Re: [SLUG] Lilo problem

2000-08-03 Thread Stephen Mills


On Fri, 4 Aug 2000, Peter Rundle wrote:
> Melinda,
> 


> I believe this limitation has been fixed in more recent distros,
> does anyone else have more info.?

More so the latest version of Lilo (21.5) addresses this problem, if the
BIOS supports EDD packet call extensions to the int 0x13 interface.

I installed this version (although for different reasons), worked first go
no problems.

It now has a menu kinda interface. Fully backward compatiable with older
versions, you might want to give it a shot.

ftp://metalab.unc.edu/pub/Linux/system/boot/lilo/ 

--Steve




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] help! Environment variables.

2000-08-03 Thread Stephen Mills

/etc/profile is the most popular and most recognised.

Being Redhat you may also notice /etc/profile.d which is used by RPM's to
export variables to newly installed programs instead of editing
/etc/profile which has less chance of breaking things.

Hope this helps
--Steve


On Fri, 4 Aug 2000, George Vieira wrote:

> have you tried /etc/rc.d/rc.local ?
> and if you need individual different environments then add it to the
> .profile or .bash_profile for each user...
> 
> thanks,
> George Vieira
> Network Administrator
> Citadel Computer Systems P/L
> http://www.citadelcomputer.com.au
> 
> 
> 
> -Original Message-
> From: Stuart Hume [mailto:[EMAIL PROTECTED]]
> Sent: Friday, 4 August 2000 12:04 PM
> To: [EMAIL PROTECTED]
> Subject: [SLUG] help! Environment variables.
> 
> 
> I need to set some default global environment variables under red hat so
> that every user in the system will be able to use it. Is there any way of
> doing this when the system starts up,i.e when the machine reboot?
> 
> Apprentice Admin wantabe.
> Louie.
> 
> 
> 
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
> 
> 
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
> 



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: [SLUG] Gigabummer?

2000-08-05 Thread Stephen Mills

start with the basics

have you tried disabling the CPU cache ?
tried different memory ?

a better description of errors and what not might give us a clue

Cheers,
Steve

- Original Message -
From: "Terry Collins" <[EMAIL PROTECTED]>
To: "Richard Blackburn" <[EMAIL PROTECTED]>
Cc: "Sydney Linux Users Group" <[EMAIL PROTECTED]>
Sent: Sunday, August 06, 2000 10:14 AM
Subject: Re: [SLUG] Gigabummer?


> Richard Blackburn wrote:
> >
> > Anyone had problems with Gigabyte motherboards and Linux?
>
> No, it is working fine and has been for a few years.
>
>
> > I installed one with dual PII cpus and it's been a nightmare getting
Linux
> > to run and/or load.
>
> But, I've never had a dual CPU one.
>
> Post the model might help.
>
> --
>Terry Collins {:-)}}} Ph(02) 4627 2186 Fax(02) 4628 7861
>email: [EMAIL PROTECTED]  www: http://www.woa.com.au
>or [EMAIL PROTECTED]
>WOA Computer Services 
>snail:  PO Box 1047, Campbelltown, NSW 2560.
>
>  "People without trees are like fish without clean water"
>
>
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
>



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] ISPs and Permanent Static IPs

2000-08-06 Thread Stephen Mills

We're paying around $24k / year for our 128k frame relay (syd-melb with
voip)

You'll find most 64k ISDN dialups are unlimited download, from memory we
were paying $750-$800 with Ozemail at the time, around a year ago

Must admit im not too happy with our provider, tpgi - they send our /24 x
2 networks out via the USA first ! Maybe because they get a cheaper rate
or something

Mmm funny what you said about 1 IP for ADSL, I hear you can have more then
1 IP if your router is a business "router" from telstra...

We'll probably move to ADSL when there is ports in Melb suburb of Kew

Cheers,
Steve


On Mon, 7 Aug 2000, George Vieira wrote:

> OK.. currently we're paying around $12K a year for a 64Kb frame relay link
> and as far as I know it's unlimited until they start winging at us..
> 
> I just found out that ADSL and HDSL don't have the facility to support
> multiple IPs on the 1 link, fair enough I guess as it's basically a dial up
> account.
> 
> So I guess the answer would be for ISDN or something else..
> 
> thanks,
> George Vieira
> Network Administrator
> Citadel Computer Systems P/L
> http://www.citadelcomputer.com.au
> 
> 
> 
> -Original Message-
> From: George Vieira [mailto:[EMAIL PROTECTED]]
> Sent: Monday, 7 August 2000 11:46 AM
> To: '[EMAIL PROTECTED]'
> Subject: [SLUG] ISPs and Permanent Static IPs
> 
> 
> Hi y'all,
> 
> We are currently looking at changing ISPs due to the fact that our current
> ISP moved us off the LAN link to Telstra and onto the satellite link which
> is 500ms slower. Since doing so we have requested them to put us back and
> now they're trying to charge us for it.
> 
> So now I'm looking for a better ISP to provide us with 16 static IPs and a
> frame relay or whateva media necessary for a decent link... I was thinking
> cable but they probbly charge an arm and a leg for it and possibly don't
> have static IPs yet but I actually don't know. ASDL is another thought but
> it's not available yet
> 
> Anybody care to throw some names or something...
> 
> thanks,
> George Vieira
> Network Administrator
> Citadel Computer Systems P/L
> http://www.citadelcomputer.com.au
> 
> 
> 
> 
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
> 
> 
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
> 



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: ADSL is not a direct connection to *The*Internet* (Was: ADSLrant(was: Re: [SLUG] Telstra Cable Anouncement))

2000-08-08 Thread Stephen Mills

> No doubt a few people will thank us both for shutting up.
> 
> Cheers.
> 
> - Raz
> 
Amen.

Agree to disagree fellas 

--Stephen



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: [SLUG] Linux -v- M$ Costs

2000-08-09 Thread Stephen Mills

Hey Howard,

For a 25 user license for SBS, its around $7400

That includes :
MS Windows NT Server 4.0 with Service Pack 4 + Windows NT Option Pack
MS Exchange Server 5.5 with Service Pack 2 and Outlook 2000
MS SQL Server 7.0
MS Systems Management Server 2.0
MS Proxy Server 2.0
MS Site Server 3.0 with Service Pack 2
MS SNA Server 4.0 with Service Pack 2, Seagate Crystal Info 6.0

Ive done a few SBS installs, and I might add they cause endless grief is
something is broken, SBS 4.0 was an absolute shocker, and SBS 4.5 is slighly
better...You fix something, and it breaks something else...

You could go out and buy MS individual packages, like NT, Exchange, although
I can see around 4k alone in costs just with those two items with 10 user
license - im quoting very roughly though, im not a salesman  :)

Ive found MS product to be overpriced, havent we all ?
For eg. An unlimited version of MS SQL 7.0 will set you back a cool $48,000.

The thing that amuses me the most is their migration tools for UNIX and
Netware are very cheap

I quoted a similar price to a client, (around 7k) I said I can do it all on
Linux and the software won't cost you anything, dont we just love saying
that ?.they obviously went with Linuxalthough I kept file and print
sharing on Netware.

Remote admin :
Linux -  SSH or telnet.
Win2k - install the terminal services add-on (its free) or get something
like pc anywhere.

If you want accurate pricing of MS stuff, try www.ht.com.au (Harris
Technology) ; there is a couple vendors around, but the names escape me at
the moment.

