Linux-Setup Digest #250, Volume #21 Thu, 17 May 01 17:13:08 EDT
Contents:
Q about redundant internet feeds, multiple gateways, redhat 7 (Ian Douglas)
Re: Installing Linux ("Peter T. Breuer")
How do I update RH 7.1 correctly using RPMs? ("George Adams")
RH 7.1 + XFS and SCSI LUN's (Mark Keever)
Re: Redhat 7.1: Whay linuxconf not run on KDE? ("Damaen")
Re: filename contains ":" characters in scp? ("Peter T. Breuer")
Re: Help Needed With SCSI Card (J Hayward)
Re: Redhat 7.1: Whay linuxconf not run on KDE? (Tim Haynes)
burned cd will boot but not install ("Rein Laaneser")
Re: LILO no boot .. says "LIL-" then just hangs there (Michael Marion)
Re: Newbie ? on Install from CD ("KW")
xircom cardbus 10/100/modem and RH 7.1 ("bullfightparty.org")
Re: How do I update RH 7.1 correctly using RPMs? (J Hayward)
Re: RH 7.1 + XFS and SCSI LUN's ([EMAIL PROTECTED])
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Ian Douglas)
Subject: Q about redundant internet feeds, multiple gateways, redhat 7
Date: Thu, 17 May 2001 18:03:40 GMT
We have 2 available Internet connections, for redundancy, but we're
having trouble getting the second feed to work.
We have two feeds coming in on one 10baseT cable into a hub, and from
the hub we are connecting to the two network cards.
Our ifconfig setup:
eth0 Link encap:Ethernet HWaddr 00:00:00:00:00:01
inet addr:1.2.3.226 Bcast:1.2.3.231 Mask:255.255.255.248
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1914722 errors:0 dropped:0 overruns:0 frame:0
TX packets:2100107 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
Interrupt:11 Base address:0xa800
eth2 Link encap:Ethernet HWaddr 00:00:00:00:00:02
inet addr:4.5.6.226 Bcast:4.5.6.231 Mask:255.255.255.248
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:9570 errors:0 dropped:0 overruns:0 frame:0
TX packets:51 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
Interrupt:3 Base address:0xa000
Our routing table:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref
Use Iface
4.5.6.224 0.0.0.0 255.255.255.248 U 0 0 0
eth2
1.2.3.224 0.0.0.0 255.255.255.248 U 0 0 0
eth0
0.0.0.0 1.2.3.225 0.0.0.0 UG 0 0 0
eth0
The default gateway route for the 4.5.6.224 network is SUPPOSED to be
.225, but when I try to add that route to the routing table after
removing the default to go through 1.2.3.225, I get this:
root /etc> route del -net 4.5.6.224 netmask 255.255.255.248
root /etc> route add -net 4.5.6.224 netmask 255.255.255.248 gw
4.5.6.225 eth2
SIOCADDRT: Network is unreachable
root /etc> _
So, I added the network back on the default gateway for the time
being:
root /etc> route add -net 12.146.197.224 netmask 255.255.255.248 eth2
I'm pretty low on the skill side of setting up redundant links like
this. Can someone provide any pointers on what I can try? I'm above
average in terms of other areas of Linux administration, just never
set up a second network feed like this.
RedHat 7.0, i386, 2.2.16-22
Any assistance or pointers would be greatly appreciated.
Ian Douglas
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: Installing Linux
Date: Thu, 17 May 2001 19:49:12 +0200
Ammon <[EMAIL PROTECTED]> wrote:
> I have a two part question.
> 1) How do you install Linux without completely reformatting (and losing
> all old data) your computer?
You don't do the reformatting, full stop! You need one or two
free partitions on which to install linux, and distros come with tools to
make them if you don't have them, shrinking your old partitions to suit.
> 2) How do you create a dual-boot system so that you can choose which OS
> you want your PC to boot up?
