Mandrake sucks. I get a mess of warnings any time I start
a Gtk app because of the weird stuff they've done with their theme.
[Heather]
I can give you my DarkGems gtk theme, would you like it?
I carefully mucked with it till it looked good on gimp and
some other gtk apps that eluded correct behavior under some themes.
I have plenty of themes, it's that Mandrake is set up to look
for their own theme, and I can't get it to stop. My GTK package is
libgtk+-x11-2.0_0-2.5.2-0.1gpw.
[Heather]
ohhh. I might be able to help you with that sometime.
If we do, then we can make a nice tip out of it.
Or if one of you readers knows the right answer, we'll cheerfully pub
what you send us after Jimmy improves his life with GTK
-- Heather
I think it's compiled in.
I've been through the fscking mess of a config layout they
have several times and can't get it to stop looking for that
bloody theme.
[Heather]
You cant just move their theme to another name and copy your
theme into the name it's reaching for? </hack type=cheap>
I have seen ur page for the sendmail configuaration and goes fine ,
And i have tried to configure the sendmail as u said , but i have some
problem regarding the configuration .
Iam trying to configure it as SMTP AUTH client for the mailserver,But
its giving error msg as DNS:service unavailable , i don't understand
what is missing in the configuration.and one more things is its not
alaising the username to the mailid .
Hi,
Firstly I would like to congratulate you on a nicely written article. I
found it very useful.
In addition to the ways you have pointed out in your article there is
another method we can use to sync linux clocks to a central server. I am
talking about using rdate (http://freshmeat.net/projects/rdate).
On the website:
"The rdate utility retrieves the date and time from another machine on
your network, using the protocol described in RFC 868. If you run rdate
as root, it will set your machine's local time to the time of the
machine that you queried."
I use the following command to sync my computers clock every day:
/usr/sbin/rdate -s tick.greyware.com
Thought I should share this with you.
Thanks,
Suramya
'rdate' is just another program using port 37 (Time protocol) which is
what RFC-868 is about. I think 'netdate' is more prevalent. In fact,
you can use shell script, because the server returns number of second
since Jan 1, 1900, in 32-bit number. So:
sec=`nc time.nist.gov 37 | od -A n -t xC | tr -d ' '`
date --date="1970-1-1 0:0:$((0x$sec - 2208988800)) GMT"
where 2208988800 is to make it relative to Jan 1, 1970.
I find that there could be an extra 32-bit number for microseconds.
So, if you change the script to:
set -- `nc time.nist.gov 37 | od -An -txC`
date -u --set="1970-01-01 0:0:$((0x$1$2$3$4 - 2208988800))"
you will use only the seconds.
You are encouraged to use UDP side of things. The shell script has no
choice but to use TCP, because knowing when to exit 'nc' is a hassle.
"[sic] folks. Debian's actual command (to do mass upgrades without breaking
holds or allowing package removals for apps that changed drastically) is:
apt-get upgrade dist -- Heather"
The command is:
apt-get dist-upgrade
I know, because I do it everyday.
I did attempt to correct a typo. I replaced it with a worse one. It
was late, and my vim went one way, my hand another. At least as far as
I can tell. Either that or I was on some amazing grade of sleep deprivation.
-- Heather
[Kapil]
You should avoid doing this for more reasons than one!
1. Since there are only three versions of Debian at any given time
you will reach the top of the charts in exactly two days! Once you
already run "unstable" (or "testing" for that matter), running:
apt-get update; apt-get upgrade
should do the trick.
I do have to say, that 'upgrade' will not help you if they drastically
rearrange a package. And, if you do it habitually you may not notice
that. I don't recommend upgrades where you don't pay attention to such
details.
The package that came to mind a long while back was vim. With merely
'upgrade' Debian would have been glad to leave me in its older style of
package splits. If only it could be told to offer the same care when a
package changes its config files under the hood.
-- Heather
2. There is more to life than keeping up with the Joneses...
[Jimmy]
Yeah, of course. There's keeping up with the Smitheseseses.[1]
[1] Why yes, I am quite drunk. Why do you ask?
The real way in which I was wrong is that neither 'upgrade' nor
'dist-upgrade' keywords break holds, and I claimed that the latter does.
D'oh!
If you're gonna throw tomatoes, let's try to completely dunk the target,
shall we?
Glad you found it helpful! I appreciate your note... even though it was
sent to a very old (and nearly unread) address of mine and thus took a
long time to come to light.
(Didi's comment refers to my "Clearing out the Master Boot Record"
article back in LG#63, which many people have found helpful over time; I
keep receiving grateful comments on it even now, years later. It's very
gratifying to know that I've been able to help so many folks with it.
The reason I'm making this reply a public one is for the new readers
of LG who may not be aware of this useful technique; the article can be
found at http://linuxgazette.net/issue63/okopnik.html .)
Perl One-Liner of the Month
Sat, 06 Nov 2004 08:11:03 +0000
Hans Wennborg (hawe from bigfoot.com)
Reply by Jimmy O'Regan
Thanks for a good e-zine!
Thanks for reading it.
Just wanted to tell you that I miss Ben Okopnik's Perl One-Liner of the
Month imensely! Those articles were what made me start loving Perl.
Woomert and Frink only get a brief cameo, but the one-liners are there.
[Ben]
Hi, Hans -
Thanks for the compliment! For the moment, I've written Woomert and
Frink into a blind alley; I think I need to put that particular story
aside, to be revised at another time (I mean, what kind of a Perl
one-liner can save an alien civilization, and from what threat??? Those
guys get themselves into the craziest predicaments...) and write another
one, not quite that deep.
POLOTM isn't dead, just resting... but it's notes like yours that can
get me motivated to start it rolling again. Look for Woomert and Frink
in the future issues of LG.
Ben O's review of the Averatec laptop is exactly what I'd like to see
more of. It is accurate about the important things and gets right to the
point. Far and away the best laptop review I've even seen.
Thanks very much; I look forward to reading others as they become
available.
-BA
Thanks, BA - reader's compliments make for great author motivation. The
laptop reviews will continue happening (although I may skip a month in
between), and will only terminate on EXIT_SUCCESS - or factors beyond my
control. I definitely need a backup laptop, and will be "processing"
more of them.
I came across your question about Stephen quite by accident, and thought
that, although there has been a response, I would take a moment to post a
personal reply.
We do not actually know the date of Stephen's death...........February 27th
was the day on which he was found by a farmer on Exmoor.
The ensuing inquest concluded that it was 'Death by
misadventure'...Hypothermia being the cause.
Thank you for caring enough to question his whereabouts............I know
many people will miss him and mourn his death.
He was my son.
Sincerely,
Ann Morgan
...............
'That man is a success who lived well, laughed often and loved much: who has
gained the respect of intelligent men and the love of children: who has
filled his niche and accomplished his task: who leaves the world a better
place than he found it, whether by an improved poppy, a perfect poem or a
rescued soul; who never lacked appreciation of earth's beauty or failed to
express it; who looked for the best in others and gave the best he had.'
-- Robert Louis Stevenson.
...............
Thank you, Ann. Stephen was a man who held to his principles, and we
can only respect him for that.
Readers, think kindly of your family this season, even those you don't deal
well with or haven't seen in years. Look for the finer qualities in people
around you. The winter may be cold - but don't forget to care.
This page edited and maintained by the Editors of Linux Gazette HTML script maintained by Heather Stern of Starshine Technical Services, http://www.starshine.org/
I'd whined several times about lack of success in setting an
X-gazette-tag header in Mozilla Thunderbird. The answer lies in the
user.js file, every one of the tens of Google hits I tracked down
promised me. But I'd do each one as specified, and no joy in Mudville.
