RE: fedora-list Digest, Vol 43, Issue 44
by McGuffey, David C.
>
> > Bought a new 320 GB EIDE disk and installed it as MASTER and moved
the
> > previous MASTER (a 160 GB EIDE disk that booted WinXP) to SLAVE.
That
> > made the old hd0 hd1 and the new disk hd0. Without any OS installed
on
> > the new hd0, the machine continued to boot WinXP from hd1.
>
> Are you sure of this above? No change at all, and swap from master to
> slave
> of the previous disk? Did you change anything in BIOS options?
>
> perhaps it would help if inside windows section of grub you have
something
> like this:
>
> title Other
> rootnoverify (hd1,0)
> map (hd1) (hd0)
> map (hd0) (hd1)
> chainloader +1
>
> HIH,
> Gianluca
>
Yes, that is correct...I moved the jumper on both drives to ensure the
old MASTER became the SLAVE and the new drive was the new MASTER.
Didn't do anything in the BIOS, and it recognized them both. The output
of parted shows that F7 got it right (sda has multiple Linux partitions
and sdb has two NTFS partitions).
I'll try the re-mapping tonight. The dual-boot HowTo documents I read
all seemed to indicate the problem associated with Windows wanting to
boot from the first drive was associated with Win 9x. I seem to
remember in the past getting WinXP to boot from the second or third disk
in the sequence. A couple of years back I had a disk failure, and when I
ended up installing the new disk, the drive with the bootable WinXP
partition was hd1. The machine was later rebuilt and XP ended up on hd0.
I could be wrong, but that was my recollection.
Dave McGuffey
Principal Information System Security Engineer
SAIC, IISBU, Columbia, MD
I181 TSAT, SETA
410.865.7972 (w)
16 years, 9 months
Re: kernel panic
by Paulo Cavalcanti
> As I pointed out 18 hours ago, if it was the case the root=(LABEL=/ ?)
> part of the kernel commandline in /boot/grub/grub.conf no longer matched
> the partitions on the drive for some reason, it could have directly
> explained it.
Because it is cached. It is necessary to delete
*/etc/blkid/blkid.tab
*before rebooting the first time.
Mine looks like this:
<device DEVNO="0x0802" TIME="1189078643" LABEL="/"
UUID="9d36358b-a777-4df2-8ac8-782a972145df" SEC_TYPE="ext2"
TYPE="ext3">/dev/sda2</device>
If the new "/" in my case is no longer /dev/sda2 it will not work (unless I
delete the cache).
--
Paulo Roma Cavalcanti
LCG - UFRJ
16 years, 9 months
2.6.22.4-65.fc7 hangs on PS3
by Fredrik Orderud
When attempting to boot the new kernel 2.6.22.4-65.fc7 on my PS3, the screen
goes black, and the console crashes imediately after selecting the kernel in
the bootloader. The original kernel 2.6.22.1-27.fc7 did and still does boot
normally.
The problem seems to be related to
https://bugzilla.redhat.com/show_bug.cgi?id=249217
, but this issue is closed since the problem was apparently fixed in kernel
2.6.22-1.37.fc7.
Does anyone else encounter problems when booting kernel 2.6.22.4-65.fc7 on
PS3?
-Fredrik
16 years, 9 months
how to locate a yum repo containing a particular package?
by Robert P. J. Day
for example, i want to install "kprof". i know an rpm is available
at kprof.sourceforge.net, but i'd like to just "yum install" it.
however, it's not available at any of my currently-enabled repos.
is there some kind of "meta yum" feature that would search a larger
set of repos for me? perhaps a set of "fedora-endorsed" repos?
thanks.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA
http://crashcourse.ca
========================================================================
16 years, 9 months
Selinux and exim
by Frank Chiulli
I have been the messages below for awhile. But since everything seems
to be working, I just ignored them. But now I've decided to fix them.
I tried:
touch /.autorelabel
reboot
But that did not fix it. Does anyone have any suggestions?
