Very quickly after that, I get a hard lock, no break to debugger, nothing. Ideas? Thanks Rob -- If only it were that simple for me... -- Remember: the only difference between being the champ and the chump is u. signature.asc Description: This is a digitally signed message part
sys/boot
Hi, I'm running -stable and I encounter the following problem : # make buildworld ... === sys/boot/buildworld cd: can't cd to /disk2/freebsd/stable/src/sys/boot/buildworld *** Error code 2 let's try : # cd /sys/boot/buildworld # SUBDIR=buildworld make -V SUBDIR buildworld ficl i386 well, src/sys/boot/Makefile only have SUBDIR+= w/o any SUBDIR= before, this is a problem when SUBDIR exists in the environment. however, I'm surprising I don't have this problem before and I'll investigate where this occur since I don't set SUBDIR anywhere ? in the mean time, could you fix that or do you prefer I submit a PR ? thanks, Index: Makefile === RCS file: /home/ncvs/src/sys/boot/Makefile,v retrieving revision 1.6.2.1 diff -u -r1.6.2.1 Makefile --- Makefile17 Jul 2002 14:23:14 - 1.6.2.1 +++ Makefile27 Jul 2002 14:43:50 - @@ -1,5 +1,7 @@ # $FreeBSD: src/sys/boot/Makefile,v 1.6.2.1 2002/07/17 14:23:14 ru Exp $ +SUBDIR= + # Build the add-in FORTH interpreter SUBDIR+= ficl CC -stable Cyrille. -- Cyrille Lefevre mailto:[EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
buildworld objdir moved! (was Re: sys/boot)
On Sat, Jul 27, 2002 at 04:50:30PM +0200, Cyrille Lefevre wrote:
I'm running -stable and I encounter the following problem :
# make buildworld
...
=== sys/boot/buildworld
cd: can't cd to /disk2/freebsd/stable/src/sys/boot/buildworld
*** Error code 2
found it! this is due by recent changes in src/Makefile.inc1.
TARGET is set using ?= and this is a problem when this variable
already exists in the environment as I do through the Makefile
I use to fetch/update/build/install/configure everything.
here is the log before w/ the following patch :
--
Rebuilding the temporary build tree
--
# XXX - These two can depend on any header file.
rm -f /usr/obj/buildworld/disk2/freebsd/stable/src/usr.bin/kdump/ioctl.c
rm -f /usr/obj/buildworld/disk2/freebsd/stable/src/usr.bin/truss/ioctl.c
^^ oops!
and the log w/ it :
--
Rebuilding the temporary build tree
--
# XXX - These two can depend on any header file.
rm -f /usr/obj/disk2/freebsd/stable/src/usr.bin/kdump/ioctl.c
rm -f /usr/obj/disk2/freebsd/stable/src/usr.bin/truss/ioctl.c
Index: /usr/src/Makefile.inc1
===
RCS file: /home/ncvs/src/Makefile.inc1,v
retrieving revision 1.141.2.54
diff -u -r1.141.2.54 Makefile.inc1
--- /usr/src/Makefile.inc1 25 Jul 2002 09:33:13 - 1.141.2.54
+++ /usr/src/Makefile.inc1 27 Jul 2002 15:04:17 -
@@ -126,9 +126,12 @@
MAKEOBJDIRPREFIX?= /usr/obj
TARGET_ARCH?= ${MACHINE_ARCH}
.if ${TARGET_ARCH} == ${MACHINE_ARCH}
-TARGET?= ${MACHINE}
+TARGET=${MACHINE}
.else
-TARGET?= ${TARGET_ARCH}
+TARGET=${TARGET_ARCH}
+.endif
+.if defined(TARGET_OVERRIDE)
+TARGET=${TARGET_OVERRIDE}
.endif
.if make(buildworld)
BUILD_ARCH!= sysctl -n hw.machine_arch
PS : the previous patch submitted still apply.
Cyrille.