Tonight, I googled again, and about 6 links into the process, I found
something close to the answer via a link from about.com. In my user.js
file (found in my setup at ~/.thunderbird/*default/user.js) I put in
these two lines:
[Ben]
This works in Mozilla as well, although the file there is "prefs.js".
Searching the Web for the above header confirms that you cannot set a
default value for the header...
[Jimmy]
No, there's a user.js as well. That's what I'm using, and it worked last
time.
[Ben]
Odd. My version doesn't have one - I searched with "find".
if user.js doesn't exist, put it in the same directory as prefs.js. Moz*
is in charge of prefs.js and may/will overwrite any changes in that
file. Additionally, order of loading has user.js last, and settings in
user.js will override prefs.js settings.
[Ben]
Ah - I didn't realize that.
Note that this does not automatically create the header in every
message, nor apparently is there a way to set a default value for the
header (which would make the first part a lie, eh?)
[Ben]
...however, you can add a plugin that will allow you to do that and a
lot more header-related stuff - which reputedly works for Mozilla and
Thunderbird:
That was ONE of the things I tried that didn't meet my needs. It turns
out that it does a lot of header VIEW customization, but not new
composition headers, from about two hours of playing with it, trying to
get to the point I wanted.
[Ben]
Hmph. Might be worth tossing into Bugzilla as a wishlist item; seems
like it's something a lot of folks could use, given how much discussion
there is about it on the Web.
But when I'm sending to TAG, I can select from "To:", "Cc:" "Bcc:" ...
"X-gazette-tag", and add a value for that header. This is the first
test, let's see if it works.
[Jimmy]
And indeed it does. Probably works in Mozilla Mail too.
I tried running cygwin from a CD, and it mostly works (X doesn't, but
Perl, Python and most of the usual tools do). The only thing that needs
to be done is that a few entries need to be added to the registry (these
can be safely deleted when done), because Cygwin doesn't have an /etc/fstab.
This file should be put on the CD. To add the entries, just double click!
If your CUPS printing suddenly stops working with the error:
lp: error - scheduler not responding!
try deleting the printer in the web interface and re-adding it. According
to Gentoo forum posts, glibc had an unexpected API change and that broke
Gentoo's configuration. You probably installed some unrelated software that
upgraded glibc. You may have to re-emerge glibc, cupsd, and turboprint,
but I didn't.
Curiously, I was still able to do the "Print Test Page" from the web
interface, although others said they couldn't.
In an ideal world, dependencies would prevent this kind of problem from
happening. But it's impossible for maintainers to know what
incompatibilities might appear in third-party libraries in the future.
I found the KDE option to change it back to showing the login screen after
logging out instead of shutting down (../106/orr.html). It's
KDE Control Center -> Session Manager -> Login as different user.
Mouse acceleration with Linux/X
Dave Blackburn (blackburn from acm.org)
Answered By Ben Okopnik,
Thomas Adam,
Jason Creighton
How can I adjust the "base" ratio between mouse motion and cursor motion
on the screen?
I have successfully used xset. It works well for me.
My problem is that the "base" ratio (e.g. with xset m 1 10) is too
high (fast).
[Ben]
I'm not sure what you mean by a "base ratio", Dave; I've found that I
prefer a different "xset m" setting for different computer/mouse
combinations, but have never come up with an absolute "rate" to reflect
that preference. Whenever I set up a new machine, I twiddle this for a
minute or two, set it in my ~/.xinitrc, and forget about it from there
on.
I mean the mouse speed (i.e. ratio between mouse movement and cursor movement)
without using: xset -- the initial mouse speed provided by X as if
e.g. xset 1 10 were active.
[Ben]
Ah - I was misled by the word "ratio", which implied to me that you were
using "xset" with some fraction as a parameter (the man page specifies
that you can do this.) Now I get what you mean.
I usually use "xset m" with a single parameter (which specifies
"acceleration" only); I find that this works well for pretty much every
machine out there. Once you start using more than that, you're off into
/Terra Incognita/: some meeces will respond in odd ways to "threshold"
tweaks.
I am using both RH 9.0 and Fedora 2 on a laptop.
With RH 9.0, the "base" ratio is fairly slow.
With Fedora 2 the ratio is quite a bit faster.
[Ben]
I presume the two of them use somewhat different mouse drivers.
Using RH 9.0, I apply xset m 3 20 getting a pleasant result.
Using Fedora 2, the "base" ratio is too fast for me.
Doing an Internet search, I found many explanations for xset.
[Ben]
The man page for it is quite informative as well.
Also, I found the suggestion that I could modify xorg.conf (in /etc/X11)
by adding a line like:
Option "Resolution" "500"
in section:
Section "InputDevice"
Identifier "Mouse0"
This seemed like the answer I was looking for.
However, when trying different values for Resolution,
I observed no change in the "base" rate.
[Ben]
I'm not familiar with "xorg.conf", but I've learned from
personal experience that, while the "Resolution" option
used to work in XF86Config, it does not in XF86Config-4.
This info is helpful. This is what I experienced.
Do one of you know how I can slow down the mouse when running
Fedora 2?
[Ben]
Find the value of "xset m" that you like and add it to your
"~/.xinitrc" or "~/.xsession".
I use xset in ~/.xinitrc.
It seems that xset can only be used to speed up the mouse.
[Thomas]
Err, no. xset is used to control the mouse threshold speed which means
you can either increase or decrease it. One other option you have is
that you might be using an incorrect mouse protocol for your configured
mouse.
I want to slow the mouse down.
In connection with slowing the mouse, I will use xset to control
the higher speed.
Since the "Resolution" option doesn't work in XF86Config-4, perhaps
it's also unavailable in the X.org release.
[Ben]
That would be my best guess. From
/usr/share/doc/xserver-xfree86/README.mouse (note the obvious escape clause):
5.3 Resolution
The following option will set the mouse device resolution to N counts per
inch, if possible:
Option "Resolution" "N"
Not all mice and OSs can support this option. This option can be set in the
XF86Setup program.
Perhaps there is a workaround?
[Jason]
Quoting from the xset manpage:
m The m option controls the mouse parameters. The
parameters for the mouse are `acceleration' and
`threshold'. The acceleration can be specified as an
integer, or as a simple fraction. The mouse, or
whatever pointer the machine is connected to, will go
`acceleration' times as fast when it travels more than
`threshold' pixels in a short time. This way, the
mouse can be used for precise alignment when it is moved
slowly, yet it can be set to travel across the screen in
a flick of the wrist when desired. One or both
parameters for the m option can be omitted, but if only
one is given, it will be interpreted as the acceleration.
If no parameters or the flag 'default' is used, the
system defaults will be set.
The key to an ugly workaround is the fact that you can specify the
acceleration as a simple fraction. Thus:
xset m 1/2 1
...will cut the speed in half. But you lose any acceleration features
because you're using it to hack the mouse speed to where you want it.
But if that doesn't bother you, this method is one option.
[Ben]
The fractional specification may work for you. Conversely, as Thomas
suggested, the two distros may be using two different mouse drivers
(compare the '"Option" "Device"' and '"Option" "Protocol"' lines in
their X config files); if that's the case, then simply choose the one
that works best for you, and do the final tweaking with "xset" from
there.
[Thomas]
By all means. I've Cc'ed The Answer Gang on this, since I assume this is
where you got my e-mail address. If you don't want this published, let me
know.
I have installed SuSe Linux 6.4 as a mail server. The current version
of quota is 2.11
- what is the latest version? where can I get it?
[Thomas]
Quota support is kernel and user-land. Therefore I can only assume that
you are referring to the userland tools. Currently the version is at 3.12,
available here:
- How to update to the old version? please give me steps.