Messages:
avc: denied { append } for comm="sendmail" dev=hda2 egid=93 euid=93
exe="/usr/sbin/exim" exit=-13 fsgid=93 fsuid=93 gid=93 items=0
name="main.log" pid=7094
scontext=system_u:system_r:system_mail_t:s0-s0:c0.c1023 sgid=93
subj=system_u:system_r:system_mail_t:s0-s0:c0.c1023 suid=93
tclass=file tcontext=user_u:object_r:var_log_t:s0 tty=(none) uid=93
avc: denied { append } for comm="sendmail" dev=hda2 egid=93 euid=93
exe="/usr/sbin/exim" exit=-13 fsgid=93 fsuid=93 gid=93 items=0
name="panic.log" pid=7094
scontext=system_u:system_r:system_mail_t:s0-s0:c0.c1023 sgid=93
subj=system_u:system_r:system_mail_t:s0-s0:c0.c1023 suid=93
tclass=file tcontext=system_u:object_r:var_log_t:s0 tty=(none) uid=93
More info:
ls -Z /usr/sbin/exim
-rwsr-xr-x root root system_u:object_r:sendmail_exec_t /usr/sbin/exim
ls -Z /var/log/exim/panic.log
-rw-r----- exim exim system_u:object_r:var_log_t /var/log/exim/panic.log
ls -Z /var/log/exim/main.log
-rw-r----- exim exim user_u:object_r:var_log_t /var/log/exim/main.log
Thanks,
Frank
16 years, 9 months
re: Dual-booting F7 and WinXP on separate harddisks
by Gianluca Cecchi
> Bought a new 320 GB EIDE disk and installed it as MASTER and moved the
> previous MASTER (a 160 GB EIDE disk that booted WinXP) to SLAVE. That
> made the old hd0 hd1 and the new disk hd0. Without any OS installed on
> the new hd0, the machine continued to boot WinXP from hd1.
Are you sure of this above? No change at all, and swap from master to slave
of the previous disk? Did you change anything in BIOS options?
perhaps it would help if inside windows section of grub you have something
like this:
title Other
rootnoverify (hd1,0)
map (hd1) (hd0)
map (hd0) (hd1)
chainloader +1
HIH,
Gianluca
16 years, 9 months
Samsung ML-2510
by Thufir
The printer isn't responding. Now, there's been a USB issue with the
kernel, a bug? Everyone keeps responding about a Samsung ML-2010, so I'm
not convinced that this particular printer model works.
yum update all will update to the latest kernel?
CUPS just shows it as non-responsive. Should I go with splix, or the
propriety Samsung Linux driver?
thanks,
Thufir
16 years, 9 months
fedora 6, Network card question?
by chloe K
Hi all
I install 4 ports and 2 ports Intel NIC card in fedora 6
but the system is only showing eth0 to eth4 (totally 5 ports)
ifconfig -a
the first one is showing one is in "__tmp180428983 Link encap: Ethernet Hwaddr"
What is this one?
and
dmesg |grep eth
there is two lines for eth3
e1000: eth3: e1000_proble: Intel(R) PRO/1000 Network Connection
e1000: eth3: e1000_proble: Intel(R) PRO/1000 Network Connection
but I check /etc/moderprode.conf and /etc/sysconfig/network-sysconfig
is fine
Thank you for your help
---------------------------------
All new Yahoo! Mail
---------------------------------
Get news delivered. Enjoy RSS feeds right on your Mail page.
16 years, 9 months
F7 Choice between PDF reader
by Antonio M
On one of my system evince has been installed alon with Adobe Reader.
When I grab a PDF file from the Net, evince is started: how do I get
Adobe reader started instead??
There is no choice in Firefox options.
Tnx
--
Antonio Montagnani
Skype : antoniomontag
16 years, 9 months
Re: cp/scp
by Gianluca Cecchi
gnu tar by default preserves symbolic links, afaik.
tried on an fc6 and it is so.
You have to force -h option when you create the archive to tell gnu tar to
archive the file pointed to and not the link itself
Different thing it could be if you create / extract with different versions
of tar (example by AIX to Linux, ecc.)
scp doesn't copy links. For this you can do a sort of
tar zcf - p1 p2 | ssh target "cd target_dir; tar zxf -"
the considerations about gnu tar still are valid, so that if you have to
transfer to/from different flavours of Unix systems, you ought to use find +
cpio as they should be ok overall
sort of
find p1 p2 | cpio -o | gzip | ssh target "cd target_dir; gunzip | cpio
-idmuv"
should preserve link.
then man cpio and man tar run on the arch where you intend to launch the
command is your friend, to see link related options.
HIH and not confuses,
Gianluca
16 years, 9 months