Hope this helps.
Stephen

- Original Message -
From: "Howard Lowndes" <[EMAIL PROTECTED]>
To: "Mail List - SLUG" <[EMAIL PROTECTED]>; "Mail List - Oz-ISP"
<[EMAIL PROTECTED]>
Sent: Wednesday, August 09, 2000 4:57 PM
Subject: [SLUG] Linux -v- M$ Costs


> I am trying to put together a cost comparison for a client comparing the
> costs of going down the wholly M$ path as against the Linux servers/M$
> desktop path.
>
> The client is a small office setup with around 10 workstations.  They will
> want a permanent Internet connection and the plan is to connect them
> through BPD so they will be running their own mail server, and later on
> their own low volume web server.  Ignore the desktops, they will be
> staying M$.
>
>
> I have a good idea what the Linux costings will work out at, but I am not
> too sure of the M$ costs.  If anyone has any info in this field they would
> care to impart I would be grateful.
>
> The areas I am looking at are:
>
> Firewalls and Internet connectivity.  What does M$ need here?
>
> Licence fees.  I can get these worked out.
>
> Mail server.  Does M$ need Win2K server and M$ Exchange server?
>
> Internal file & print server.  I assume Win2K server will do here.
>
> What does it cost for this lot to be set up by a COMPETENT M$ bod?
>
> Costs of maintenance and support down the line and ease of remote
> maintenance.
>
> --
> Howard.
> __
> LANNet Computing Associates 
>




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: [SLUG] SCSI CD-WR... not

2000-08-09 Thread Stephen Mills

I know its a pain but try removing everything except the IBM drive and new
Yamaha and triple check your termination.

I presume all those devices are 50 pin or you -will- have terminating issues
as your last device is narrow.

Cheers,
Stephen

- Original Message -
From: "Geoffrey Robertson" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, August 09, 2000 8:17 PM
Subject: [SLUG] SCSI CD-WR... not


> I've just tried to add a SCSI Yamaha CD-RW (8424)
> to a box today with no success.
>
> At boot up the system hangs with the message:
> ...timeout pid 31 SCSI 0 Ch 0 id 3 lun 0
> test not ready 00 00 00 00 00
> printed repeatedly
>
> anybody able to shed any light on what this means?
> geoffrey
>
>
> box details: pII 266 with an adaptec 2940UW
> running RH 6.something
>
> The card seems to recognise the device ok
> and i think everythings terminated ok
>
> the Yahama is plugged into the middle of this chain:
> 0 IBM disk
> 1 HP scanner
> 2 NAKAMICHI jukebox
> 3 YAMAHA  cd-rw new thing
> 4 IOMEGA zip
> and all the other stuff works ok
>
>
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
>



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: [SLUG] Linux -v- M$ Costs

2000-08-09 Thread Stephen Mills

> >Ive found MS product to be overpriced, havent we all ?
> >For eg. An unlimited version of MS SQL 7.0 will set you back a cool
$48,000.
>
> We did a costing for a 100 user SBS system, ignoring all hardware
> requirements, and it came to a shade under $100k - I kid you not !!  That
> was for CAL's, Server, Exchange, Office... And bugger all else...  I
forget
> what the comparitive Linux price was, including licenced copies of things
> like StarOffice, BRU, a good mail client, etc. Something like $5k

I guess the biggest problem if your converting Office-Windows-Users to Linux
cold turkey they will need someone that knows Linux fairly well to support
the userbase - although with 100 users Im sure they have someone there
already :)
I am getting very excited about the software that is coming out for the
desktop for the mainstream...

Actually, I read an article on slashdot (?) a few weeks ago on a proposal of
a userbase of 2000 users moving their desktops to Linux, I did a quick
search for the article but came up with zip, sorry..

> Why did you stay with Netware for file & print ?  reason I ask is my
> network Australia-wide is about to start dumping Netware in favor of NT
for
> file & print (NT4 initially, upgrading to 2000 later). I'm trying to
> convince the PHB that I can do the same job on a Linux box and get more
> bangs-per-buck for the hardware they're going to use ($30k COMPAQ
servers).

it was mainly because of a few factors
1) its a seperate server that has had uptime of over 1 year, why fix
something that isnt broke
2) all the backup software / hardware [tape drive] is geared up
3) the cost for me (in downtime and money) to move everything over to samba
isnt worth it
4) its nice to have more then 1 machine doing all the work for you,
especially if hardware fails

come to think of it we're running netware 3.12 for our file/print services,
it gets rebooted (turned off) only at xmas.
our exchange server and iis server gets rebooted more times then i make
coffee ! :)

> HT tend to be VERY overpriced compared to what some of the smaller
MicroSloth vendors are willing to offer.

I agree - you could do a lot better, if you looked about.


Cheers,
Stephen



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] squid fun

2000-08-10 Thread Stephen Mills

with a default install of squid, it will allocate 100 megs for cached  
objects, my cache at present is taking up 120 megs, im keeping an eye on
it.

is this normal ?

The server has been up for about 3 months now, and I only just noticed it,
so it may well been at 120 megs for the past 2 months

--Steve



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] CIFS/SMB

2000-08-10 Thread Stephen Mills

Does anyone know of any limits to the amount of shares you can have per
machine using Samba ?

Ive checked the Using Samba book but can't notice anything interesting -
if there is a limitation, I wonder if Windows NT has a similar limit, the
reason I ask is because we sell NAS boxes (www.snapserver.com) and at this
stage they have a 250 share limit.

My guess is there is no limit on Samba,NT

--Steve




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: [SLUG] OT: If you need to look after NT what mailing are yousubscribed?

2000-08-11 Thread Stephen Mills

hit the newsgroups - there is a microsoft news server that handles all
this - be warned the amount of traffic is huge

On Fri, 11 Aug 2000, Jobst Schmalenbach wrote:

> 
> All,
> 
> This is a uestion for people who look after a mixture of environments
> ie Linux AND NT:
> 
> 
>   What "Windows NT mailing list" are you subscribed to?
>   Which lists are good?
> 
> 
> 
> 
> Jobst
> 
> 
> -- 
> Share your knowledge. It's a way to achieve immortality.
> 
> |__, Jobst Schmalenbach, [EMAIL PROTECTED], Technical Director|
> |  _ _.--'-n_/   Barrett Consulting Group P/L & The Meditation Room P/L  |
> |-(_)--(_)=  +61 3 9532 7677, POBox 277, Caulfield South, 3162, Australia|
> 
> 
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
> 



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: [SLUG] why not to multi-boot

2000-08-12 Thread Stephen Mills

> Question: can I get any of the data on /dev/hda2 back by
> reconstructing the superblock? I suspect that manual examination
> of the segments on a 2G partition would be quite boring.

you have your redhat cd don't you ? 
if your cdrom drive is bootable throw it in, if not make a boot disk
run rescue.
run e2fsck -n /dev/hda2 - it will try and search for a backup superblock, 
the -n flag is used to scan in read-only mode
if your happy with the results remove the -n
e2fsck should scan other superblocks, but if for some reason it doesnt
work,
try forcing it:
run e2fsck and specify another superblock
e2fsck -b 8193 /dev/hda2

if you dont have the redhat cd, go and grab toms root/boot disks
www.toms.net

its not uncommon for linux not to mount the root fs if it finds an error -
hopefully for you its something that can be repaired, fingers cross :)

--Steve
> 
> The partitions were:
> /dev/hda1: 1GB FAT16
> /dev/hda2: 2GB ext2 Red hat 6.2 running 2.2.12 I think.
> /dev/hda3: 1GB swap space
> 
> Jim Donovan
> 
> 
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
> 



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] why not to multi-boot

2000-08-12 Thread Stephen Mills

kernel 2.2 can handle up to 2G swaps
since his root filesystem is 2gig, the 1 gb swap is overkill :)
id remove the wsap, and create a new swap around 120 meg and use the rest
and mount under home or something

-Steve

On Sun, 13 Aug 2000, Dave Kempe wrote:

> > /dev/hda3: 1GB swap space
> A bit off topic, but aint that a huge amount of swap space? I generally only
> use about 150meg at most.
> 
> dave
> 
> 
> 
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug
> 



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: [SLUG] interesting scsi question

2000-08-13 Thread Stephen Mills

> I'm having a bit of trouble with a scsi CD writer. When I try
> 
> mount -t iso9660 /dev/scd0 /mnt/cdr
> 
> I get a "not a block device" error.
> 
> I have also tried MAKEDEV sg to make generic nodes but these dont work
> either.
> rob


what does your scd0 device look like ? mine looks like :
[smills@proxy /dev]$ ls scd0 -al
brw---1 root disk  11,   0 May  6  1998 scd0

you might want to have a fiddle with mknod
it seems your scsi support is there, but you might want to make sure you
have SCSI CDROM support - Redhat by default compiles it into the kernel

--Steve




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] diskspace

2000-08-14 Thread Stephen Mills

I have a strange problem with available disk space.

the root partition is 407 megs, for which 380 megs is reported being used,
although this is only 5.5 megs available on the partition.