[Thomas]
It's a case of removing the old quota tools and installing the new one.
There's a number of ways you can do this. The INSTALL file within the tar
file tells you all, and it should just be a simple matter of:
./configure && make && su -c 'make install'
Note that by default, this will install into /usr/local/* -- if
"/usr/local/bin" is not in your $PATH, before "/usr/bin" and you still
have the old quota-tools installed, this will conflict. If, however,
/usr/local/bin is listed before /usr/bin in your $PATH then this should
not be a problem. But you should make sure you remove the old quota-tools,
regardless.
Thanks for your quick response and it helped me a lots.
Tip: ls -l, AIX style
Jimmy O'Regan (The LG Answer Gang)
Answered By Bradley Chapman,
Kapil Hari Paranjape
On a certain other Linux site, I saw an unanswered question about
getting the time format of ls -l to have a leading zero on single digit
dates, like AIX does, instead of padding with a space as GNU ls does.
I was not aware of this (before I went to the man page), but GNU ls will
format the date any way you like, if you use the --time-style switch. To
use 'date' style formatting, you prefix it with a '+'.
So, AIX-style ls -l is:
ls -l --time-style=+"%b %d %H:%M"
[Brad]
Does ls support any environment variables to do this? Something like
export TIME="%b %d %H:%M"
ls -l
or do you have to use an alias in your ~/.*rc files?
[Kapil]
According to "info ls":
You can specify the default value of the `--time-style'
option with the environment variable `TIME_STYLE'...
So the correct environment variable is TIME_STYLE not TIME.
Kapil.
You still need to have the '+', so it's
export TIME_STYLE=+"%b %d %H:%M"
ls -l
POP3 for Yahoo Mail
Breman (012832161 from mobitel.com.kh)
Question by Breman (012832161 from mobitel.com.kh)
Answered By Thomas Adam,
Kapil Hari Paranjape
- What is pop3/smtp of yahoo mail?
[Thomas]
There isn't any freely available as yahoo removed that service. That said,
you can "emulate" it using a program called 'yahoopops' available here:
How you forward dhcp serive across differenent interfaces on ones firewall
17:11 < billp>
dhcp services*
17:11
* Wicket sniffs billp
17:13 < billp>
So I have a dhcp server running on a hostname "dhcpserver", and it sends lease offers to networks it is not actully connected to, via connected interfaces on the firewall
17:14 < billp>
So, a Linux firewall connecting a subnet with a dhcp server to two subnets without dhcp servers, would go something like:
where dhcpserver is the hostname of your dhcp server. Dead useful, especially for having untrusted network segments, like an open-access wireless one
17:16 <@Dee>
i see
This page edited and maintained by the Editors of Linux Gazette HTML script maintained by Heather Stern of Starshine Technical Services, http://www.starshine.org/
The Answer Gang
Linux Gazette 109: The Answer Gang (TWDT)The Answer Gang 109:
...making Linux just a little more fun!
The Answer Gang By Jim Dennis, Karl-Heinz Herrmann, Breen, Chris, and...
(meet the Gang) ...
the Editors of Linux Gazette...
and
You!
We have guidelines for asking and answering questions. Linux questions only, please.
We make no guarantees about answers, but you can be anonymous on request. See also: The Answer Gang's
Knowledge Base
and the LGSearch Engine
Greetings, dear readers, and welcome once again to the world of The Answer
Gang.
A world, I note, which has changed quite a bit in the last year. Last
December it was like the Gazette was under a sort of shadow - our
crew handled the transition to our new site with all the energy we could
muster, but there's only so much life we few who run this could pour in
- we had to wait for you readers to rediscover us.
We were already a great team, but the effort bound us together in a way
that's really rather hard to describe... except that I like the results. We
learn from each other on the lists as much as we ever did, but it's more
than that, we're closer as friends I think, and we back each other up on
things. We don't always agree (you put 20 plus
curmudgeonly and very smart people in a room together and watch the fun.
Oh? You say that's your typical holiday party? You see what I mean then)
and at times Real Life(tm) mugs one or another of us ... and we really are
different people, in what can take us down or carry us forward ... but we
try to do more for each other.
It brings me great joy this year to still be bringing Linux
Gazette to your browser's doorstep. I have to admit that for
awhile even my own enthusiasm was failing; in past years I had done it for
simply knowing Jim's "Answer Guy" pages would look far better when I put my
hand to them. For awhile I had done it just for knowing there were
uncounted readers out there looking for our good bits every month. In my
own harder times I did it simply because I had promised to, I couldn't let
you people down, and what little energy I could gain from knowing I was
getting at least one job done right was very much a boost to me.
The ordeals we put ourselves through here have given me strength in my
friendships and renewed hope for a lot of little things in my life, Linux
just one of them. The world has turned again, and I'm glad to know that
you people are out there, still reading this, telling your newbie friends
that their Linux troubles are solvable, we can Make Linux Just A
Little More Fun and even get a giggle now and then.
That's part of what has seriously changed the face of LG
though.
The kind of people who were Linux newbies 5 years ago are not the same
today. For one thing those people aren't newbies, but they still like to
read about Linux and expand their knowledge even further. But the people
who even thought about Linux then have a smorgasboard of choices
now - and the newbies of today are in many cases less technical - you can
buy "a linux box" at Wal-Mart, on special at CompUSA, you might be leaping
ship from the (amazing to me, why does Microsoft saw off its own foot like
this??) myriad flavors of MSwin to be found nowadays. Whether or not they
ever caught a virus, the crashes and the pain they encounter when they dare
to change their hardware are just as stumping to them as the box of Red Hat
or SuSE on the store shelves. If they're going to get their hands dirty
anyway... they may as well try this Linux thing on for size.
It won't fit all of them.
That's the thing about a world of choices. vi versus emacs
debate fans the world over know that when it comes down to someone being a
power user, the fact they've learned all the deep magics is worth a lot more
than what type of filesystem runs under the hood or what some geek thinks is
an 'operating system'. If the system operates, that's what they like. And
that's good. I made the choice to stick with Gazette the way that
gave it its power, a monthly source of good knowledge and fun. It's been
worth it, and I thank all of you, of whatever age, country, newness to linux
or being such an old hat that "Red Hat" upstart's a whippersnapper, for
joining us every month. Send us your tips and good stories. Have a great
time with Linux.
So it's the holidays now, and I'm giving you folks out there the same thing
I give my dear friends and family every year, however rich or poor I might
be. Presence.
Laptop review: Averatec 5400 series
From Rick
Answered By: Ben Okopnik
Hi Ben,
Good article. My experience with laptops have had varied results. By
that I mean, in many cases it has more to do with the distribution of
Linux as it does with the hardware. I have a Compaq Presario 915ca
(the specs on this can be easily looked up). The only upgrade is to
512MB of RAM. In any event, I am currently running WinXP and 3
different versions of Linux (I am a bit of an OS junkie!) using a 3rd
party boot loader.
Xandros recognizes everything out of the box, even the WinModem! I had
[Ben]
"Out of the box" functioning is nice to have, but I don't expect it -
and once you don't, all distros become essentially equal. There is no
difference between, e.g. the kernel used in Mandrake and the one in
Debian beyond possibly compilation choices and maybe a patch or two -
which I can also get and use at need. There's certainly no difference in
the standard toolkit, and there's no real difference in configuration -
the tools may vary, but the end result is the same. I'm expert enough
with Linux to do the configuration myself, so that issue does not
obtain.