--
Cyrille Lefevre mailto:[EMAIL PROTECTED]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-stable in the body of the message
Re: Cannot initialize disklabel
My MB is modern (PIII 450MHz - GigaByte Intel 440BX of one and a half year ago). The disk works fine on many linuxes (Mandrake 8.1, Red Hat 7.3 and Debian 2.2r5) and on Windows 2000 on the same system. Another 10GB Maxtor disk with 19841 cylinders has the core FreeBSD system on it and it works well. The problem relies into the BSD label for the main disk (i.e. /dev/ad2) I've edited this label from linux fdisk and set it up correctly. Then I reboot FreeBSD and the command disklabel /dev/ad2 shows 4092 cylinders instead if 39693. If I execute disklabel -r /dev/ad2 the cylinder count is ok, meaning the disk copy of the label is correct. The kernel in-core BSD label is always incorrect and, as FreeBSD fdisk relies on it I can't create a partition greather tha 2 GB (4092 cylinders) Worse, if I set up the partition from linux fdisk and the correct BSD label, when I boot FreeBSD the in-core BSD label is still 4092 cylinders and the ata driver complaints about having a partition too big. If try to edit the BSD label the ata driver block the writes. I don't understand why from BSD there's no way to force a new BSD label. Bye Gino --- Cyrille Lefevre [EMAIL PROTECTED] wrote: On Fri, Jul 26, 2002 at 03:07:06PM -0700, Andrea Monaldi wrote: I have a 4.6 box and I added a new HD (Maxtor 20GB). It is always listed as 2G instead of 20 and I am not able to initialize the partition or the disk label to something different to 2 GB. The number of cylinder is always 4092 although the disk has more than 3. how old is your machine ? personally, since my BIOS is somewhat old, I have to use `MaxBlast' to deal w/ disks bigger than 2 GB (or 8 GB, don't remember). does your first driver `MaxBlast'ed ? if so, your second drive have to be `MaxBlast'ed too. see the following URL for details and read carefully the installation procedure, etc. http://www.maxtor.com/en/support/downloads/maxblast_plus_ii.htm The disk in the BIOS is not set to LBA, but it is set as NORMAL. I have another Maxtor disk that is running fine. why don't want to use LBA mode ? what are the caracteristics of the other drive ? # grep ata /var/run/dmesg.boot How can I create a disklabel ? before to create a disklabel, what say fdisk ? # fdisk ad0 # fdisk ad1 Cyrille. -- Cyrille Lefevre mailto:[EMAIL PROTECTED] __ Do You Yahoo!? Yahoo! Health - Feel better, live better http://health.yahoo.com To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
Re: Changes to tar (was Re: strange SSH / tar problem)
Hey Mike, Just as a POI, I think you'd get better results with 'rsync' for this application (from the people that brought you Samba, readily available via ports) - it'll do intelligent copying, when archiving can maintain a directory of the previous versions for files that change, uses SSH as the transport, can be bandwith-limited and will compress the stream in-line. Course, it'd still be good to fix any tar issues :) Cheers, AS Actually, someone pointed out this is a tar problem. I forgot that RELENG_4 has a whole new version of tar. tar (GNU tar) 1.13.25 vs 1.11.2. Old version of tar both forms work fine. granite# tar -czf t.tgz /tmp/root granite# tar -czf - /tmp/root t2.tgz granite# tar -tzf t2.tgz root/ root/ps.out root/m.sort root/q.out root/nestat.out root/m.list root/s.dead root/s1 root/s2 root/exp.list root/dead.txt root/dsl.out root/code-red.txt root/mail.access.times root/own.txt root/ms.rot root/dead.sort root/files.txt root/mail.list root/f.out root/m.dort root/du.out root/list.txt root/list.txt2 root/dead.list root/deadmail/ granite# ls -l *.tgz -rw-r--r-- 1 root wheel 1424563 Jul 27 09:44 t.tgz -rw-r--r-- 1 root wheel 1424563 Jul 27 09:45 t2.tgz granite# new version of tar news% tar -czf - test-files t2.tgz news% tar -czf t.tgz test-files news% tar -tzf t2.tgz test-files/ test-files/ps.out test-files/netstat.out test-files/mount.out test-files/ipfw.out test-files/if.out test-files/t.zip gzip: stdin: decompression OK, trailing garbage