I removed about 40 megs with of packages, mostly X apps and the like, and
about 30 mins after all that space I recovered was being chewed up again,
but I can't find where its been taken.

I have only two partitions on this server, its strictly a mailserver and
proxy server.

one is mounted to /  (407 meg)
the other is mounted to /var/spool/mail (1.5 gig) (1 gig free)

Over time I have been asked to add things, like a proxy server etc, so Ive
mounted the cache under /var/spool/mail/proxy to utilise that space. I am
going to be replacing the server with a new box very soon.

The pop3 toaster seems to copy the users mailbox to /tmp first and then
pulls the mail down, so I removed /tmp and created a symlink to from /tmp ->
/var/spool/mail/tmp just in case. Although I still can't find where my space
has gone.

Ive checked out the find syntax, but havent worked out a way to find files
sorted in size order, from largest to smallest or files over a certain size
to try and find out where the space is going.

Any clues ?

--Stephen


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] ASP and apache?

2000-08-14 Thread Stephen Mills

http://www.apache-asp.org/

--Stephen


Hi all,

I'm not sure if it exists or what but is there ASP support for Apache 1.3?
Someone from out clients site has decided to run ASP and wants to remove the
Linux box there and I'd like to keep it there is possible...

thanks,
George Vieira
Network Administrator
Citadel Computer Systems P/L
http://www.citadelcomputer.com.au




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] diskspace

2000-08-14 Thread Stephen Mills

Ive removed all the log files .1,.2,.3,.4 and changed the cycle to only
logrotate once a week, so thats not it

the cache objects are mounted under /var/spool/mail/proxy so that should not
effect the root filesystem

ive removed /tmp directory and created a symlink from /tmp ->
/var/spool/mail/tmp so i doubt thats it :(

-Original Message-
From: Jason Rennie [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, August 15, 2000 1:44 PM
To: Stephen Mills
Cc: '[EMAIL PROTECTED]'
Subject: Re: [SLUG] diskspace


> I removed about 40 megs with of packages, mostly X apps and the like, and
> about 30 mins after all that space I recovered was being chewed up again,
> but I can't find where its been taken.

Big log files ? /tmp filling up ? Squid proxy caching objects ?

Jason

-- 
GnuPG Key 2450EEDC Jason Rennie<[EMAIL PROTECTED]>
Key fingerprint = 1A2B 5E34 B45A 2871 A488  99C7 7579 5FFC 2450 EEDC



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] diskspace

2000-08-14 Thread Stephen Mills

Thanks for your input Raz. 

I have solved the problem, what I needed was some lunch and to think about
it, and I checked the queue on the MTA, seemed some bozo had tried to send
120 megs worth of TIFF's (this company does aeroplane designs). I killed the
queue and symlinked the queue directory to the 1 gig free partition and all
is well and with 120 meg free on the root filesystem.

Thanks again for your suggestions guys,
Stephen


Stephen Mills wrote:

> I have a strange problem with available disk space.
>
> the root partition is 407 megs, for which 380 megs is reported being used,
> although this is only 5.5 megs available on the partition.

It would help to see the output(s) from the tool(s) that you are using
to gather this information. The above sentence alone suggests that you
are either misinterpreting the output from the tool(s), or that you are
combining different kinds of "size" data (e.g. bytes stored in files vs.
disk blocks used * block size).

> I removed about 40 megs with of packages, mostly X apps and the like, and
> about 30 mins after all that space I recovered was being chewed up again,
> but I can't find where its been taken.

You've checked the contents of /var/log?

Perhaps something is growing a deleted file. Does the space magically
reappear upon rebooting?

- Raz


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Is Arkeia Worth $0?

2000-08-15 Thread Stephen Mills


>What to do? Any recommendations from Penguinillas that have trod this path
>before me? (I'm growing a beard...)
>- Jeff

I found that mt didnt work very well for me - for some reason I would get
corrupted files and errors reading my backup device - im running a scsi dat
4 gig tape backup unit at home and backing up with Perfect Backup
http://www.unisrc.com/

About 6-12 months ago they went commerical but were offering their previous
product for free for personal use, i just checked out the site and they
still offer it but have taken down the free linux version, (other *nix avail
still) probably because it was so good, i still have the free version, works
with X, and an excellent console version ! scheduled backups etc all in an
interface, the retail version is still pretty cheap tooaround $70US

--Steve


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Kylix - is it available yet?

2000-08-17 Thread Stephen Mills

a collegue that i met on a course was lucky enough to get a beta copy - i
don't know if its available yet, but he was very impressed with it, his been
developing delphi since version 1

--Stephen

-Original Message-
From: Rick Welykochy [mailto:[EMAIL PROTECTED]]
Sent: Thursday, August 17, 2000 6:07 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] Kylix - is it available yet?


Any Sluggers know if Borland's Kylix is available for Linux?

A promotional CD I have contains everything Linux from Borland
but Kylix. And their web site talks the project up, but there is
no sign of a download yet.

--
Rick Welykochy || Praxis Services Pty Limited


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] interesting rumour

2000-08-17 Thread Stephen Mills

I think you will already find versions of IE / Bleak Outlook Express for
HP-UX and Solaris - although Ive never used them

Who wants a verison of MS Office for Linux ? Id rather see Star Office
succeed and keep the bloatware out :)


-Original Message-
From: Arthur Barton [mailto:[EMAIL PROTECTED]]
Sent: Friday, August 18, 2000 9:21 AM
To: [EMAIL PROTECTED]
Subject: [SLUG] interesting rumour


>From net-new:
WinInfo says Microsoft has been working for the past year with a company
called Mainsoft to port Windows to Linux and, possibly, other versions of
UNIX. It's apparently a complicated business to get Windows applications to
run normally in Linux, and so far results are disappointing. See
http://www.wininformant.com/display.asp?ID=2874

hmmm



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] OT: interesting rumour

2000-08-17 Thread Stephen Mills

I just read an interesting article from PC User September Issue, detailing
some of the features of OSX, one thing I have learnt is that the 'X'
represents Version 10, not "Ecks" You can find the article on page 108. It
mentions "Cocoa" as the API's you mentioned below

Cheers,
Stephen



Not true.  IE for OSX is written to the Carbon or Cocoa APIs (Carbon if I
remember correctly), not to the BSD APIs.  Carbon and Cocoa have more in
common with Windows than they do with Unix so your assumption that OSX being
based on a Mach/BSD kernel means IE would be talking to Unix APIs is quite
wrong.




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Linux Certification

2000-08-20 Thread Stephen Mills

>What I am concerned about is that if by doing a RedHat endorsed course, I 
>would be exposed to RedHat tools that wouldn't apply to Un*x in general, 
>such as the rpm system and GUI-based configuration tools.
>Any experiences with these programs would really be appreciated.

Ive done the RHCE - yes you must know RPM, I can't go into too much detail
(or Redhat will give me a boot up the behind) but if you know your Linux
very well, you shouldnt have much trouble

Good thing is, you can use any tools on the standard Redhat distro you wish
to get the job done - thats all I can really say :)

Stephen


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Re: Linux Certification

2000-08-21 Thread Stephen Mills

>The people sending .doc files were definately at a disadvantage, however
>the format of their CV was not the determining factor in choosing who 
>got the job.

Last week I sent my resume to an agency for a position in .txt form, and
they didn't know how to open it. I had to resend in .doc.

--Steve


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Linux Certification

2000-08-21 Thread Stephen Mills

Who knows, "bad stuff" (revoke your quals?)- they make you sign a contract
not to disclose the exam, I guess to protect its integrity, I personally
think its a good idea, keeps the playing field equal.

We all know that MCSE exams aren't based on real life stuff, I bet more then
half don't know how to recover a BSOD on startup without reinstalling in
situations where it wouldnt be necessary.

--Stephen

> Stephen Mills wrote:
> 
> Ive done the RHCE - yes you must know RPM, I can't go into too much detail
> (or Redhat will give me a boot up the behind) but if you know your Linux
> very well, you shouldnt have much trouble


>RedHat will do WHAT?