Admittedly, this is more difficult than tossing in a different distro
and seeing if it works - but it carries the near certainty that if
something can be made to work, and I take the time to do so, it will
work, regardless of the distro. In my experience with multiple distros,
Debian is one of the more easily configurable ones (the only "tool"
required being a text editor), and so I use, and will be using, Debian
and possibly its derivatives (e.g., Knoppix or Morphix) for this
testing.
You may be assured that if I can make Debian work on a given laptop,
then any other distro can also be convinced to do so. It may take more
or less work, but it's doable. On the other hand, if I'm unable to make
it work, that fact will not be conclusive.
to mess with the /etc/fstab file a bit to get my 80GB USB hard drive
working properly, but overall Xandros works well on this machine.
I guess the point of my email here is, sometimes it's just as
important to find the right distro as it is the right hardware. I
recently tried to install Linux on a friends new laptop. I went
through about 10 different distros (including Fedora Core 2, Ubuntu,
and even tested with Knoppix, which is Debian based and with great
hardware detection!). Everything failed! ... ie: no mouse, it crashed
the X server, etc. Finally I tried SuSE 9.1 and it just worked. Yes, I
had to install the Nvidia drivers, and I never did manage to get his
built-in wireless to work!
[Ben]
That can be a valid strategy if you don't have a lot of expertise to
rely on, and it's nice that it's available. It is, however, not the only
strategy, nor is it one that I prefer.
(sometime it seems the hardware manufactures engineer "anti-Linux"
into their products!)
[Ben]
As much as a frustrating experience with hardware can make you lean in
the direction of that belief, I have to believe that it's false at its
core. PC OEMs are in the business of selling hardware; I can't think of
an argument convincing enough to make them purposely chase customers
away.
Yes, I agree with you. However, the vast majority of users do not know
how to recompile a kernel, nor do they want to know. I would say that
people like you account for a very small percentage of computer users.
Most of my friends think I am some kind of computer wizard, but in
reality, I am nowhere near that. Everything is relative, isn't it? I
am probably in the top 10% of computer users from a knowledge point of
view, and you are likely in the top 1%.
[Ben]
I'm afraid you missed the point I was trying to make, Rick - I guess I
didn't state it clearly enough. In essence, if I can get a laptop
working well under Linux, whatever the distro, then it's at least
possible and in fact not too difficult - you may have noticed in the
article that I'll simply stop trying once it's consumed a certain amount
of time (a few hours at most.) If I can't do it at all (as was the case
with the HP 5000 I'd picked up from Staples), then it gets blackballed.
In both cases, a report of the experience becomes useful and doable,
whereas testing every laptop against a dozen distros is out of the realm
of possibility.
As for the comment about hardware manufacturers purposely engineering
anti-Linux products, I was just being facetious.
[Ben]
I understood that, but wanted to use your comment to clarify my view of
the situation; I've heard a lot of people grumbling about it as if they
believed it.
As for aviation, well I am only 43, but I took up flying when I was
16. I am not currently flying, as things got all f&*^%$d up for me
after 9/11. It's a rather long story.
[Ben]
Yeah. I've heard of a number of pilots who were affected... and you've
probably heard about the recent TSA alien flight training/citizenship
verification rule. That's going to shaft a LOT of pilots - and the
damned thing is as confusing as can be imagined, and a couple of miles
beyond that. AOPA is fighting the good fight, but there's a lot of fear
in the air - and about it.
Linux boots from RAMdisk,
From keesan
Answered By: Kapil Hari Paranjape, Thomas Adam
I tried telling CMOS that there was no second drive because someone
suggested that method to get linux to recognize a larger drive, but my
drive is 3GB. I have DOS on a master drive and two linux partitions on
the slave drive, with one linux in each, and RAMdisk and loop versions in
DOS partitions.
I am able to mount the linux partitions when running from the RAMdisk or
loop versions and then switch to run linux on them:
mount /dev/hdb1 /mnt
chroot /mnt
This puts hdb1 on / and I can then use linux on the ext2 partition.
[Kapil]
But this is essentially what most initrd-based start up procedures do
anyway! So, though I haven't used BasixLinux, I would guess that the
problem is with the start scripts on your ext2 partition in /etc/rcS.d
or some such.
This is a minor nuisance and I suppose I could put it in an rc file, or
just use the smaller version unless I needed the larger one.
[Kapil]
The glass could also be half full! Given the variety of hardware that
Linux runs on it is surprising that so many computers boot with it at
all.
I am writing out of curiosity - why are so many computers difficult to
boot with linux? Is there a better fix for this one? Is there a better
fix for the other three besides installing Win98 DOS on them (and having
to use a boot floppy to defragment the DOS partitions after that) or
rebooting with a Win98 boot disk to go from DOS to linux?
I also have one Northgate 386 SX 20MHz 4.7MB RAM laptop which has no
cursor in linux. Cirrus video, 256K video RAM, mono VGA. The cursor is
plain white when used in color VGA. The computer in theory can output to
a color monitor in 800x600 resolution.
[Kapil]
Is this with or without X? The question is not clear enough. Are you in
graphics mode or text mode?
[Thomas] The distinction here is whether he is running with Framebuffers or not. Framebuffers are pretty much standard and most monitors that are not of the dark age can handle them, assuming their VGAness is OK. That said, certain monitors can react badly to the Framebuffer modeline it has been given and blank out. In fact, I remember discussing this some time ago in LG:
Does laptop video treat software cursors oddly? A cursor appears when I
use a text editor. There is a cursor while booting to DOS and in DOS.
Another 386 with identical speed, RAM and video won't boot linux at all -
the screen goes black and I need to reboot. What might cause this
problem?
[Kapil]
Try booting with the additional option "vesafb=off". Some of the older
hardware may not respond well to being switched to graphic mode.
[Thomas] Again, that won't be enough. He'll need to tell it to not only ignore vesafb, but to ensure that the vag16 definitions are turned off. So at the boot prompt:
linux nofb video=vga16:off vesafb=off vga=normal
Ought to do it.
Quota needs update, help!
a place for everything and everything in its place...
From Breman
Answered By: Thomas Adam, John Karns, Ben Okopnik
May I ask you some questions below:
I have installed SuSe Linux 6.4 as a mail server. The current version of quota is 2.11
[Thomas]
Quota support is kernel and user-land. Therefore I can only assume that
you are referring to the userland tools. Currently the version is at 3.12,
available here:
How to update to the old version? please give me steps.
[Thomas]
It's a case of removing the old quota tools and installing the new one.
There's a number of ways you can do this. The INSTALL file within the tar
file tells you all, and it should just be a simple matter of:
./configure && make && su -c 'make install'
Note that by default, this will install into /usr/local/* -- if
"/usr/local/bin" is not in your $PATH, before "/usr/bin" and you still
have the old quota-tools installed, this will conflict. If, however,
/usr/local/bin is listed before /usr/bin in your $PATH then this should
not be a problem. But you should make sure you remove the old quota-tools,
regardless.
[John]
That's the way I usually do it, too. But there is also something to be
said for staying within the boundaries of the package management system,
which in SuSE's case (as well as Red Hat and Mandrake, and probably a few
others) is the rpm system.
[Thomas]
I disagree. Package management should be sufficiently advanced that if one
doesn't want a particular package (or one wants to override it), then one
can. This is where stow shines, for all the reasons outlined above. You are
not tied down to having to install .rpm just because you can.
Of course, if you are compiling from source, for the odd program then this is
not so much an issue as all of the libaries are linked (or should be) for the
appropriate versions that exist on the system at the time.