ignored tar: Child returned status 2 tar: Error exit delayed from previous errors news% ls -l *.tgz -rw-r--r-- 1 mdtancsa wheel 4320 Jul 27 09:46 t.tgz -rw-r--r-- 1 mdtancsa wheel 10240 Jul 27 09:46 t2.tgz news% At 04:13 PM 7/26/2002 -0400, Mike Tancsa wrote: On a number of my servers I use tar via ssh to do frequent backups of files to a central backup server. The backup server is running an older ssh, but the same problem happens with the most recent versions. SSH-1.99-OpenSSH_2.9 FreeBSD localisations 20020307 If I do a tar over ssh the resulting file has an error at the end. e.g. source is running SSH-1.99-OpenSSH_3.4p1 FreeBSD-20020702 destination (20020307) tar -cpzf - /etc /usr/local/etc | /usr/bin/ssh [EMAIL PROTECTED] cat - /path-to-files/userid/backup.`date +%y%m%d`.tgz When I go to test the file, I get tar -tzf backup.020726.tgz usr/local/etc/apache.bak/httpd.conf.default gzip: stdin: decompression OK, trailing garbage ignored usr/local/etc/apache.bak/access.conf.default usr/local/etc/apache.bak/access.conf tar: child returned status 2 It seems to work OK, but I was a bit surprised about the warning message From the source server md5 * MD5 (t.zip) = 868cd7e432dc2ad7049a2c61a2403194 MD5 (t2.zip) = 90279347e332762c0fc43798b4ae4d57 MD5 (webalizer.conf) = dd948a77ba729a471f7ade18088b3680 tar -cpzf - /tmp/test-files | /usr/bin/ssh [EMAIL PROTECTED] cat - /path-to-files/userid/testb.`date +%y%m%d`.tgz and the target server tar -xzf testb.020726.tgz gzip: stdin: decompression OK, trailing garbage ignored tar: child returned status 2 cd test-files/ backupserver# md5 * MD5 (t.zip) = 868cd7e432dc2ad7049a2c61a2403194 MD5 (t2.zip) = 90279347e332762c0fc43798b4ae4d57 MD5 (webalizer.conf) = dd948a77ba729a471f7ade18088b3680 The sshd_config and ssh_config on both machines are stock. If I got from a stable box as of today to another stable box as of today, the same error happens. ---Mike Mike Tancsa, tel +1 519 651 3400 Sentex Communications,[EMAIL PROTECTED] Providing Internet since 1994www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message Mike Tancsa,tel +1 519 651 3400 Sentex Communications, [EMAIL PROTECTED] Providing Internet since 1994www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message msg48025/pgp0.pgp Description: PGP signature
Re: Cannot initialize disklabel
On Fri 2002-07-26 (15:07), Andrea Monaldi wrote: It is always listed as 2G instead of 20 and I am not able to initialize the partition or the disk label to something different to 2 GB. The number of cylinder is always 4092 although the disk has more than 3. It sounds like your drive has the Capacity Limitation Jumper set. All the 2GB Maxtors I've seen have a jumper which reduces the reported capacity to 4092 cylinders so old BIOSes can detect them. Look up your drive at http://www.maxtor.com/, download the jumper settings, and make sure that that jumper is not set on your drive. -- David Siebörger [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
Re: Cannot initialize disklabel
David, If I had this it should not work well under Linux and W2K too, isn't it? Anyway it's not this problem. It works bad ONLY in FreeBSD. BTW this HD was taken from an Apple G4 and has a little Apple on the top of the drive. Does it mean it has a different firmware? Why Linux and W2K works perfectly with it? Does it make any sense if this drive has been partitioned by OS X previously, doing something strange, and preventing any usage with another BSD OS? Bye Gino --- David Siebörger [EMAIL PROTECTED] wrote: On Fri 2002-07-26 (15:07), Andrea Monaldi wrote: It is always listed as 2G instead of 20 and I am not able to initialize the partition or the disk label to something different to 2 GB. The number of cylinder is always 4092 although the disk has more than 3. It sounds like your drive has the Capacity Limitation Jumper set. All the 2GB Maxtors I've seen have a jumper which reduces the reported capacity to 4092 cylinders so old BIOSes can detect them. Look up your drive at http://www.maxtor.com/, download the jumper settings, and make sure that that jumper is not set on your drive. -- David Siebörger [EMAIL PROTECTED] __ Do You Yahoo!? Yahoo! Health - Feel better, live better http://health.yahoo.com To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