>Reminds me of a couple of database vendors who refuse publication of
>benchmark results.



>- Jeff



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Linux Certification

2000-08-22 Thread Stephen Mills

RHCE has is a multiple choice section as well that is structured in the
manner your describing

--Stephen

On Tue, Aug 22, 2000 at 06:44:52PM +1000, Jon Biddell wrote:
> 
> Any organization that does its' testing this way is sus in itself -
> have they never heard ot parametric testing ?  Have a pool of, say,
> 10,000 questions and randomly select 100 for the test - no two tests
> are the same.

not if the testing is hands on

Conrad.


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] VFS can not mount root

2000-08-23 Thread Stephen Mills



I booted the RH rescue CD but my drives were not in /dev How do I create
these so that I can mount them and check what the problem is?



mknod /dev/hda

take note that while in rescue mode, vi is not available, only pico, I
believe this is due to the size difference, vi being larger, if you want vi,
mount your filesystem

you might get the 'VFS cant mount root' if your root= line points to a wrong
fs in your /etc/lilo

If you want to rerun lilo config and your kernel is on /boot then you'll
need to chroot it before lilo will run.

Have fun
Stephen


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



FW: [SLUG] VFS can not mount root

2000-08-23 Thread Stephen Mills

whoops, /etc/lilo.conf - im half asleep it seems

=


I booted the RH rescue CD but my drives were not in /dev How do I create
these so that I can mount them and check what the problem is?



mknod /dev/hda

take note that while in rescue mode, vi is not available, only pico, I
believe this is due to the size difference, vi being larger, if you want vi,
mount your filesystem

you might get the 'VFS cant mount root' if your root= line points to a wrong
fs in your /etc/lilo

If you want to rerun lilo config and your kernel is on /boot then you'll
need to chroot it before lilo will run.

Have fun
Stephen


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] VFS can not mount root

2000-08-23 Thread Stephen Mills

take a look at your file system table
fdisk -l /dev/hda
I doubt there is any problem with your filesystem in regards to corruption,
so running e2fsck won't do much - at most your partition might be there, but
all your data maybe have jumped ship.

I can only see you to have this symptom if your trying to mount the extended
partition, which would come up with nothing, but since you said /dev/hda5 I
can't see this as possible as an extended table can not be hda5, hda5 is the
first logical partition

[root@proxy /root]# fdisk -l /dev/sda

Disk /dev/sda: 255 heads, 63 sectors, 553 cylinders
Units = cylinders of 16065 * 512 bytes

   Device BootStart   EndBlocks   Id  System
/dev/sda1 1   102819283+  83  Linux
/dev/sda2   103   553   3622657+   5  Extended
/dev/sda5   103   437   2690856   83  Linux
/dev/sda6   438   539819283+  83  Linux
/dev/sda7   540   552104391   82  Linux swap

In my case, if I tried to mount /dev/sda2 it would come up with nothing

anyone else have any ideas ?

Stephen

==

I did a 
mknod /dev/hda
and 

mknod /dev/hda5 

but when I mount it it comes up with nothing. 
or should I do an fdisk ?




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Help with conf.modules

2000-08-23 Thread Stephen Mills

>After a reboot, I have two problems. One is that neither of the NIC's is 
>brought up at startup, despite these lines in /etc/conf.modules:
>alias eth0 eepro100
>alias eth1 ne
>options ne io=0x240 irq=3

you need to check to make sure the "ONBOOT=" flag is set correctly (to yes)
in /etc/sysconfig/network-scripts/ifcfg-eth0 (& eth1) - this is a redhat
thing, I don't know
about other distros.

>I know that the 8390 module is required to use the ne, so I am suspecting 
>that it has to be inserted into this file for the ne to be brought up 
>successfully, maybe appended to the second line? However, this still
doesn't 
>explain the eepro100 not being loaded in.

nope, linux will bring up the interface intelligently and load any modules
it depends on, I think it uses modprobe to do this, but I havent check the
init script. so basically you don't need to specify 8390. I know insmod
isn't that smart, but modprobe is.


>My second problem is that once I do insert the modules manually, I can 
>configure both cards and see the network, but I can't get dhcpcd to assign 
>the ne's IP address. After running it:
>dhcpcd -h  interface eth1
>it seems to run and exit (or fork(), whatever it does) successfully, but 
>checking ifconfig shows the interface remains unchanged. Mind you, this 
>command worked flawlessly previously.

If your using Redhat's way, it uses a program called kudzu to probe for new
hardware, so when you installed the new nic, it might of botched your
installation slightly, Ive found that it doesnt do everything correctly
especially configuring nics. You should see if your external cable modem nic
is configured to accept DHCP requests in the above eth0 file, apart from
that, I'm unsure.

--Stephen


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] hp openmail on linux

2000-08-23 Thread Stephen Mills

I downloaded most of the files required, havent had time to look into it
further, but its not a cheap solution either, its a little cheaper then
Exchange, but not by a great deal, and there is no HP support for the Linux
port yet

Cheers,
Stephen

-Original Message-
From: Craige McWhirter [mailto:[EMAIL PROTECTED]]
Sent: Friday, August 25, 2000 12:27 AM
To: David Kempe
Cc: Slug
Subject: Re: [SLUG] hp openmail on linux


Umm no but we're also considering it so if you do give it a whirl,
feedback would of interest ;)

David Kempe wrote:
> 
> Anyone tried it?
> Got any comments on setting it up using it?
> 
> I'm willing to give it a go if its a viable alternative to exchange then
im
> there.
> 
> thanks,
> 
> Dave
> 
> __
> solutionsFirst.net Consulting
> http://solutionsfirst.net
> Ph: (02) 9413 9604
> Fax: (02) 9413 9719
> Mob: 0413 022 143
> 
> --
> SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
> More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Help with dhcpcd (was: Help with conf.modules)

2000-08-24 Thread Stephen Mills

>>You should see if your external cable modem nic
>>is configured to accept DHCP requests in the above eth0 file, apart from 
>>that, I'm unsure.

>Yes, it is set, but I've still had no success with dhcpcd. Any other 
>suggestions anyone? Apart from the NIC being configured and cable modem 
>being physically connected to it, what else is there other than to issue
the 
>command:

>>dhcpcd -h  interface 

You may or may not be aware that Redhats Pump DHCP client does not work very
well with cable modems, here is an extract from Trinity OS - all copyrights
from the paste belong to David A. Ranch, [EMAIL PROTECTED]

You have indicated your using dhcpdcd so I guess this doesnt pose as a
problem, although I suggest you goto his website and downloading his
excellent documentation on setting up firewalls, cable modems, security and
the like - it might give you a idea on where your going wrong.

http://www.ecst.csuchico.edu/~dranch/LINUX/


Cheers,
Stephen




 # NOTE: Red Hat users of DHCP to get TCP/IP addresses (Cablemodems, DSL,
etc)
  #   will need to install and use a different DHCP client than the
stock
  #   client called "pump".  One recommended DHCP client is called
"dhcpcd"
  #   and can found in Appendix A.
  #
  #   The stock Red Hat DHCP client doesn't allow the ability to have
scripts
  #   run when DHCP gets a TCP/IP address.  Specifically, DHCP delves
out
  #   TCP/IP addresses to its clients for a limited amount of time; this
  #   called a "lease".  When a DHCP lease expires, the client will
query the
  #   DHCP server for a lease renewal.  Though the DHCP client will
usually
  #   get back its original TCP/IP address, this is NOT always
guaranteed.
  #   With this understood, if you receive a different TCP/IP address
than
  #   the IPCHAINS firewall was configured for, the firewall will block
ALL
  #   network access in and out of the Linux server because that was
what it
  #   was configured to do.

===


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Syslogd

2000-08-29 Thread Stephen Mills

you will have to tell us which messages are coming up so we can try and
suppress them

Cheers
Stephen

-Original Message-
From: Darren Noonan [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, August 30, 2000 12:28 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] Syslogd


I'm running RH6.2, the problem I have is that certain system messages are
going to the console.