In the normal way. Please read the instructions that come with your
distro of linux. It will tell you how to put both winodws and linux as
option sin th eboot loader.
Peter
------------------------------
From: "George Adams" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.admin,comp.os.linux.misc
Subject: How do I update RH 7.1 correctly using RPMs?
Date: Thu, 17 May 2001 14:26:17 -0400
I have a basic question about keeping my RH 7.1 Linux box up-to-date. I see
that there are already several patches and upgrades of various sorts
available in the form of RPMs off Redhat's website, which I have downloaded.
Now what? For instance, if I do an "rpm -Uvh
netscape-common-4.77-1.i386.rpm", I get several warning messages about other
programs (such as Netscape Navigator and KDE) which are dependent on the
netscape-common-4.76 files. Should I ignore these warning, and issue a
"rpm -e -nodeps" for the old netscape-common-4.76 package, then install the
new package? Or will this break some program or tool that needed the old
package and doesn't know what to do with the new package?
Another example:
> rpm -Uvh arts-2.1.2-1.i386.rpm
error: failed dependencies:
arts = 2.1.1 is needed by kdelibs-sound-2.1.1-5
If I force the removal of the old 2.1.1 package, then install 2.1.2, will
kdelibs-sound be OK with that? Or will it still be looking for arts 2.1.1
(and therefore break).
That's just a couple of examples, but I guess what I'm looking for is a more
general answer - what's the normal procedure for keeping a Linux box
up-to-date? Is it the same set of commands that I just issue each time I
download an updated RPM? And can I automate the process (through autorpm or
some such tool)?
Thanks to anyone who can help me get started.
------------------------------
From: Mark Keever <[EMAIL PROTECTED]>
Crossposted-To: git.unix.linux
Subject: RH 7.1 + XFS and SCSI LUN's
Date: Thu, 17 May 2001 14:52:52 -0400
I have a system that has RH 7.1 + XFS. I am trying to set up a disk
device that needs to use LUN's. I checked the kernel, and probe all
luns is there, but the system only seems to be looking at lun 0. I know
the device is there because the scsi controller card identifies it
correctly with the luns. I have tried the max_scsi_luns parameter and
the aic7xxx_mod.o driver says that this is an invalid parameter.
The scsi card is an adaptec 29160 and works correctly when not using
luns with this system.
Any suggestions?
------------------------------
From: "Damaen" <[EMAIL PROTECTED]>
Subject: Re: Redhat 7.1: Whay linuxconf not run on KDE?
Crossposted-To: comp.os.linux.security
Date: Thu, 17 May 2001 18:55:46 GMT
In article <[EMAIL PROTECTED]>, "Mark Schlegel"
<[EMAIL PROTECTED]> wrote:
>> Damaen wrote:
>> Ok... I'm confused... linuxconf works on my box, with no mention of
>> being deprecated, nslookup was the only program I noticed proclaiming
>> that. So basically this is one big fat... HUH??
>
> see section #17 near the bottom of the RedHat 7.1 manual:
>
> http://www.redhat.com/support/manuals/RHL-7.1-Manual/release-notes/s1-system.html
>
Wow... thanks for the link... I had no clue. Not that I use it very
often... but it's nice for quick changes... That list had netscape 4.x
deprecated too... but I can't say I like Mozilla much. I still use 4.x,
it seems much faster on my machine. Oh well...
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: filename contains ":" characters in scp?
Date: Thu, 17 May 2001 20:59:03 +0200
In comp.os.linux.setup Colin Watson <[EMAIL PROTECTED]> wrote:
> Peter T. Breuer <[EMAIL PROTECTED]> wrote:
>>In comp.os.linux.setup Colin Watson <[EMAIL PROTECTED]> wrote:
>>> Debian bug #47000), the inclination is to fix something else instead.
>>That was uncalled for, colin. Bug 47000 looks kosher to me!