ATA CDROM Problems 4.6 (and 4.6.1)
I just upgraded from 4.5 (off the CD Release) to 4.6 (off the CD release) and yadda yadda yadda ATA CDROM fails per the Errata (big read timeout, etc). The workaround DOESN'T /home/dob: cat /boot/loader.conf # -- sysinstall generated deltas -- # userconfig_script_load=YES hw.ata.ata_dma=1 hw.ata.atapi_dma=1 /var/log: gzcat messages.0.gz | grep -i timeout Jul 21 19:28:55 yorky /kernel: acd0: READ_BIG command timeout - resetting Jul 21 19:29:25 yorky /kernel: acd0: READ_BIG command timeout - resetting Jul 21 19:29:55 yorky /kernel: acd0: READ_BIG command timeout - resetting Jul 21 19:30:26 yorky /kernel: acd0: READ_BIG command timeout - resetting I even tried a 4.6.1 GENERIC kernel a friend prepared and it fails. /var/log: gzcat messages.0.gz | grep -i timeout|grep GENERIC Jul 25 21:48:43 yorky /kernel.GENERIC: acd0: READ_BIG command timeout - resetting Jul 25 21:49:13 yorky /kernel.GENERIC: acd0: READ_BIG command timeout - resetting I'm now stuck without a way to upgrade to 4.7 (off the CD's) (not really, I kept the old 4.5 kernel and if I boot.old/kernel.old, the CDROM works perfectly). I always do a source buildworld, installworld, mergemaster, build/install kernel yadda yadda. I'm willing to play Guinea Pig for any and all fixes. I've got a generic 50X CDROM: dmesg | grep acd acd0: CDROM CD-ROM 50X at ata1-master PIO4 It mounts, I can do ls, but cat ports.tgz /dev/null results in the big read stuff. I've had one panic. Now what? I've seen others in the same boat. Thanks in advance, = Dan O'Brien, [EMAIL PROTECTED] Cell: 614-783-4859 Work: 614-476-8473 Home: 740-927-2178 Pataskala, OH USA '02 Suzuki DL1000 #00082 Grey __ Do You Yahoo!? Yahoo! Health - Feel better, live better http://health.yahoo.com To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
Re: Cannot initialize disklabel
On Sat 2002-07-27 (09:26), Andrea Monaldi wrote: If I had this it should not work well under Linux and W2K too, isn't it? Maybe; maybe not. Linux, Windows 2000 and FreeBSD all have very different IDE drivers. Anyway it's not this problem. Have you actually checked the jumper? Make sure. Perhaps I'm wrong - maybe it is something else - but the 4092 cylinders is just too much of a co-incidence to rule it out. It works bad ONLY in FreeBSD. BTW this HD was taken from an Apple G4 and has a little Apple on the top of the drive. Does it mean it has a different firmware? Why Linux and W2K works perfectly with it? I wouldn't know. Does it make any sense if this drive has been partitioned by OS X previously, doing something strange, and preventing any usage with another BSD OS? The cylinder/head/sector numbers in the dmesg output, for example, ad0: 16479MB Maxtor 91728D8 [33483/16/63] at ata0-master using UDMA33 ^^^ are read from the drive without reference to the partition table. So if you're seeing 4092 cylinders there, I doubt OS X is the problem. -- David Siebörger [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
Re: Changes to tar (was Re: strange SSH / tar problem)
On Sat, Jul 27, 2002 at 09:48:31AM -0400, Mike Tancsa wrote: Actually, someone pointed out this is a tar problem. I forgot that RELENG_4 has a whole new version of tar. tar (GNU tar) 1.13.25 vs 1.11.2. [snip] gzip: stdin: decompression OK, trailing garbage ignored tar: Child returned status 2 tar: Error exit delayed from previous errors [snip] this is a long standing known problem in gzip 1.2 which has been solved in gzip 1.3. I'm currently making a port of gzip 1.3.3, so, you can use --use-compress-program=/usr/local/bin/gzip until gzip 1.3.3 is imported into -current or -stable. see the following URLs for details : http://www.gzip.org/#faq8 and http://www.gnu.org/manual/tar/html_node/tar_132.html#SEC127 (see details on the -b options) or info tar.info 'Blocking Factor' (see details on the -b options) Cyrille. -- Cyrille Lefevre mailto:[EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message