I have change the syslog.conf to try and stop it but nothing seems to work.
I monitor this machine remotely and need some of the info in the logs and
not the console.

any ideas


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] NT err NDS and PAM module

2000-09-04 Thread Stephen Mills

Hello Grahame

Ive used NT's domain model for authentication from Linux, but not NDS.

There is NDS for Linux, I believe this is a start, I suggest going to
http://www.novell.com/products/ndsas/ but I don't know if its in a PAM lib
:) Be expected to pay for it though !

--Stephen

-Original Message-
From: Grahame M. Kelly [mailto:[EMAIL PROTECTED]]
Sent: Monday, September 04, 2000 9:09 PM
To: SLUG
Subject: [SLUG] NT err NDS and PAM module


Hi Sluggers.

I have a school that needs to Authenticate its school students from
a Linux system under AUC using PAM. 

Has anyone used/aware of which pam module to use (if one exists,
where), and if you experienced any issues with NDS and PAM modules.

Thanks Grahame.

Member SLUG (Sydney Linux User Group) http://www.slug.org.au 




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Removing duplicate entries from a file

2000-09-05 Thread Stephen Mills

I used to use pwebstats to generate sites that my users had browsed, it can
be a very detailed tool, with graphs on usage and is nicely HTML'ised.

http://martin.gleeson.com/pwebstats/

It requires Perl and a coupla other little things.

Oh (looks at address) say hello to Breeanna for me.

--Steve


-Original Message-
From: MacFarlane, Jarrod [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 06, 2000 9:43 AM
To: '[EMAIL PROTECTED]'
Subject: [SLUG] Removing duplicate entries from a file


Hey sluggers,

I need to look at a particular machines web hits.. I am currently using:

cat /usr/local/squid/logs/access.log |grep 1.2.3.4 |cut -f4 -d"/" >
logfile.txt

This outputs something like:
www.reallynaughtysite.com
www.smackmeimbad.com
and so on

The problem is that it has many double ups... are there a long confusing
string of commands that will go through my logfile and remove all but one
instance of every domain listed?

Thanks,
Jarrod.


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Q: Using SQUID to authenticate internet users ?

2000-09-05 Thread Stephen Mills

If you refer to the squid.conf file there is a section on authenication, you
can authenicate off /etc/passwd or an NT domain, among other things..

The only thing you're going to have to watch for is if your users click to
save the user/pass when it asks to authenicate, if they save the settings,
then your efforts arewell futile :) (Unless you change their registry
not to allow this)

Stephen

-Original Message-
From: Matt Hyne [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 06, 2000 1:53 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] Q: Using SQUID to authenticate internet users ?



Folks,

My father has a small business network with about 12-15 users and a
dial-on-demand internet connection running from a Linux box to his local
ISP.

What he would like to try and do is require PC users wanting to use the web
to have to authenticate before they can access any external webpages.  This
is because he only wants a select number of staff to have web access as it
is expensive and uses bandwidth.

Now, the first thing someone is going to say is "why don't you block the ip
addresses of the un-authorised users' PCs" - well the staff do a lot of
hot-desk work where they will not be using the same PC every day.  Also, in
other areas, 5 or 6 people use the same PC.

My question is - can this be done with squid (and transparently) and does
anyone have an example config that I can take a look at and try to build my
own.

I've seen commercial products that do this - but I would prefer a Linux
solution.

Thanks
Matt





--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Q: Using SQUID to authenticate internet users ?

2000-09-05 Thread Stephen Mills

Cisco run their printing system on Samba/Linux - we shouldnt complain, they
arent ;)

--Steve


-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 06, 2000 1:50 PM
To: Matt Hyne
Cc: [EMAIL PROTECTED]
Subject: Re: [SLUG] Q: Using SQUID to authenticate internet users ?


Quoting Matt Hyne <[EMAIL PROTECTED]>:


H... 

Surely CISCO have a product to do this ???..:p)


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Q: Using SQUID to authenticate internet users ?

2000-09-05 Thread Stephen Mills

my point exactly !

--Steve

-Original Message-
From: Matt Hyne [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 06, 2000 3:28 PM
To: Stephen Mills; '[EMAIL PROTECTED]'
Cc: '[EMAIL PROTECTED]'
Subject: RE: [SLUG] Q: Using SQUID to authenticate internet users ?



We probably do, but why should I buy something when I have a Linux box I can
do it on already !!!

Matt

At Wednesday, 6/09/2000 03:07 PM (+1000), Stephen Mills wrote:
>Cisco run their printing system on Samba/Linux - we shouldnt complain, they
>arent ;)
>
>--Steve
>
>
>-Original Message-
>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
>Sent: Wednesday, September 06, 2000 1:50 PM
>To: Matt Hyne
>Cc: [EMAIL PROTECTED]
>Subject: Re: [SLUG] Q: Using SQUID to authenticate internet users ?
>
>
>Quoting Matt Hyne <[EMAIL PROTECTED]>:
>
>
>H... 
>
>Surely CISCO have a product to do this ???..:p)
>
>
>--
>SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
>More Info: http://slug.org.au/lists/listinfo/slug
>
>
>--
>SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
>More Info: http://slug.org.au/lists/listinfo/slug




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] Intel 440BX2 + CDROM

2000-09-06 Thread Stephen Mills

Is anyone running an ATAPI IDE CDROM in an Intel 440BX2 Motherboard running
Linux ?

Its funny how since I changed my motherboard over a few months ago, my CDROM
is no longer recognised when the kernel boots, yes it works in Windows.

The strange thing is, I'm running Redhat 6.2, the CDROM is bootable, and it
will install Linux with no problems with the CDROM, but when you reboot, it
detects the harddisks I have, then stalls for about 5 seconds scanning the
rest of the IDE bus for devices, then continues on without detecting the
CDROM, very odd. Its definately motherboard related, I think I'm running P22
BIOS which was the latest last time I looked. I never had this problem with
my previous motherboard.

Its not a huge deal, my main fileserver has Its CDROM NFS'd out. just a
thought. 

Regards,
Stephen




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] change default mail handler

2000-09-07 Thread Stephen Mills


 thing to do that instead? Where is that info stored? How do i stop
sendmail?
I should be able to figure out if I can enable this thing or not.



on redhat:

make sure sendmail is stopped 

/etc/rc.d/init.d/sendmail stop

then un-install it

rpm -e sendmail

then install your alternative mail software

--Stephen



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Old scsi hard drive problems.

2000-09-10 Thread Stephen Mills

could have a dodgy harddisk, those 4 gig scsi micropolis drives were very
dodgythe company doesn't exist anymore either.

since this hdd is a narrow one, it only needs 8 bits terminated, like your
burner, but you might want to set active termination.

hope u got the drive for next to nix :)

--Stephen

-Original Message-
From: Arunava Sen [mailto:[EMAIL PROTECTED]]
Sent: Sunday, September 10, 2000 1:04 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] Old scsi hard drive problems.


hi,

i just got an old scsi hard drive. its a micropolis 4341NS. i try to
insmod my scsi card module. my burner is properly detected and stuff but
the hard drive causes problems.

i get (approximately):
---
sda: (scsi:0:2:0) No active SCB for reconnecting target.
issuing BUS DEVICE RESET.
---
and then i get lots of "timeout" errors.

Anyone have this hard disk or know what is going wrong?

Thanks in advance.

Arun.



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] File systems and redundency

2000-09-13 Thread Stephen Mills

you'll probably get a million other responses, but

a few reasons

* it easier to rebuild a system if all your /home data is on a seperate
system -  which ive had to do !
* if u mount under /var & /tmp under diff partitions, it protects the
systems root drive from   running out of space, ive heard that is root
fills up 100% then kernel won't boot although ive never seen this happen
personally
* on linux, you can only set quotas on seperate partitions
* not having all your cookies in one basket

--Stephen

-Original Message-
From: George Vieira [mailto:[EMAIL PROTECTED]]
Sent: Thursday, September 14, 2000 10:22 AM
To: '[EMAIL PROTECTED]'
Subject: [SLUG] File systems and redundency


Hi all,

I was just wondering why is it so crucial to have different mount points on
a unix system? Eg. create /, /usr , /tmp , /home

Why is it so much better to have multiple partitions instead of having
everything mounted as (/) root? 
Sure some times the file system could crash and at least it's only 1 file
system and root or /home or /usr is still OK but what other reasons are
there? Speed? Fragmentation? Etc.

thanks,
George Vieira
Network Administrator
http://www.citadelcomputer.com.au
PGP Fingerprint :   43DC 92AC 1A82 27B2 E97B  52F1 B60F 301A 38A9 A10C
PGP KeyID:  0x38A9A10C



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] File systems and redundency

2000-09-13 Thread Stephen Mills

"even though its a RAID 5" ?