> It is a kosher bug, yes. Unfortunately, it's also several bugs (some of
> which I agree with and some I don't) rolled up into one, together with
> lots of sarcastic remarks. Nothing can be closed until the whole thing
I don't find the remarks sarcastic, rather more a transcription of
what the students said to me in class!
> is done, which makes tracking very difficult for the maintainer (i.e. I
> can't do what I do for every other bug, put closes: #nnnnn against the
> fix in the changelog, and have the bug tracking system remember the rest
> of it for me).
If you fixed the bug of trying to write in /usr, and made token changes
to the docs to make them more comprehensible, call it closed! You have
my "permission", should you need it! Man-db does work nowadays. It's
stopped behaving strangely (mostly), so there's no need to try and
fathom the docs any more. Doc bug avoided! And if I disagree with the
bug closure I'll just make life hell by entering another bug report,
and we go round again :-). But it's not likely that I'll disagree.
BTw, I see only two bugs reported in that report (my numerals):
(1)
/usr is mounted readonly, yet man insists at times on trying to update
links and caches and dbs. It should check that the dir is writable
before trying it!
(2)
Secondly: many doc bugs. What does "considered \"global\"" in the
manpath.config connote, if anything. Why is it considered? Isn't
it? Is there anything apart from "global", whatever you mean by it?
What does it mean if not nothing? Please remove this GUFF!
> And, really - if you had thirty-odd open bug reports, and one of them
> contained stuff like "What does it mean if not nothing? Please remove
> this GUFF!", would you be inclined to fix that one first? It's the best
Well, I am - if nothing else - qualified to comment on scientific writing!
So perhaps you should take the criticism of the writing seriously.
I took the time to say what was wrong with it! The point (2)
above is the start of the list of specific comments I made about the
writing, quoting paragraphs and commenting on each. In other words, I
treated it as if it were a scientific paper, applying the standard
refereeing practices, the intent of which is to produce a better
document through negotiation with the author ("peer review" :_).
> way of making a non-critical bug fall to the bottom of the pile.
The write on /usr problem was functionally critical. It stopped man-db
working in environments in which /usr is readonly. That the docs were
incomprehensible I'd count as critical in other ways, given that we're
talking about the manual pages package - after all, if they couldn't
explain themselves, they couldn't explain anything.
> I guess I'm trying to make a point (well, perhaps overstated - I
> apologize for going off half-cocked) about bug reporting, just as about
> helping people on newsgroups - making sarcastic remarks about the
> package or its maintainer is almost always the least productive thing
I had no sarcastic intent then. I can't even see any sarcasm in the report
you mentioned now. Are you perhaps reading something into it? I made
the debian report after talking directly with the author (I believe I
recall). Our discussions were fruitful (not fruity!), and I added the
debian report as a sort of deadline/guarantee that something would be
done.
> it's possible to do, particularly when the maintainer is a volunteer.
> http://www.chiark.greenend.org.uk/~sgtatham/bugs.html - I get it wrong
> from time to time as well, but it's worth remembering that there's an
> ordinary person at the other end (usually) trying to fix it.
>> It is very "unsightly" to users. It undoes the work of persuading them
>> to read manpages in the first place.
>>
>>I still see problems with man-db. I have no idea why, but occasionally
>>it just won't see manpages that I can read with nroff. The only cure is
>>to su - man, and run man-db.
> 'man --debug' is the key to a useful bug report here. Some code related
> to this will be changing soon anyway.
> man-db is indeed broken in many interesting ways. I hope and believe
> that it's a lot less broken now than it used to be (assuming you're
> running testing/unstable).
No, I'm running "stable". Perhaps I'll try testing's source.
>>Have a look at the report and see if it makes more sense to you now!
>>Are you the maintainer? I remember a conversation partly in german
>>with the author, whoever it is.
> The previous maintainer passed away a couple of months ago, and I took
> the package over then. Believe me, I have read the report recently. :)
>>I went on to complain about the documentation, pointing out what made
>>it so wonderfully incomprehensible. Has that been fixed ...