[John]
Note: with such an old version of the distro as in the case of the
querent, dealing with this issue may prove to be more trouble than it's
worth, depending on how the package in question has evolved. I.E., in
cases where there have been significant changes to the number of files
involved, and / or their placement in the filesystem, then the process
would likely involve fundamental changes to the rpm spec file, which may
require a good understanding of the inner workings of the rpm system.
But for a more current version of an rpm based distro, the process would
be much less problematic, and provide the advantage of keeping the rpm
database "in sync" with the installed system, thus easing efforts
concerning system maintenance regarding package management issues. It
would entail something along the lines of replacing the 'make install'
with 'rpmbuild --rebuild', after replacing the older package source files
with those of the newer version.
What is pop3/smtp of yahoo mail?
There isn't any freely available as yahoo removed that service. That said,
you can "emulate" it using a program called 'yahoopops' available here:
I have install in as default configure (usr/local/) and now I want to
reinstall it in /usr/local/quota312 folder.
How to uninstall it?
[Thomas]
Uninstalling compiled software can be hit and miss, since it depends
whether the source makefile has an "uninstall" target in it, or not. So
the first thing I would do, is for the source where you compiled your
quota-tools, do:
cd /tmp/quota-tools/ && su -c 'make uninstall'
If that doesn't work, then the makefile probably doesn't support the
feature, although it is worth looking at the makefile to see. The other
option, if not, is to delete the files by hand. There are various tar and
rm incantations to do that. But I would do it by hand, myself.
One other thing I will mention to you for the future is that if you find
yourself compiling software from source a lot, I really do recommend you
use stow [1,2]. It is a really useful piece of software and would have
helped you here easily.
Thanks for your quick response.
I try your uninstall connand but not success.
I got the warning: "make: *** No rule to make target `uninstall'. Stop"
I will delete it manually.
Can we just install again without uninstall first?
[Ben]
There's a way to do it that I invented in a fit of desperation; it's a
little manual but quite effective.
1. Run "make -n install > uninstall".
2. Edit the newly-created "uninstall" file and replace the "install" or "cp" commands with "rm", editing each line as necessary.
3. Run "sh uninstall".
Submitters, send your News Bytes items in
PLAIN TEXT
format. Other formats may be rejected without reading. You have been
warned! A one- or two-paragraph summary plus URL gets you a better
announcement than an entire press release. Submit items to
bytes@linuxgazette.net
Legislation and More Legislation
European Software Patents
The difficulties and delays dogging the European Union's attempts
to agree on a software patent policy continued in November, this
time to the benefit of anti-software-patent interests.
In September last year, the European Parliament (a directly
elected body, with representatives from all European Union member states)
voted to accept a directive on the "patentability of computer-implemented
inventions". This directive was only accepted subject to a set of
amendments that constituted real and substantial limits on the
patentability of software. This fact was recognised both by
gratified anti-software-patent campaigners
and by
chagrined patent attorneys.
However, following this initial success, May 2004 brought a
setback as the European Council of Ministers (a body composed of
ministerial representatives from EU member states' governments)
instead moved to advance a directive that excluded the
substantial amendments adopted by the Parliament. As well as
disheartening Free Software advocates, this brought howls of
protest from the Parliamentarians who (in a not untypical tug of
war between the branches of EU administration) accused the
Council of flying in the face of democracy. After May, the obstacles
still standing between the Directive and its final ratification
were for the Council to formally adopt the directive, and then
for the directive to pass before Parliament once more. This
brings us to the present day.
This heroic about-turn by Poland, as well as similar
pronouncements emanating from Austria should remind us that
this fight is certainly still there to be won.
The efforts of prominent Free and Open Software figures like
Linus Torvalds, Michael Widenius, and Rasmus Lerdorf
who have raised their voices to say
"No to Software Patents"
are welcome, especially as they provide an example to all of us
that if we feel this issue is important then we should make our
opinions known. Those supporting a future software industry
stretched on
a rack of software patents are certainly not shy about
pushing their case.
Truly, can we afford the luxury of silence? Wladyslaw Majewski,
president of the Internet Society of Poland puts the point
starkly:
"The questionable compromise that the EU Council reached in
May was the biggest threat ever to our economic growth, and
to our freedom of communication. The desire of the patent
system and the patent departments of certain large
corporations must never prevail over the interests of the
economy and society at large."
During November,
Slashdot highlighted and discussed
a very illuminating
article at Groklaw, written by a retired attorney. Written
in the context of Novell's current anti-trust case against
Microsoft, the article points out that quite apart from the
interest stirred up by the substance of this particular case,
this is also a great opportunity to store up ammunition for
future legal adventures. The key to this opportunity is that the
discovery process will, at least temporarily, move a lot of
Microsoft material into the public eye. The trick is then to
store it up for future use.
Nmap
The Register and Security Focus
have reported
that Fyodor, the author of the
port-scanning network security tool Nmap, has found that his
software writing has brought him into surprisingly close terms
with the FBI. Apparently the Feds have begun to look for access
to his webserver logs in order to identify individuals or
organisations downloading tools like Nmap.
OSRM and Patent Risks
Open Source Risk Management has steered a somewhat
controversial course since arriving on the scene offering to
indemnify open source developers against hostile intellectual
property lawsuits. In a bid to encourage use of its services,
OSRM has at times appeared to over-emphasise the risks open
source software developers and users expose themselves to.
As reported by internetnews.com, this perception has now
resulted in Pamela Jones (of
Groklaw) resigning her position as litigation risk research
director for the group. In particular, Pamela wished not to
contribute to activities that she felt were creating fear and
doubt about Linux, and providing ammunition to anti-open-source
interests.
The potential damage caused by such publicity is demonstrated by
the usefulness of one particular OSRM report to Microsoft's Steve
Ballmer, who used it to claim in front of Asian business leaders
that
Linux violates over 200 patents.
However, one should not be too critical of OSRM, since quite
apart from any
FUD there does
appear to be a real and growing likelihood that open source
software will in the future be attacked using patents.
Microsoft's statements on the issue of intellectual property
point in this direction, as does its policy of substantially
increasing its own patent portfolio. This portfolio can be used
either defensively, or offensively, but either way its existence
and rapid growth indicate a messy time ahead,
and it seems the lawyers can already smell blood in the water.
Highlighting this threat can lead to some bad publicity, such as
Ballmer's showboating in Asia, but it is also useful for
generating awareness. And as pointed out by Ballmer,
and highlighted by The Register,
our current framework of international agreements governed by the
WTO means that these issues will have ramifications right across
the globe. How any possible patent-war will pan out is still an
open question, but having good awareness in the Free Software
ranks of the risks and dangers that are faced is sure to be a
good thing.
John C. Dvorak
writing at CBS MarketWatch
thinks Microsoft is planning to make an entry into the Linux
marketplace, and that this explains the effort invested in the
Lindows trademark lawsuit.
Intel has announced an initiative that shows a growing interest
in Linux. The new
quick-start kit
for systems-integrators, launched in the Asian market, is
intended to facilitate the building of Linux-based desktop PCs.
This is a small move towards providing the kind of support
Windows PC integrators have grown used to.
Netherlands
The European Union
Interchange of Data
Between Administrations Project
has publicised the decision of the local government in the Dutch
city of Haarlem to
switch 2000 desktops to OpenOffice.org. As well as currently using
GNU/Linux in server applications, the local government is
exploring the feasibility of migrating desktops to the operating
system, with initial estimates indicating that 20% of desktops
could be migrated without causing particular difficulties or
inconvenience to employees.