Raid 5 is the -slowest- but the offset is its the most redundant...  (OK
don't catch me out with the RAID 5+0 etc)  :)

Hope you'll be running a decent hardware RAID Controller (!)

Stephen

-Original Message-
From: George Vieira [mailto:[EMAIL PROTECTED]]
Sent: Thursday, September 14, 2000 11:06 AM
To: '[EMAIL PROTECTED]'
Subject: RE: [SLUG] File systems and redundency 


That could be a pain. I have a +30GB database for this client to handle here
and just worried about the performance on the file system even though it's a
RAID 5.



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Proceedure for preventing 'linux single' at lilo prompt

2000-09-14 Thread Stephen Mills

1) be reminded that if a person has physical access to your linux machine,
they can usually have full access to all information within a few mins with
a boot floppy

2) you can set a password on the lilo prompt to prevent them from booting an
image without the proper password, but rule 1 still applies, check out
/usr/doc/lilo-* for more info

3) put your server under lock and key if its a problem

--Stephen

-Original Message-
From: Matt [mailto:[EMAIL PROTECTED]]
Sent: Friday, September 15, 2000 11:59 AM
To: [EMAIL PROTECTED]
Subject: [SLUG] Proceedure for preventing 'linux single' at lilo prompt


How might I disable this quick way of getting in to root ? 

Thanks ! 

-- Matt Kozera


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] MTU and flamebait

2000-09-14 Thread Stephen Mills

something wrong with the list ? I seem to have got this message four times
already.

--Stephen

-Original Message-
From: Mehmet Ozdemir [mailto:[EMAIL PROTECTED]]
Sent: Friday, September 15, 2000 12:48 PM
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Subject: Re: [SLUG] MTU and flamebait




Well a Visual Basic Users list wouldn't be the best place to ask a windows
networking question, kind of like asking a php question in a perl list,
perhaps one of the windows.networking or windowsnt.networking newsgroups is
a better. 




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Strange Kernel Compilations

2000-09-18 Thread Stephen Mills

Hi Arthur

When you make the kernel use ;

make bzImage

(instead of vmlinuz)

Its not a compressed version and may look the same-ish in size, but lilo
will load the kernel differently

alternatively you may try compiling all your possible options as modules (as
redhat kernel rpm does anyway)

Cheers,
Stephen


-Original Message-
From: Arthur Barton [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 19, 2000 2:00 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] Strange Kernel Compilations


Hey there,
   i seem to be having the strangest problems with re-compiling my kernel.
Machine in question is a relatively standard RedHat 6.2 (Zoot) install,
which i have had on other machines without problems.
The actual problem being, whenever i recompile my kernel (RH6.2 standard
2.2.14) or even trying it out with 2.4-test4(/7)(/8) it compiles fine, but
when i add it to lilo i get helpfull messages (on wrong box at the moment so
havent got exact error :-( ) something like:
Kernel size is too large.
Does this mean anything to anyone else?
never encountered this before, the image size is *smaller* than the default
kernel.
Idea's?

Arthur



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Debian

2000-09-18 Thread Stephen Mills

Why should Linux be hard to use ?

Although I believe that a famous programmer once said that there is two
types of interfaces, a user interface and good interface, its clear that for
Linux to take off on the desktop, it needs to be also user friendly.

If Linux was difficult to install, manage, administer I hardly doubt you
would see the Linux craze come about or the "slug-fests" be as popular.

I'm hoping for the most users that use Windows, should be able to make the
transition over to Linux without too much trouble. They should be able to
add a ppp0 interface without fiddling with chat scripts.

Sure, leave the administrators to the backend, fiddling with scripts, its
what I like doing also, because I'm *interested* in it, but for users that
just want the job done, they should be able to do that also. If it wasn't
for some of the front end work happening I think you'll find most people
will stick to Windows on the desktop.

Im running Helixcode on my desktop and think its fantastic. My only gripe
with it was it crashed about 10 times (seriously) when it was grabbing the
RPM's, luckily it was saving them to my local drive before it would run the
actual installation.

I think the reason why front ends get a rough deal is that they are usually
trying to calcuate what is happening at the core of the programme, this can
fail miserably, would you run linuxconf over your mailhub of 50,000 users ?
I certainly wouldn't. But for the most part its an important step to get the
numbers of users. As far as I'm concerned, if the OS is successful,
companies will start supporting it, commerically, and supply their own
drivers. Its already happening in a big way and is only going to better. I'm
certainly looking forward to the future of Linux !

Stephen

-Original Message-
From: Dean Hamstead [mailto:[EMAIL PROTECTED]]
To: Sydney Linux Users Group



Not to brag, but to a certain degree i feel a line must be drawn. I
think things
can become *over* simplified and *over* centralized. If people want to
point and 
click maybe free unixs arent the best solution for them. They might just
have to
live with NT (or whatever) for now.

People should be more serious about their abilities and be content with
a solution
that best fits their needs. If you want linux and you cant work it out,
hire someone.
It will still probably work out cheaper (especially for people looking
for firewall




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] keyboard timeout

2000-09-18 Thread Stephen Mills

Its only a cosmetic warning message, but when my newly compiled 2.2.17
boots, I get 'keyboard timeout' in the middle of the message output kernel
bootup. It doesn't seem to affect the keyboard, only pauses for about 3
seconds before continuing, the default kernel that was supplied on Redhat
didn't seem to have this warning message.

Any ideas what would cause this, maybe there is a keyboard option I would
need to tweak ? Never seen this before...

Stephen


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] squid and/or routing problem

2000-09-19 Thread Stephen Mills

I think defining the routing via ACL's might be a little tricky and not
something I know thats supported by default :)

I don't know if it will work, but what about 2 instances of squid running on
different ports ? (one for each company)
with each instance of squid following a different routing rule

Cheers
Stephen

-Original Message-
From: John Ferlito [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 20, 2000 9:40 AM
To: David Kempe
Cc: Slug
Subject: Re: [SLUG] squid and/or routing problem


On Wed, Sep 20, 2000 at 10:01:24AM +1000, David Kempe wrote:
> Situation:
> Office, shared by two companies. Company A has sucky Telstra ISDN 64K
> through cisco 801 router at 192.168.0.1.
> Company B gets fed up with sharing this bandwidth amongst 10 others and
> wants to split. Linux box has squid on it running as (opt in)proxy for
both
> companies.
> Modem on linux box. Comp B sets linux box as default gateway, masquerades,
> traffic flows out modem and back in sweet. Problem is all the company
squid
> traffic also goes out the modem.
> How do i set up squid ACLS so a limited amount of source IPs only goes out
> the modem and the others go out the router at 0.1 ? Or is an ipchains
> question? Can you even do this?
> So I want squid to get traffic from some ips from one default
> gateway(router), and traffic from other ips from another default gateway
> (modem).
> Or (cop-out) should we just cut company A loose from squid and let em use
no
> proxy over the ISDN?

Hmm I don't think you can do this with squid itself. Which means 
it's pretty hard to do. You can run two squids on different ports using
different parent caches that get routed differently. But they wouldn't
be able to share the same cache.

You can't really use ipchains or any similar routing thing
either because it's squid that makes the request and so once it hits
squid there's noway of working out which network it came from.

Sounds like interesting bit of functionality to hack into squid
though.

-- 
John


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Maxtor MaxAttach

2000-09-20 Thread Stephen Mills

we sell Snapservers www.snapserver.com its based on BSD(i) and apparently
they say they are the fastest with models up to 240 gig rack mounted

but be aware the snapserver's NFS support works, but isnt that great, you
can't set permissions to like 700, has to be 755, maybe they will fix this,
makes its useless for ppl that want to mount stuff under /home, but good for
ppl that just want to access the server from UNIX via NFS

Id say maxattach has the same features *shrug*, afaik all of the NAS servers
operate at a share level, not file level...im told this may change next year
for the snapserver.