> I rewrite bits and pieces as I see them. On one point I agree with
> Fabrizio - since I have lots of code to fix and/or rewrite, any offers
> to help improve the documentation will be appreciated. Note that
I believe I did that, in a very concrete manner.
> Fabrizio's comment that the manual only has a verbatim licence is
> inaccurate, so the whole thing doesn't need to be rewritten.
>>> 'tar cf - ... | ssh remote-host tar xvf -' can be a useful way to get
>>> around scp's (apparent) lack of an escape.
>>
>>Yes, good idea. But if you're scp'ying a directory, there won't be any
>>trouble with the filenames inside.
> Assuming that the directory name doesn't contain colons, of course, and
> you could always call tar on a single filename too.
> Cheers [followups set to poster, as this is getting onto a different
> topic],
I unset them, probably. Dunno why.
Thanks for the followup (honestly!).
Peter
------------------------------
From: J Hayward <[EMAIL PROTECTED]>
Subject: Re: Help Needed With SCSI Card
Date: Thu, 17 May 2001 12:09:33 -0700
Hello,
I have never played with one of these cards, but try this see if it works:
1. boot windoze,open the device manager, write down the io and IRQ setting
it has listed for your SCSI card.
2. boot Linux
3. open /etc/modules.conf
4. Add these lines. It is possible you might not need the options line or
you might need the options line with only the IRQ. Looking a the source
code for the driver it looks like it does some autodetection:
alias scsi_hostadapter seagate
options seagate=ADDRESS,IRQ
replace ADDRESS,IRQ with the values you wrote down from windoze
Example: seagate=0xC8000,5
5. save the changes
6. in a console window type: modprobe scsi_hostadapter
It should work. /sbin/lsmod will display a list of the loaded modules.
Regards,
Jim H
------------------------------
From: Tim Haynes <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.security
Subject: Re: Redhat 7.1: Whay linuxconf not run on KDE?
Date: 17 May 2001 20:29:03 +0100
Reply-To: [EMAIL PROTECTED]
"Damaen" <[EMAIL PROTECTED]> writes:
[snip]
> > http://www.redhat.com/support/manuals/RHL-7.1-Manual/release-notes/s1-system.html
> >
>
> Wow... thanks for the link... I had no clue. Not that I use it very
> often... but it's nice for quick changes... That list had netscape 4.x
> deprecated too... but I can't say I like Mozilla much. I still use 4.x,
> it seems much faster on my machine. Oh well...
I know the feeling, and think it'd be good to have it as a fall-back for a
little while yet. There are just too many sites out there that don't grok
Mozilla itself, but only look for IE or NS (I know, these sites are broken,
I either mail the webmaster a load of abuse as they deserve or give up & go
away...)
An alternative, btw, is definitely Konqueror. It rocks, I figure. Mozilla
is only just catching up with it - while some people were making noise &
hot air, folks at KDE were making a browser.
~Tim
--
You take your message to the waters, | [EMAIL PROTECTED]
And you watch the ripples flow | http://piglet.is.dreaming.org
------------------------------
From: "Rein Laaneser" <[EMAIL PROTECTED]>
Subject: burned cd will boot but not install
Date: Fri, 18 May 2001 07:56:41 +0300
* Burned cd will boot but not install *
Why would burned (from ISO images, RH 7.1, disk1 and disk2)
boot normally but not install???
In booting will be message
Boot from ATAPI CD-ROM:
1. FD 2.88MB System Type-(00)
Then Syslinux will run, there is "Welcome to Red Hat Linux 7.1" screen
Then install will start but soon will be message
Error
I could not find a Red Hat Linux CDROM
in any of your CDROM drives.
Please insert the Red Hat CD and press
"OK" to retry.
Is it familiar problem to someone.????