The Dutch prime minister and the office of the Irish prime
minister have also
encouraged European governments to consider open source software
in the spirit of inter-agency collaboration. Now, while such
pronouncements are good to hear, this writer is relatively
familiar with Irish government policy on these matters, and it is
anything but coherent. If this is the case elsewhere, then
if your elected representative pays lip-service to open source
and Free Software, then perhaps it would be helpful to follow up
on this with a letter, to let them know you paid attention, that
you'll remember what they said, and that if they do a U-turn on
it then they'd better knock on somebody else's door come election
time!
This new publication has also
received comment on Andy Oram's weblog at O'Reilly.com.
Distro News
Damn Small Linux
OSDir.com has published
a screenshot tour
of
Damn Small Linux 0.8.4.
Damn Small Linux is a business card size (50MB) bootable Live CD
Linux distribution that strives to have a functional and easy to
use desktop.
Linux Journal
has published an interesting article
on
gnuLinEx. GnuLinEx is the Debian-based operating system used
by the government of the Autonomous Community of Extremadura,
Spain. The system is used widely in the region's schools, and
also in local government administration.
Knoppix
O'Reilly has published a book dealing with some of the cool
things you can accomplish with the Debian-based GNU/Linux live-CD Knoppix:
Knoppix Hacks.
This book was
reviewed on Slashdot
(the review was very favourable, to the extent that it was linked
on the press-release advertising this title).
Slackware
Patrick Volkerding, coordinator of
Slackware Linux, has been having
serious health problems in
recent weeks. Patrick has found it difficult to obtain
satisfactory treatment from the doctors that he has visited to
date, so he has solicited help from members of the community.
Topologilinux
Linuxforums.org has published
a review of Topologilinux.
Based on the much respected Slackware, Topologilinux is a free
Linux distribution designed to be run on top of or inside an
existing windows system. Thus, Topologilinux does not require any
repartitioning of the system's hard drive, and instead it uses a
single file as a linux root system.
Software and Product News
Lulu
Lulu, a company that offers people the chance to self-publish
at relatively modest costs (at least for small production runs)
is now aiming to
extend this service to open source software developers.
The intention is to sell boxed sets comprising the software
itself, as well as printed and bound manuals and documentation.
An example of this class of product is the
Fedora GNU/Linux distribution.
It provides migration process planning, automated migration
scripts, anti-virus/anti-spam solutions, and specific
migration and deployment details for all relevant
technologies. The CD includes valuable automated scripts for
migrating any flavor of Windows to Linux.
Another title likely to be of interest to readers is
SELinux.
This book provides a background to SELinux, as well as guidelines
on its installation and subsequent use.
Born some time ago in Ireland, Michael is currently working
as a lecturer in the Department of Mechanical Engineering, University
College Dublin. The topic of his PhD research was the use of Lamb
waves in nondestructive testing. GNU/Linux has been very useful in
this work, and Michael has a strong interest in applying free
software solutions to other problems in engineering.
Virtual Filesystem: Building A Linux Filesystem From An Ordinary File
Under Linux, you can create a regular file, format it as an ext2, ext3,
or reiser filesystem, and then mount it just like a physical drive. It's
then possible to read and write files to this newly-mounted device. You
can also copy the complete filesystem, since it is just a file, to another
computer. If security is an issue, read on. This article will show you how
to encrypt the filesystem and mount it with ACL (Access Control Lists),
which gives you control of rights beyond the traditional read (r), write
(w), and execute (x) permissions for the three user groups "file", "owner",
and "other".
This is an excellent way to investigate different filesystems
without having to reformat a physical drive, which means you avoid
the hassle of moving all your data. This method is quick -- very
quick compared to preparing a physical device. What is truly great
about this technique is that you can explore different filesystems
such as reiserfs, ext3, or ext2 without having to purchase an
additional physical drive. Since the same file can be mounted on
more than one mount point, you can also investigate sync rates.
Creating a filesystem in this manner allows you to set a hard
limit on the amount of space used, which, of course, will be equal
to the file size. This can be an advantage if you need to move this
information to other servers. Since the contents cannot grow beyond
the file, you can easily keep track of how much space is being
used.
First, you want to create a 20MB file by executing the following
command:
$ dd if=/dev/zero of=disk-image count=40960
40960+0 records in
40960+0 records out
A count of 40960 created a 20 MB file because, by default, dd uses a
block size of 512 bytes. That makes the size: 40960*512=20971520.
Next, to format this as an ext3 filesystem, you just execute the
following command:
$ /sbin/mkfs -t ext3 -q disk-image
mke2fs 1.32 (09-Nov-2002)
disk-image is not a block special device.
Proceed anyway? (y,n) y
You are asked whether to proceed because this is a file, and not
a block device. That is OK. We will mount this as a loopback device
so that this file will simulate a block device.
Next, you need to create a directory that will serve as a mount
point for the loopback device.
$ mkdir fs
You are now one step away from the last step. You just want to
find out what the next available loopback device number is.
Normally, loopback devices start at zero (/dev/loop0) and work
their way up (/dev/loop1, /dev/loop2, ... /dev/loopn). An easy way
for you to find out what loopback devices are being used is to look
into /proc/mounts, since the mount command may not give you what
you need.
On my computer, I have no loopback devices mounted, so I'm OK to
start with zero. You must do the next command as root, or with an
account that has superuser privileges.
# mount -o loop=/dev/loop0 disk-image fs
That's it. You just mounted the file as a device. Now take a
look at /proc/mounts, you will see this is using /dev/loop0.
You can now create new files, write to them, read them, and do
everything you normally would do on a disk drive. First, I'll give
access to the chirico account.
# chown -R chirico.chirico /home/chirico/junk/fs
Now, under the chirico account, it is possible to create
files.
$ cd /home/chirico/fs
$ mkdir one two three
$ ls -l
total 15
drwx------ 2 chirico chirico 12288 Sep 3 14:28 lost+found
drwxrwxr-x 2 chirico chirico 1024 Sep 3 14:34 one
drwxrwxr-x 2 chirico chirico 1024 Sep 3 14:34 three
drwxrwxr-x 2 chirico chirico 1024 Sep 3 14:34 two
$ df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda2 17G 11G 4.6G 71% /
/dev/sda1 99M 83M 11M 89% /boot
none 62M 0 62M 0% /dev/shm
/home/chirico/junk/disk-image
20M 1.1M 18M 6% /home/chirico/junk/fs
If you need to unmount the filesystem, as root, just issue the
umount command. If you need to free the loopback device, execute
the losetup command with the -d option. You can execute both
commands as follows:
Before we get started with ACL, how would you set up rights on
the filesystem so that users could create and save documents that
others could modify? For instance, let's say that users chirico and
sporkey are collaborating on a project together.
Well, you have to add everyone to the same group. You would
execute commands like these:.
Note that if these changes do not take effect for your users
(for example, if they were logged in when you executed the
commands), they'll have to log out and log in again or execute the
"$ newgrp sharefs" command. No big deal, right? Well,
keep reading, and see how ACL avoids this step.
More importantly, even though the old way worked for you,
at some point, new users may need to be added to the project. What
if some of these users only need a subset of the rights? For
instance, you have developers, testers, managers, and a few special
people. There are limits to what the 'rwx'-type rights can do. ACL
solves a lot of these problems.
ACL, Reiserfs, and AES Encryption: The 2.6 Kernel
For the next steps, I will assume that you are running Red Hat
Fedora Core 2. If not, reference the 2.6 kernel upgrade section
below. Four things will be covered in this section:
Create A File With Random Data
Set Up An AES Encrypted Loopback Device With Password
Build A Reiser Filesystem On The Loopback Device
Mount With ACL Capabilities
Your installation of Fedora Core 2, by default, will be
configured for loop, cryptoloop, and aes, but it is highly unlikely
that you will have all of these modules loaded. So, execute the
following commands to load these modules (you will need to do this
as root):
Next, create a directory to store the files. The Reiser
filesystem will require more space than the ext3 filesystem.