--Steve


-Original Message-
From: Jill Rowling [mailto:[EMAIL PROTECTED]]
Sent: Thursday, September 21, 2000 10:02 AM
To: Slug@slug. org. au (E-mail)
Subject: [SLUG] Maxtor MaxAttach


http://www.maxtor.com/products/MaxAttach/MaxAttachExplorer.htm

Has anyone actually used one of these things? They don't say what's inside
it (apart from it's a pentium with lots of storage) but I figured it's
either running Linux or FreeBSD.
They don't say what its continuous throughput is either.

Cheers,

Jill.

___
Jill Rowling
Snr Design Engineer & Unix System Administrator
Electronic Engineering Department, Aristocrat Technologies Australia
3rd Floor, 77 Dunning Ave Rosebery NSW 2018
Phone:  (02) 9697-4484  Fax:(02) 9663-1412
Email:  [EMAIL PROTECTED]
 



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] AHA1542

2000-09-20 Thread Stephen Mills

In redhat (and I presume Debian) if you install in Expert Mode (by typing
expert at install boot prompt)you can select the IRQ/DMA for each device
yudda yudda

--Stephen


-Original Message-
From: David Fisher [mailto:[EMAIL PROTECTED]]
Sent: Thursday, September 21, 2000 12:07 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] AHA1542


G'day all,

I am attempting to install Linux (either Debian potato or RH6.2, whatever I 
can get to work) on an old 486 fitted with an Adaptec 1542CP SCSI adapter.


Now, I thought this card was supposed to be supported but neither the Debian

or Redhat installers will recognise it.

I am thinking that maybe what I need to do is pass appropriate boot 
parameters, which seems to be what the RH hardware compatibility page is 
saying but I don't know how to concoct the correct numbers.

First of all, can anyone tell me definitely whether both/either installers 
support this card?  If so what would the boot time argument be?

The card is currently on base address 130, IRQ 11 and DMA 5.  I don't even 
know if these are appropriate, I'm not much of an ISA person.

Help!

-- 
David
mailto:[EMAIL PROTECTED]
mailto:[EMAIL PROTECTED]


Note:  No Microsoft programs were used in the 
creation or distribution of this message.
If you are using a Microsoft program to view this message, 
be forewarned that I am not responsible
for any harm you may encounter as a result.





--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] qmail and bulk mail

2000-09-24 Thread Stephen Mills


>i want to send the same message to many ppl. i wanted to test this by
>echo "This is a test" | /var/qmail/bin/qmail-inject -fdougb-sender@host1 $s
>from host2 and where $s is "dougb-1@host1 dougb-2@host1 ...
dougb-2500@host1"
>now. qmail tries to send to all aliases at once through a seperate smtp
>connection. how do i stop this to force it to send it through the one
>connection?
>the logs have
>deferral: Sorry,_I_wasn't_able_to_establish_an_SMTP_connection._(#4.4.1)
>all the way through it.

afaik this is how smtp should work, qmail will execute qmail-remote to send
each email to its remote host, even though its to the same host.

from memory qmail has a concurrency limit of 20 seperate connections set by
default, create a file concurrencyremote in control directory and put a
numeric value to represent max smtp connects outgoing.

you might want to just try one email to make sure your system/dns is working
/var/qmail/bin/qmail-inject dougb-@host1

man qmail-send for more info

Stephen




--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] ozemail problems.

2000-09-24 Thread Stephen Mills

hey we shouldn't talk, ours is SLUG afterall :)

--Stephen

-Original Message-
From: Conrad Parker [mailto:[EMAIL PROTECTED]]
Sent: Monday, September 25, 2000 2:53 PM
To: Paul Robinson
Cc: Rick Welykochy; [EMAIL PROTECTED]
Subject: Re: [SLUG] ozemail problems.


On Mon, Sep 25, 2000 at 02:54:15PM +1100, Paul Robinson wrote:
> Rick Welykochy wrote:
> 
> > OutLook Express is not a connection protocol - it's an application.
> > 
> > This is a Linux User's Group List - contact Micro$oft for this
> > kind of help, or join a Microsoft Users Group.
> > 
> Yes, the Sydney Microsoft Users Group (SMUG) :)

or MUGS, more likely ;)

Conrad.


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] ozemail problems.

2000-09-24 Thread Stephen Mills

>>hey we shouldn't talk, ours is SLUG afterall :)
>Yes, but that could be SLUG as sluggish, or SLUG as in pow!

 maybe to some, but i personally see it as a slug, for an example, goto
www.openldap.org :)



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] win4lin

2000-09-25 Thread Stephen Mills

Maybe you havent tweaked your virtual machine. I think by default it selects
1/3 of your ram by default for the VM, although I could be very wrong.

Networking definately works, and 128 meg is bare minimum I'd consider
running VMware.

As for the jerkiness during intensive disk operation, make sure you have
loaded the VMware Display drivers which increase the refresh rate in X /
Windows in the VM by 300%. You can also try providing the VM with more RAM
to move. Might be an idea to watch the CPU Utilisation when your machine
slows to see what is causing it.

Hope this helps

--Stephen



-Original Message-
From: Bernhard Lüder [mailto:[EMAIL PROTECTED]]
Sent: Monday, September 25, 2000 11:53 PM
To: 'Bill Hiley'; [EMAIL PROTECTED]
Subject: RE: [SLUG] win4lin

Hi,

I have just been playing with VMware Eval License and I found it realy easy
to install and doco was great as well. However, despite 700Mhz and 128MB, I
find the performance quite bad. Lots of jerkiness especially during disk
intensive operation. Maybe it's not quite correctly tweaked yet, but I will
have another go anyway.

However I also had a go at Win4Lin a while ago and found that quite easy to
install as well (except for the Kernel patch) and performance was definitely
better. However no networking at that stage (Is there networking support
now?).

Price for VMware is also high at US$299! ~A$550 and you also need your
windows license.

So you will need to consider what you want to do. If you are steering away
from MS you might be better off with WIN4LIN, but if you need full support
of Windows (perhaps even 2000 and NT, which supported by VM, but not yet by
WIN4LIN) go for VMware.

Evaluate them both. They let you download for free.

Bernhard Lüder
ICQ 26070583


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Help with dhcrelay

2000-09-25 Thread Stephen Mills

maybe im a little confused, but any of these machines running dhcpd ? or
linux is acting as the client ?
maybe a silly question, but do you have the multicast route set on the Linux
box ? I haven't setup DHCP relays on Linux before, but its just a thought,
although this route is not required if its just a dhcp client.

Stephen


-Original Message-
From: Howard Lowndes [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 26, 2000 3:56 PM
To: Mail List - SLUG
Subject: [SLUG] Help with dhcrelay


I am trying to get dhcrelay to work between 2 subnets.

The agent sits on subnet A and hears the DHCPDISCOVER packets OK (they are
from NT boxen, BTW) and forwards them on to the DHCP server on subnet B
(again an NT box).  The server comes back with a DHCPOFFER to the agent,
but the latter fails to forward this back to the requesting client's MAC
address, it just ignores it.

Looking at the packets with tcpdump they look fine, in both directions,
and forwarding is set on, but I do not think that would be an issue here
anyway.

There used to be a hassle with Linux and the ultimate broadcast IP.  Is
that still a hassle and how do I set it up under linuxconf (preferably)

All and any assistance appreciated.

-- 
Howard.
__
LANNet Computing Associates 



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Help with dhcrelay

2000-09-25 Thread Stephen Mills

yes im talking about multicast (224.x.y.z) although in this case you won't
need that. I found that I needed this set if the Linux box was acting as the
DHCP Server.

You're description on DHCP is correct, they obviously run on UDP
ports...sorry I don't have any suggestions for you

goodluck :)

Stephen



-Original Message-
From: Howard Lowndes [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 26, 2000 4:34 PM
To: Stephen Mills
Cc: Mail List - SLUG
Subject: RE: [SLUG] Help with dhcrelay


The Linux box on subnet A only has the dhcrelay agent.  It acts like a
server on its resident subnet, but actually forwards the request on to a
real server on another subnet.  When that server comes back with a
response then the agent passes it on to the original client, which thinks
that it came from a server on the local subnet.