------------------------------
From: Michael Marion <[EMAIL PROTECTED]>
Crossposted-To: dc.org.linux-users,comp.os.linux.advocacy
Subject: Re: LILO no boot .. says "LIL-" then just hangs there
Date: Thu, 17 May 2001 20:21:17 GMT
Bernadette and Warren McRobie wrote:
> I found that having the drive linux is installed on a slave with no master,
> then this error occurs - yes I know it should be the master - but I did not
> set up this machine hardware wise ;P
Better yet.. directly from the docs:
LIL- The descriptor table is corrupt. This can either be caused by a
geometry mismatch or by moving /boot/map without running the map
installer.
--
Mike Marion-Unix SysAdmin/Senior Engineer-Qualcomm-http://www.miguelito.org
Homer: "I'm not normally a religious man, but if you're up there, save me,
Superman!" -- Simpsons
------------------------------
From: "KW" <[EMAIL PROTECTED]>
Subject: Re: Newbie ? on Install from CD
Date: Thu, 17 May 2001 15:24:09 -0500
It sounds like you've burned the ISO as a data file. The ISO is an image
of the CD. Tell your cd burning software to burn from an image and
select the ISO file.
If your using adaptec easy cd 3.5c
File
Create CD from Disc Image
Select files of type *.ISO (it default to .cif)
Select the CD1 iso and burn away...
In article <9e12rs$gtf$[EMAIL PROTECTED]>, "Nick T"
<[EMAIL PROTECTED]> wrote:
> I have download the 2 iso files to install Red Hat 7.1 and burned them
> onto a CDR. My pc will not boot to the cd. I have the bios set to boot
> to the cd first but it does not see the cd as bootable. I also tried to
> download the boot.img file and that does not help when burned on the cd
> with the iso file. Any help you could provide in how to install off a
> bootable cd would be great. Thanks.
> Nick
>
>
------------------------------
From: "bullfightparty.org" <[EMAIL PROTECTED]>
Subject: xircom cardbus 10/100/modem and RH 7.1
Date: Thu, 17 May 2001 16:23:15 -0400
hello. i have a new dell inspiron 4000 running RH7.1 with a cardbus
xircom 10/100 +modem56 realport. the ethernet device works fine, but
when i try to test the modem with rp3-config i get "Modem not
responding" errors. rp3 initially identified the modem as /dev/ttyS0.
thanks!
chuck
------------------------------
From: J Hayward <[EMAIL PROTECTED]>
Subject: Re: How do I update RH 7.1 correctly using RPMs?
Crossposted-To: comp.os.linux.admin,comp.os.linux.misc
Date: Thu, 17 May 2001 13:45:15 -0700
Hello,
If you have all the updated RPMS in the same directory just use:
rpm -Fvh *.rpm
The "F" means "freshen". This will install the updates only for the RPMS
you already have installed.
Look at the manpage for rpm too see all the options it uses.
Regards,
Jim H
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: git.unix.linux
Subject: Re: RH 7.1 + XFS and SCSI LUN's
Date: 17 May 2001 20:50:27 GMT
In git.unix.linux Mark Keever <[EMAIL PROTECTED]> wrote:
: I have a system that has RH 7.1 + XFS. I am trying to set up a disk
: device that needs to use LUN's. I checked the kernel, and probe all
: luns is there, but the system only seems to be looking at lun 0. I know
: the device is there because the scsi controller card identifies it
: correctly with the luns. I have tried the max_scsi_luns parameter and
: the aic7xxx_mod.o driver says that this is an invalid parameter.
: The scsi card is an adaptec 29160 and works correctly when not using
: luns with this system.
: Any suggestions?
2.2 or 2.4 kernel? Stock RH kernel? Doug's driver is being picked up by a
new maintainer and he's rewriting it from scratch. Under the 2.2 kernel, use
Doug's driver. Under the 2.4 kernel, use the new aic7xxx driver and not the
old one.
--
David Cantrell, KG6CII
http://www.burdell.org/
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.setup.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Setup Digest
******************************