# mkdir /home/diskimg
# cd /home/diskimg
Instead of creating the file zeroed out, like you did with the
ext3 filesystem, this one is going to contain random bits, which
may add a little extra security.
# dd if=/dev/urandom of=disk-aes count=102400
We need to encrypt the loop device, so you need to use losetup.
You will be prompted for a password, which you will need to
remember when you mount the device.
# losetup -e aes /dev/loop1 ./disk-aes
Password:
This step is new also. Instead of formatting the file directly,
you will format the loop device. The file stays encrypted. Again,
you will be prompted to continue, so just enter "y".
# mkfs -t reiserfs /dev/loop1
mkfs.reiserfs 3.6.13 (2003 www.namesys.com)
A pair of credits:
Elena Gryaznova performed testing and benchmarking.
The Defense Advanced Research Projects Agency (DARPA, www.darpa.mil) is the
primary sponsor of Reiser4. DARPA does not endorse this project; it merely
sponsors it.
Guessing about desired format.. Kernel 2.6.8-1.521 is running.
Format 3.6 with standard journal
Count of blocks on the device: 12800
Number of blocks consumed by mkreiserfs formatting process: 8212
Blocksize: 4096
Hash function used to sort names: "r5"
Journal Size 8193 blocks (first block 18)
Journal Max transaction length 1024
inode generation number: 0
UUID: 435e3495-5e2e-489d-bf55-1b5f9a44b670
ATTENTION: YOU SHOULD REBOOT AFTER FDISK!
ALL DATA WILL BE LOST ON '/dev/loop1'!
Continue (y/n):y
Initializing journal - 0%....20%....40%....60%....80%....100%
Syncing..ok
Tell your friends to use a kernel based on 2.4.18 or later, and especially not a
kernel based on 2.4.9, when you use reiserFS. Have fun.
ReiserFS is successfully created on /dev/loop1.
Create the mount point /fs, and mount this device. Note that you
will be entering the acl option as well. Plus, you will prompted
for a password.
# mkdir /fs
# mount -o loop,encryption=aes,acl ./disk-aes /fs
Password:
Ok, now take a look at the mount command. It should show up as
the Reiser filesystem, encrypted, using ACL. Note that it says
loop2; it mounted it on /dev/loop2, which is one above what losetup
specified, /dev/loop1.
$ mount
/home/diskimg/disk-aes on /fs type reiserfs (rw,loop=/dev/loop2,encryption=aes,acl)
Exploring ACL
With ACL (Access Control Lists), you have finer control over
access permissions. With the rwx permission scheme, you cannot
easily change rights without creating new groups to handle the
users. With ACL, you can set user permissions without creating a
group, and individual users can add or remove access.
These rights are set with the setfacl command. The
command below will give the users donkey, chirico, and bozo2 access
to this new filesystem that we mounted. Again, I'm assuming that
you are using Fedora Core 2, or some distribution that is set up
for ACL.
This is just scratching the surface of what can be done with
ACL. For more information, see some of the references below.
2.6 Kernel Upgrade
This article will get you started with the 2.6 kernel if you
are currently running Red Hat 8 or 9. You may want to take a look
at it to see what is involved. If you decide to upgrade, you will
need to configure your kernel for the following:
This is done in the .config file, and you can download my config
file
here. Just look for
kernel-2.6.8.1-i686-chirico-reiserfsacl.config in the tar.gz.
In addition to upgrading the kernel, you will need the latest
version of the Linux
utilities. Currently, there is no need to patch this version.
In the past, there was a patch, but this version worked fine for
me.
Check out tips 12, 22, and 91, on how to use ssh with rsync.
You can create a virtual filesystem on a server, then copy it to
your laptop. As you work on the laptop, sync your changes using
rsync.
by Mark Mitchell, Jeffrey Oldham, and Alex Samuel, of
CodeSourcery LLC, published by New Riders Publishing, ISBN
0-7357-1043-0, First Edition, June 2001. This book is free and you
can view it online.
Chapter 6 describes loopback devices.
Tips on using Comcast Email with a home Linux box.
Mike Chirico, a father of triplets (all girls) lives outside of
Philadelphia, PA, USA. He has worked with Linux since 1996, has a Masters
in Computer Science and Mathematics from Villanova University, and has
worked in computer-related jobs from Wall Street to the University of
Pennsylvania. His hero is Paul Erdos, a brilliant number theorist who was
known for his open collaboration with others.
Mike's notes page is souptonuts.
This article explores common methods in SQLite such as running
commands from the shell prompt and creating triggers for
populating time stamped fields with either UTC or local time
values. In addition, delete, update, and insert trigger examples
will be explored in detail.
All examples can be found in
sqlite_examples.tar.gz (local copy), and I would
encourage you to download and run these examples as you read through this
document.
The home page for sqlite3 is www.sqlite.org and the source for
sqlite3 can be downloaded at www.sqlite.org/download.htm.
This tutorial was done with the source version 3.0.8
Getting Started - Common Commands
To create a database file, run the command "sqlite3" followed by
the database name. For example, to create the database "test.db"
run the sqlite3 command as follows:
$ sqlite3 test.db
SQLite version 3.0.8
Enter ".help" for instructions
sqlite> .quit
$
The database file test.db will be created, if it does not
already exist. Running this command will leave you in the sqlite3
environment. There are 3 ways to safely exit this environment (.q,
.quit, or .exit).
You do not have to enter the sqlite3 interactive environment.
Instead, you could perform all commands at the shell prompt, which
is ideal when running bash scripts and commands in an ssh string.
Below is an example of how you would create a simple table from the
command prompt.
After table t1 has been created, data can be inserted as
follows:
$ sqlite3 test.db "insert into t1 (data,num) values ('This is sample data',3);"
$ sqlite3 test.db "insert into t1 (data,num) values ('More sample data',6);"
$ sqlite3 test.db "insert into t1 (data,num) values ('And a little more',9);"
As expected, doing a select returns the data in the table. Note,
the primary key "t1key" auto increments; however, there are no
default values for timeEnter. To populate the timeEnter field with
the time, an update trigger is needed. An important note on the
PRIMARY KEY: do not use the abbreviated "INT" when working with the
PRIMARY KEY. You must use "INTEGER", for the primary key to
update.
$ sqlite3 test.db "select * from t1 limit 2";
1|This is sample data|3|
2|More sample data|6|
In the statement above, the limit clause is used and only 2 rows
are displayed. For a quick reference of SQL syntax statements
available with SQLite, see the link syntax. There is an offset
option to the limit clause. For instance, the third row is equal to
the following: "limit 1 offset 2".
$ sqlite3 test.db "select * from t1 order by t1key limit 1 offset 2";
3|And a little more|9|
The ".table" command shows the table names. For a more
comprehensive list of tables, triggers, and indexes created in the
database, query the master table "sqlite_master" as shown
below.
All SQL information and data inserted into a database can be
extracted with the ".dump" command.
$ sqlite3 test.db ".dump"
BEGIN TRANSACTION;
CREATE TABLE t1 (t1key INTEGER
PRIMARY KEY,data TEXT,num double,timeEnter DATE);
INSERT INTO "t1" VALUES(1, 'This is sample data', 3, NULL);
INSERT INTO "t1" VALUES(2, 'More sample data', 6, NULL);
INSERT INTO "t1" VALUES(3, 'And a little more', 9, NULL);
COMMIT;
The contents of the ".dump" can be filtered and piped to another
database. Below table t1 is changed to t2 with the sed command, and
it is piped into the test2.db database.
An insert trigger is created below in the file "trigger1". The
Coordinated Universal Time (UTC) will be entered into the field
"timeEnter", and this trigger will fire after a row has been
inserted into the table t1. Again, this trigger will fire after the
row has been inserted.
-- ********************************************************************
-- Creating a trigger for timeEnter
-- Run as follows:
-- $ sqlite3 test.db < trigger1
-- ********************************************************************
CREATE TRIGGER insert_t1_timeEnter AFTER INSERT ON t1
BEGIN
UPDATE t1 SET timeEnter = DATETIME('NOW') WHERE rowid = new.rowid;
END;
-- ********************************************************************
The AFTER specification in ..."insert_t1_timeEnter AFTER..." is
necessary. Without the AFTER keyword, the rowid would not have been
generated. This is a common source of errors with triggers, since
AFTER is NOT the default, so it must be specified. In summary, if
your trigger depends on newly created data, in any of the fields
from the created row, which was the case for us in this example
since we need the rowid, then, the AFTER specification is needed.
Otherwise, the trigger is a BEFORE trigger, and will fire before
rowid or other pertinent data is entered into the field.
Comments are preceded by "--". If this script was created in the
file "trigger1", you could easily execute this script as
follows.
$ sqlite3 test.db < trigger1
Now try entering a new record as before, and you should see the
time in the field timeEnter.
sqlite3 test.db "insert into t1 (data,num) values ('First entry with timeEnter',19);"
Doing a select reveals the following data:
$ sqlite3 test.db "select * from t1";
1|This is sample data|3|
2|More sample data|6|
3|And a little more|9|
4|First entry with timeEnter|19|2004-10-02 15:12:19
If you look at the statement above, the last value has the
timeEnter filled in automatically with Coordinated Universal Time -
or (UTC). If you want local time, then, use select
datetime('now','localtime'). See the note at the end of this
section regarding UTC and local time.
For examples that follow the table "exam" and the database
"examScript" will be used. The table and trigger are defined below.
Just like the trigger above, UTC time will be used.
-- *******************************************************************
-- examScript: Script for creating exam table
-- Usage:
-- $ sqlite3 examdatabase < examScript
--
-- Note: The trigger insert_exam_timeEnter
-- updates timeEnter in exam
-- *******************************************************************
-- *******************************************************************
CREATE TABLE exam (ekey INTEGER PRIMARY KEY,
fn VARCHAR(15),
ln VARCHAR(30),
exam INTEGER,
score DOUBLE,
timeEnter DATE);
CREATE TRIGGER insert_exam_timeEnter AFTER INSERT ON exam
BEGIN
UPDATE exam SET timeEnter = DATETIME('NOW')
WHERE rowid = new.rowid;
END;
-- *******************************************************************
-- *******************************************************************
After the script file, it can be executed, by redirecting the
contents of the script file into the sqlite3 command, followed by
the database name. See the example below:
And, as a check, the PRIMARY KEY and current UTC time have been
updated correctly, as seen from the above example.
Logging All Inserts, Updates, and Deletes
The script below creates the table examlog and three triggers
update_examlog, insert_examlog, and delete_examlog to record
update, inserts, and deletes made to the exam table. In other
words, anytime a change is made to the exam table, the changes will
be recorded in the examlog table, including the old value and the
new value. By the way if you are familiar with MySQL, the
functionality of this log table is similar to MySQL's binlog. See
(TIP 2, TIP 24 and TIP 25) if you would like more information
on MySQL's log file.
-- *******************************************************************
-- examLog: Script for creating log table and related triggers
-- Usage:
-- $ sqlite3 examdatabase < examLOG
--
--
-- *******************************************************************
-- *******************************************************************
CREATE TABLE examlog (lkey INTEGER PRIMARY KEY,
ekey INTEGER,
ekeyOLD INTEGER,
fnNEW VARCHAR(15),
fnOLD VARCHAR(15),
lnNEW VARCHAR(30),
lnOLD VARCHAR(30),
examNEW INTEGER,
examOLD INTEGER,
scoreNEW DOUBLE,
scoreOLD DOUBLE,
sqlAction VARCHAR(15),
examtimeEnter DATE,
examtimeUpdate DATE,
timeEnter DATE);
-- Create an update trigger
CREATE TRIGGER update_examlog AFTER UPDATE ON exam
BEGIN
INSERT INTO examlog (ekey,ekeyOLD,fnOLD,fnNEW,lnOLD,
lnNEW,examOLD,examNEW,scoreOLD,
scoreNEW,sqlAction,examtimeEnter,
examtimeUpdate,timeEnter)
values (new.ekey,old.ekey,old.fn,new.fn,old.ln,
new.ln,old.exam, new.exam,old.score,
new.score, 'UPDATE',old.timeEnter,
DATETIME('NOW'),DATETIME('NOW') );
END;
--
-- Also create an insert trigger
-- NOTE AFTER keyword ------v
CREATE TRIGGER insert_examlog AFTER INSERT ON exam
BEGIN
INSERT INTO examlog (ekey,fnNEW,lnNEW,examNEW,scoreNEW,
sqlAction,examtimeEnter,timeEnter)
values (new.ekey,new.fn,new.ln,new.exam,new.score,
'INSERT',new.timeEnter,DATETIME('NOW') );
END;
-- Also create a DELETE trigger
CREATE TRIGGER delete_examlog DELETE ON exam
BEGIN
INSERT INTO examlog (ekey,fnOLD,lnNEW,examOLD,scoreOLD,
sqlAction,timeEnter)
values (old.ekey,old.fn,old.ln,old.exam,old.score,
'DELETE',DATETIME('NOW') );
END;
-- *******************************************************************
-- *******************************************************************
Since the script above has been created in the file examLOG, you
can execute the commands in sqlite3 as shown below. Also shown
below is a record insert, and an update to test these newly created
triggers.
$ sqlite3 examdatabase < examLOG
$ sqlite3 examdatabase "insert into exam
(ln,fn,exam,score)
values
('Anderson','Bob',2,80)"
$ sqlite3 examdatabase "update exam set score=82
where
ln='Anderson' and fn='Bob' and exam=2"
Now, by doing the select statement below, you will see that
examlog contains an entry for the insert statement, plus two
updates. Although we only did one update on the command line, the
trigger "insert_exam_timeEnter" performed an update for the field
timeEnter -- this was the trigger defined in "examScript". On the
second update we can see that the score has been changed. The
trigger is working. Any change made to the table, whether by user
interaction or another trigger is recorded in the examlog.
GTK in Mandrake, Theme change?-- Heather
Adding custom headers in Thunderbird
Laptop review: Averatec 5400 series
Greetings from Heather Stern
[Ben]
"Out of the box" functioning is nice to have, but I don't expect it -
and once you don't, all distros become essentially equal. There is no
difference between, e.g. the kernel used in Mandrake and the one in
European Software Patents![[Picture]](../gx/2002/tagbio/conry.jpg)
Born some time ago in Ireland, Michael is currently working
as a lecturer in the Department of Mechanical Engineering, University
College Dublin. The topic of his PhD research was the use of Lamb
waves in nondestructive testing. GNU/Linux has been very useful in
this work, and Michael has a strong interest in applying free
software solutions to other problems in engineering.
![[BIO]](../gx/authors/chirico.jpg)
Mike Chirico, a father of triplets (all girls) lives outside of
Philadelphia, PA, USA. He has worked with Linux since 1996, has a Masters
in Computer Science and Mathematics from Villanova University, and has
worked in computer-related jobs from Wall Street to the University of
Pennsylvania. His hero is Paul Erdos, a brilliant number theorist who was
known for his open collaboration with others.