Both the clients and the server are on NT boxen, the relay agent is on a
Linux box acting principally as a router.

What are you calling the multicast route in this, surely not 224.x.y.z
etc?  The client sends the DHCPDISCOVER with source 0.0.0.0:67 and
destination 255.255.255.255:68.  The agent grabs this and forwards it to
the server on port 68.  The server comes back to the agent with the
DHCPOFFER on port 68 and the agent should then shoot it back as source
0.0.0.0:68 destination 255.255.255.255:67 but put onto the ethernet at
level 2 using the MAC that was in the original DHCPDISCOVER packet.  I
think thats the way it works.

-- 
Howard.
__
LANNet Computing Associates <http://www.lannet.com.au>

On Tue, 26 Sep 2000, Stephen Mills wrote:

> maybe im a little confused, but any of these machines running dhcpd ? or
> linux is acting as the client ?
> maybe a silly question, but do you have the multicast route set on the
Linux
> box ? I haven't setup DHCP relays on Linux before, but its just a thought,
> although this route is not required if its just a dhcp client.
> 
> Stephen
> 
> 
> -Original Message-
> From: Howard Lowndes [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, September 26, 2000 3:56 PM
> To: Mail List - SLUG
> Subject: [SLUG] Help with dhcrelay
> 
> 
> I am trying to get dhcrelay to work between 2 subnets.
> 
> The agent sits on subnet A and hears the DHCPDISCOVER packets OK (they are
> from NT boxen, BTW) and forwards them on to the DHCP server on subnet B
> (again an NT box).  The server comes back with a DHCPOFFER to the agent,
> but the latter fails to forward this back to the requesting client's MAC
> address, it just ignores it.
> 
> Looking at the packets with tcpdump they look fine, in both directions,
> and forwarding is set on, but I do not think that would be an issue here
> anyway.
> 
> There used to be a hassle with Linux and the ultimate broadcast IP.  Is
> that still a hassle and how do I set it up under linuxconf (preferably)
> 
> All and any assistance appreciated.
> 
> 



--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] pcmcia network install

2000-09-26 Thread Stephen Mills

If you use the pcmcia.img image file, you can install on a laptop, configure
your nic, and install over NFS, FTP or HTTP.

This is the image you would want to use if you have no CDROM in your laptop,
and you need to access the ISO files using your pcmcia network controller.

Although if you have a CDROM drive in a laptop, you would use the standard
boot.img and configure your laptop.

This is redhat 6.2 im talking here, I don't know if its changed with 7.0

Cheers,
Stephen


===

-Original Message-
From: Sonam Chauhan [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 27, 2000 11:01 AM
To: slug
Subject: [SLUG] pcmcia network install


I wanted to install Linux on a laptop with pcmcia networking 
and a floppy drive (no CDROM). My only option seems to be a 
network installation. 

I looked at Redhat's boot images - their pcmcia and network install 
versions seem to be mutually exclusive. 

Is there a distro that combines pcmcia with networked installation
on a boot image? 

Regards,
Sonam


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



RE: [SLUG] Execute a file

2000-09-26 Thread Stephen Mills

create a new user, and try with that :)

--Stephen

-Original Message-
From: Marshall, Joshua [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 27, 2000 1:19 PM
To: [EMAIL PROTECTED]
Cc: Sydney Linux Users Group
Subject: Re: [SLUG] Execute a file 


> Permissions on ./test.pl? Location of perl compiler really
> /usr/bin/perl?  Permissions on /usr/bin/perl? ./test.pl created with
> DOS/Windoze editor that puts carriage returns in so that what you are
> really trying to execute is /usr/bin/perl^M? (The last one is my bet.)

Permissions are 777, /usr/bin/perl permissions are 755, test.pl was 
created in linux.

I do have a further development. I decided to test something else, so I 
created a file test.sh:

[marshallj@datasrv bin]$ cat test.sh
#!/bin/sh
echo test
[marshallj@datasrv bin]$ ls -l test.sh
-rwxr-x---   1 marshall marshall   20 Sep 27 03:21 test.sh
[marshallj@datasrv bin]$ ./test.sh
bash: ./test.sh: Permission denied
[marshallj@datasrv bin]$ sh test.sh
test

It almost seems as if my shell isn't using the first line to determine 
what to use to execute the file. Now this is weird! Any thoughts?

Cheers,
Josh.


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] FD limit - 2.0.x

2000-09-28 Thread Stephen Mills

Hi,

Can someone point me to a patch for the 2.0.x kernel to increase the FD
limit on the kernel ?

Cheers,
Stephen


--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] core dump fun

2000-11-21 Thread Stephen Mills

Hi

I have a ghost custom install of Redhat 7.0 on my
laptop, I have noticed one thing that is a little
strange, I try to run chkconfig and it core dumps
everytime. I tried recompiling the rpm package from
source, but it still core dumps. I know how to
manipulate the startup scripts manually this is not
what I'm asking.

I have got a core dump file, what tools can I use to
trace the cause of this problem using the core file? I
have tried another laptop, exactly the same (its a dell
latitude p500 192 meg ram) and it seems to be ok.

No need to give me step by step instruction, just a
little guidance.

Cheers
Stephen


__
Get your free Australian email account at http://www.start.com.au



-- 
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



Re: [SLUG] core dump fun

2000-11-21 Thread Stephen Mills

Hi Malcolm

Thanks for your help. The file is stripped, although
I'll have a play with it.

Stephen

Original message from: Malcolm Tredinnick
<[EMAIL PROTECTED]>
>
>On Wed, Nov 22, 2000 at 09:59:00AM +1000, Stephen
Mills wrote:
>> I have got a core dump file, what tools can I use to
>> trace the cause of this problem using the core file? I
>> have tried another laptop, exactly the same (its a dell
>> latitude p500 192 meg ram) and it seems to be ok.
>
>Assuming the chkconfig binary contains debugging
information -- run 'file
>chkconfig' and hope that it says 'not stripped' -- you
can use gdb, the binary
>and the core file.
>
>Run 'gdb /sbin/chkconfig core' and then 'bt' inside
gdb will show you a
>backtrace of the point at which it crashed, for example.
>
>Cheers,
>Malcolm
>
>--
>Malcolm Tredinnickemail:
[EMAIL PROTECTED]
>CommSecure Pty Ltd
>
>

__
Get your free Australian email account at http://www.start.com.au



-- 
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] RPM Fun

2000-12-05 Thread Stephen Mills

Hi, 
If anyone has any idea, feel free to give me a yell.

Redhat 6.1, sendmail has been cleanly removed, want to install new
version. I did a search on the web, with no cluesyes the rpm's
were downloaded in bin mode.

[root@davros nighty]# rpm --rebuild sendmail-8.11.0-8.src.rpm
Installing sendmail-8.11.0-8.src.rpm
only packages with major numbers <= 3 are supported by this version of
RPM

[root@davros nighty]# file sendmail-8.11.0-8.src.rpm
sendmail-8.11.0-8.src.rpm: RPM v4 src i386 sendmail-8.11.0-8

[root@davros nighty]# rpm --version
RPM version 3.0.3

[root@davros nighty]# rpm -U rpm-4.0-4.i386.rpm
only packages with major numbers <= 3 are supported by this version of
RPM
error: rpm-4.0-4.i386.rpm cannot be installed

--Stephen



__
Get your free Australian email account at http://www.start.com.au



-- 
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug



[SLUG] RPM Fun - Solved.

2000-12-05 Thread Stephen Mills

Hi Guys,

Thanks a bunch for your suggestions, upgrading to RPM 3.0.6 solved my
problem. I would of never considered this process.

Hehe sendmail is complaining about compiling with missing header
files, and as Dave correctly pointed out the compiled RPM version
complains about glibc. No matter I'll figure it out.

I found it quite frustrating in Redhat's approach in upgrading RPM, as
another person pointed out, probably to prompt more sales.

Thanks again,
Stephen


__
Get your free Australian email account at http://www.start.com.au



-- 
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug