Linux-Misc Digest #558, Volume #25 Fri, 25 Aug 00 12:13:04 EDT
Contents:
Re: If XWin hang, how to kill it ("Peter T. Breuer")
Re: glibc - rpm -> src (Garry Knight)
Re: rxvt question (Garry Knight)
Re: dat driver problem (Jean-David Beyer-valinux)
Re: hard disk problems? (-ljl-)
Re: No printing on HP Laserjet 1100 (Terry Steyaert)
Re: Replacing a binary while it is in use does not cause problems? (Jean-David
Beyer-valinux)
linux/apache/php/mysql setup (paul simdars)
Re: dat driver problem (-ljl-)
Hardware question ([EMAIL PROTECTED])
Re: If XWin hang, how to kill it (Jean-David Beyer-valinux)
Re: No printing on HP Laserjet 1100 (Leonard Evens)
Re: Firewall for Linux ("Steve Brennan")
Re: glibc - rpm -> src (-ljl-)
Four notes from LinuxWorld ([EMAIL PROTECTED])
aic7xxx scsi drivers ([EMAIL PROTECTED])
Re: dat driver problem (-ljl-)
----------------------------------------------------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: If XWin hang, how to kill it
Date: 25 Aug 2000 14:50:30 GMT
Jean-David Beyer-valinux <[EMAIL PROTECTED]> wrote:
: "Peter T. Breuer" wrote (in part):
:> Usually the ctl-alt stuff is working, btw. You just can't see it!
: So the next time this happens, I inferI could do the following, right?
: C-A-F1
: and remember that it will say ... login:
Well, you may not see it, so yes, you will have to remember it.
: (Perhaps I should wait a while. I notice if I do it now, my monitor clicks off
That's also useful!
: and type the root password
: remember that I will get a prompt
: and type init 3
err, well, it depends. I didn't realise you ran in initlevel 4. Yes,
it does no harm and is probably simpler to put yourself in level 3.
: and wait a while and see if I get a CLI prompt on my console?
Your console state will be messed up. I'd just have issued a
"killall -9 X" blind, then "init 3; startx;" and take it from there.
You might need another ctrl-alt-f1 if xdm restarted, but then you
really want that to happen ...
: And if I do, just type init 5 and I should be all set, I guess.
I wouldn't go back to level X until I had checked out the machine a
bit. Just starting X on its own is usually enough:
"X; setenv DISPLAY :0 ; xterm &;"
: In the old days, if the console was still screwed (usually the result of the
: previous process putting the console or terminal into raw mode), I could type
: something like
: stty sane
: and it would setup the console as a dumb terminal and it would work. I cannot tell
: if this still works; the documentation says it should.
Interesting (isn't it "stty sane < /dev/tty")? I don't think the
problem is at that level. It's the graphics card that's in a funny
state, not the systems representation of it. You really need
"restoretextmode".
: and is where I run X. If I go to runlevel 3, it should shut down X, though when
This depends on your distro and its init and its init scripts. Not all
successfully kill xdm when moving back to level 3 from a higher level
(and yes, it should be reported as a bug, are you listening, debian
maintainer ...).
: I looked at /etc/rc.d/rc3.d, I did not see how that was done. I expected a line
: that said K01prefdm or something a bit like that, and did not find one. Maybe init
It's probably missing.
Peter
------------------------------
From: Garry Knight <[EMAIL PROTECTED]>
Subject: Re: glibc - rpm -> src
Date: Fri, 25 Aug 2000 15:18:22 +0100
Ryan Tarpine <[EMAIL PROTECTED]> wrote:
>Caldera somehow left libcrypt out of glibc for OL 2.2. How can I
>upgrade using the source? I tried making the source, running rpm -e
>glibc, and then make install, but without glibc, make itself wouldn't
>even run! Should I just install the source-compiled glibc over the
>existing rpm version? If yes, should I install it in /usr or
>/usr/local?
libcrypt is a library. It belongs in /usr/lib.
--
Garry Knight
[EMAIL PROTECTED]
------------------------------
From: Garry Knight <[EMAIL PROTECTED]>
Subject: Re: rxvt question
Date: Fri, 25 Aug 2000 15:19:26 +0100
[EMAIL PROTECTED] (Andrew Purugganan) wrote:
>Eric ([EMAIL PROTECTED]) wrote:
>[ Luke Olson wrote:
>[ >
>[ > trying to make the background transparent in rxvt. any ideas?
>
>[ Yeah, read the info include with the package :-)
>
>I think in rxvt you have to fake it by setting the background to match
>the desktop wallpaper. There's always eterm or aterm which i've used.
>
>I HATE telling a guy named Luke to fake it. Obi Wan must be spinning in
>his grave ;-)
Better than telling him to use the source... :o)
--
Garry Knight
[EMAIL PROTECTED]
------------------------------
From: Jean-David Beyer-valinux <[EMAIL PROTECTED]>
Subject: Re: dat driver problem
Date: Fri, 25 Aug 2000 11:03:30 -0400
Andreas Moroder wrote:
> Hello,
>
> I tried to write multiple tar files onto my dds dat tapedrive connected via
> a adaptec
> Adapter Configuration:
> SCSI Adapter: Adaptec AIC-7880 Ultra SCSI host adapter
> Ultra Wide Controller
> PCI MMAPed I/O Base: 0xdf100000
> Adaptec SCSI BIOS: Disabled
> on a linux 2.0.36
I have 2.2.14-5.0.14csmp, a HP C1599 DDS-2 on a Symbios 810 Narrow SCSI
controller.
> I did
>
> mt -t /dev/st0
> tar cvf /dev/nst0 file1
> tar cvf /dev/nst0 file2
> tar cvf /dev/nst0 file3
> All works well
>
> if I try to get back the data I have the following problem
> mt -t /dev/st0
> tar xvf /dev/nst0
> all works well
> tar cvf /dev/nst0
> there comes back nothing
> tar cvf /dev/nst0
> there comes back nothing
> tar cvf /dev/nst0
> now the second file comes back.
To reduce confusion, I did:
mt -f /dev/nst0 rewind
cat file1 > /dev/nst0
mt -f /dev/nst0 weof 1
cat file2 > /dev/nst0
mt -f /dev/nst0 weof 1
cat file3 > /dev/nst0
mt -f /dev/nst0 weof 1
cd frammis (a newly created empty directory)
mt -f /dev/nst0 rewind
cat /dev/nst0 > file1
cat /dev/nst0 > file2 (skipping file mark)
cat /dev/nst0 > file2
cat /dev/nst0 > file3 (skipping file mark)
cat /dev/nst0 > file3
and got all the files. I infer the driver is OK.
Perhaps tar writes two filemarks at the end of each backup. In the old days,
you wrote files to tape, and an end-of-file after each file. You wrote 1 or
two MORE end-of-file marks after the last file to indicate end of logical
tape. Maybe tar wants to write end of logical tape after each file.
> I get this two empty sessions between every session.
I could understand one, to get over the file mark.I do recall that Unix was
very rudimentary and unconventional in handling magnetic tapes.
> I asked Paul Eggert, the tar mantainer, if this is a tar problem. After a
> few tests he told me that there must be a problem in the device-driver.
>
> Please let me know via email at amoroder@se-nord[nospam].provinz.bz.it
> take away [nospam] )
>
> Thank you all
>
> Andreas
--
Jean-David Beyer .~.
Shrewsbury, New Jersey /V\
Registered Linux User 85642. /( )\
Registered Machine 73926. ^^-^^
------------------------------
From: -ljl- <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.misc
Subject: Re: hard disk problems?
Date: Fri, 25 Aug 2000 14:57:11 GMT
In article <[EMAIL PROTECTED]>,
Ben Ritchie <[EMAIL PROTECTED]> wrote:
> -ljl- wrote:
>
> > In article <[EMAIL PROTECTED]>,
> > Ben Ritchie <[EMAIL PROTECTED]> wrote:
...
> Yes - it is only about six months old.
It should still be under warranty, ask for a replacement. Don't
take a chance with it; flakey hardware is a nightmare. You'll
always be waiting for the other shoe to drop.
--
Louis-ljl-{ Louis J. LaBash, Jr. }
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Terry Steyaert <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Re: No printing on HP Laserjet 1100
Date: Fri, 25 Aug 2000 15:04:01 GMT
It sounds as if you have a problem with "stair stepping", depending on
the version of printtool you are using. My main suggestion would be to
run printtool (as root) and have it test the printer. If you test the
printer and it comes up with the first line and part of the second, you
have a problem with stair-stepping.
Define the printer as one of the known types (most of the HP laser jets
are close enough together that it doesn't really matter which one) and
turn on the Fix Stair-stepping. One other note is that depending on
version it may be "Convert LF to CR-LF."
Basically, Unix and Linux both default to a line feed at the end of a
normal line of text. Your laser printer is expecting a carriage return
(and possibly a line feed.) Since the printer is only seeing the line
feed, it is moving the paper up one line and continuing to print. When
the stuff to print is off the side of the paper, you don't see anything
more.
Terry Steyaert
[EMAIL PROTECTED]
In article <8o604v$eqe$[EMAIL PROTECTED]>,
"..." <"chris"@|\\|0$PAMsantafe.edu> wrote:
> I have a HP laserjet 1100 that prints perfectly (with and w/o filters)
> from a S.u.S.E. Linux notebook. When connecting it to my RedHat 6.1
box
> I just get a single line of cut-off characters on the top. And this
> happens already when printing a plain text-file directly to /dev/lp0.
> The printer is connected to the parallel port. I run Linux 2.2.12. I
> also have the fancy "alias parport_lowlevel parport_pc" line in
> /etc/conf.modules.
> So what am I doing wrong?
>
> Christian (remove the fancy NOSPAM from my FROM:-line)
>
>
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Jean-David Beyer-valinux <[EMAIL PROTECTED]>
Subject: Re: Replacing a binary while it is in use does not cause problems?
Date: Fri, 25 Aug 2000 11:12:28 -0400
Ken Pratt wrote:
> Using a tool such as 'rpm' to perform an upgrade to a package that is
> currently running, such as 'The GIMP' never results in a core dump. Why
> is this? I figured the kernel would only load pages that it needs (a
> working set) from the file that represents either a binary executable or
> library according to what parts of the code in an executable were used.
>
> What if a running program's binary files are changed while it is running
> and then a new page is needed because the program began to execute code
> that had not yet been loaded into memory? The upgraded program is in no
> way going to be compatible with what is in memory.
>
> How is it that upgrades to running systems work so flawlessly?
>
> --
> Ken Pratt
In the oldest days, the binaries were copied to swap and run from there
(this is a slight oversimplification), so it did not matter what happened
to the "real" ones once execution started.
When memory management got a little more sophisticated, what happened was
that if you tried to update a binary, you were informed that it was busy
and it did not update. Try again later.
Now, it seems as though it is done a little better. Perhaps the update
removes the old file and creates a new one with the same name. This will
not hurt the running program, since the i-node remains until the last user
closes the file (or the last directory that references it no longer does,
but that is not the case here) before it is really removed. The new file
with the same name as the old one gets a new i-node, so the part of the
OS running the old process is not concerned.
--
Jean-David Beyer .~.
Shrewsbury, New Jersey /V\
Registered Linux User 85642. /( )\
Registered Machine 73926. ^^-^^
------------------------------
Date: Fri, 25 Aug 2000 10:15:55 -0500
From: paul simdars <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.questions
Subject: linux/apache/php/mysql setup
I have ReedHat 6.1 installed . Apache and mysql are running. Then I
thought I'd add php to put it all together. In added the apache-devel
package from the CD and then I went to php.net (I think that was the
place). They have released php4 and were encouraging everyone to
upgrade to it. I downloaded the source and installed it. It went
well. BUT, now the apache server no longer runs. Any clues as to what
went wrong?
A second point, one interesting thing I noted along the way was that in
two different documents I looked at each had their command to start
apache. Neither were recognized by my computer.
Thanks for any help.
Paul
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: -ljl- <[EMAIL PROTECTED]>
Subject: Re: dat driver problem
Date: Fri, 25 Aug 2000 15:07:17 GMT
In article <8o5r87$h65$[EMAIL PROTECTED]>,
"Andreas Moroder" <andreas[nospam][EMAIL PROTECTED]>
wrote:
> Hekko Jean-David,
>
> good question,
>
> it's because we have HP-UX too and the command on hp-ux is -t
> ( 't' for tape ).
> It's not documented in linux, but it works.
Excerpt from TurboLinux 6 "man mt":
The path of the tape device to operate on can be given with
-f or -t option.
--
Louis-ljl-{ Louis J. LaBash, Jr. }
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Hardware question
Date: Fri, 25 Aug 2000 15:12:37 GMT
Does anyone know if the 5.25/3.5 dual diskette drives work in Linux?
These are the ones with a single floppy connection for both devices.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Jean-David Beyer-valinux <[EMAIL PROTECTED]>
Subject: Re: If XWin hang, how to kill it
Date: Fri, 25 Aug 2000 11:19:45 -0400
"Peter T. Breuer" wrote:
> Jean-David Beyer-valinux <[EMAIL PROTECTED]> wrote:
> : "Peter T. Breuer" wrote (in part):
> :> Usually the ctl-alt stuff is working, btw. You just can't see it!
>
> : So the next time this happens, I inferI could do the following, right?
>
> : C-A-F1
> : and remember that it will say ... login:
>
> Well, you may not see it, so yes, you will have to remember it.
>
> : (Perhaps I should wait a while. I notice if I do it now, my monitor clicks off
>
> That's also useful!
>
> : and type the root password
> : remember that I will get a prompt
> : and type init 3
>
> err, well, it depends. I didn't realise you ran in initlevel 4. Yes,
I run in initlevel 5, not 4, though I imagine this is just a typo...
> it does no harm and is probably simpler to put yourself in level 3.
>
> : and wait a while and see if I get a CLI prompt on my console?
>
> Your console state will be messed up. I'd just have issued a
> "killall -9 X" blind, then "init 3; startx;" and take it from there.
> You might need another ctrl-alt-f1 if xdm restarted, but then you
> really want that to happen ...
>
> : And if I do, just type init 5 and I should be all set, I guess.
>
> I wouldn't go back to level X until I had checked out the machine a
> bit. Just starting X on its own is usually enough:
> "X; setenv DISPLAY :0 ; xterm &;"
I would have checked things out at level 3 before continuing. I just did not wish to
type it in...
> : In the old days, if the console was still screwed (usually the result of the
> : previous process putting the console or terminal into raw mode), I could type
> : something like
> : stty sane
> : and it would setup the console as a dumb terminal and it would work. I cannot tell
> : if this still works; the documentation says it should.
>
> Interesting (isn't it "stty sane < /dev/tty")? I don't think the
> problem is at that level. It's the graphics card that's in a funny
> state, not the systems representation of it. You really need
> "restoretextmode".
I guess so. too bad it is not there. ;-)
> : and is where I run X. If I go to runlevel 3, it should shut down X, though when
>
> This depends on your distro and its init and its init scripts. Not all
> successfully kill xdm when moving back to level 3 from a higher level
> (and yes, it should be reported as a bug, are you listening, debian
> maintainer ...).
Since I run Red Hat, let me add: are you listening, red hat maintainer?!
> : I looked at /etc/rc.d/rc3.d, I did not see how that was done. I expected a line
> : that said K01prefdm or something a bit like that, and did not find one. Maybe init
>
> It's probably missing.
>
> Peter
--
Jean-David Beyer .~.
Shrewsbury, New Jersey /V\
Registered Linux User 85642. /( )\
Registered Machine 73926. ^^-^^
------------------------------
From: Leonard Evens <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Re: No printing on HP Laserjet 1100
Date: Fri, 25 Aug 2000 10:16:17 -0500
"..." wrote:
>
> I have a HP laserjet 1100 that prints perfectly (with and w/o filters)
> from a S.u.S.E. Linux notebook. When connecting it to my RedHat 6.1 box
> I just get a single line of cut-off characters on the top. And this
> happens already when printing a plain text-file directly to /dev/lp0.
> The printer is connected to the parallel port. I run Linux 2.2.12. I
> also have the fancy "alias parport_lowlevel parport_pc" line in
> /etc/conf.modules.
> So what am I doing wrong?
>
> Christian (remove the fancy NOSPAM from my FROM:-line)
Try bringing up printtool, edit the entry for the printer, and
in the select option, click on the boxes for stairstepping and
EOF. I just did this on the same printer and it works.
By the way, I selected HP Laser 4/5/6 under the printer selection.
--
Leonard Evens [EMAIL PROTECTED] 847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
------------------------------
From: "Steve Brennan" <[EMAIL PROTECTED]>
Crossposted-To:
alt.os.linux,comp.os.linux.networking,comp.os.linux.setup,uk.comp.os.linux
Subject: Re: Firewall for Linux
Date: Fri, 25 Aug 2000 10:38:46 -0500
I use CyberNet's NetMax FireWall product. It costs approx $50 but it is
worth it in tech support; their's is great!
-Steve
"MH" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Jason Ng wrote:
> >
> > Hi,
> >
> > I would like to install a firewall on my Linux box. Does anyone have any
> > suggestions?
> > I am a newbie for this and I want to know which one is good...
> >
> > Regards,
> > Jason
>
> I've been using Seattle Firewall, aka "Seawall". Seems to do the job.
> It's GPL and it's easy to set up. You can get it at:
>
> http://seawall.sourceforge.net/
>
> --
> "The worst form of inequality is to try to make unequal things equal."
>
> --Aristotle
------------------------------
From: -ljl- <[EMAIL PROTECTED]>
Subject: Re: glibc - rpm -> src
Date: Fri, 25 Aug 2000 15:26:17 GMT
In article <[EMAIL PROTECTED]>,
Ryan Tarpine <[EMAIL PROTECTED]> wrote:
> Caldera somehow left libcrypt out of glibc for OL 2.2.
This, I'm reasonably sure, is due to the US's export law debacle.
You won't find mutt-i, SSL-open, ... on US servers.
> How can I upgrade using the source? I tried making the source,
I just built glibc-2.12 from a source RPM. A 'libcrypt.a' was created
so it looks like encrytion is included. I believe that our government
plans on rescending the export restriction for encrytions software, or
maybe it already did.
You'll need a lot of workspace to build the library.
--
Louis-ljl-{ Louis J. LaBash, Jr. }
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Four notes from LinuxWorld
Date: Fri, 25 Aug 2000 15:35:16 GMT
I've just found four notes in my nodebook from LinxWorld.
It took place just before the fall 2000, in 14-20th of August.
================
Got Parking ticket?
LinuxWorld expo - Convention Center in San Jose. I cannot say that
location is good,
the downtown of San Jose always parking problems: full lots, parking
tickets ($28-$51),
no free parking at all. My opinion, that the Convention Center in Santa
Clara is more
convenient place, same light trams, huge free parking of "Great
America".
I also expected, the huge expo. The Silicon Valley TechWeek magazine
was evidenced
about Linux has 24% market share of OS shipped for computer servers.
In fact, it is unpretentious. Some companies that usually have a name
as variations
of word "Linux". Software Giants like IBM, HP, SUN, SGI... they have a
special role
to demonstrate the point of presence in this market, strong muscles,
large booths and
magicians... and Intel, Compaq, Corel, Dell... and all other...
Future in the Past!
My second day on the expo. Both days I see practically the same
contingent
of visitors: New Age (60%) - cool doom players, wide trousers,
notebooks with
wireless connection - really new generation that prefers only Linux;
High Tech (30%) - represent the current elite of hardware and software
engineering
from well known Silicon Valley companies, they try to understand what
really going on;
Other (10%) - people just want to see the Linux miracle.
It seems that in next five years computer industry will have Linux
domination in all areas
from huge servers to embedded systems as fifteen years ago the Unix did
by the same way...
GNU is not Unix!
GNU is not Unix, but does it Linux? Source Code Freedom what does it
mean for an end user?
In experience, for any system name in the Operating System World is
suffixing on "ux" or "ix"
that is definitely means relationship to Unix. What about Lin-ux? Who
is Lin-ux relative?
Minix? or Unix?
Good-bye SVR4?
The SVR4 has a sad history, system was in so many hands - AT&T,
INTEL, USL, NOVELL, SCO,
Caldera, so many ideas, so many different accents in development...
SCO tried to complete
it by renaming to SVR5, but It seems it really will be terminated by
Caldera...
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: aic7xxx scsi drivers
Date: Fri, 25 Aug 2000 15:38:48 GMT
Redhat 4.2 (kernel 2.0.30)
I'm having problems using a scsi tape drive, I keep getting
intermittent errors :-
Aug 16 02:11:47 ras kernel: scsi : aborting command due to timeout :
pid 75026,
scsi0, channel 0, id 0, lun 0 Write (6) 00 00 4b 02
00
Aug 16 02:11:47 ras kernel: aic7xxx: (abort) Aborting scb 1, TCL
0/0/0
Aug 16 02:11:52 ras kernel: scsi : aborting command due to timeout :
pid 75027,
scsi0, channel 0, id 0, lun 0 Write (6) 00 80 5d 24
00
Aug 16 02:11:52 ras kernel: aic7xxx: (abort) Aborting scb 2, TCL
0/0/0
Contents of /proc/scsi/aic7xxx/0 are :-
Adaptec AIC7xxx driver version: 4.0/3.2/4.0
Compile Options:
AIC7XXX_RESET_DELAY : 15
AIC7XXX_TWIN_SUPPORT : Enabled
AIC7XXX_TAGGED_QUEUEING: Disabled
AIC7XXX_PAGE_ENABLE : Disabled
AIC7XXX_PROC_STATS : Disabled
Adapter Configuration:
SCSI Adapter: AIC-7850 Ultra
(AIC-785x chipset)
Host Bus: Single
Base IO: 0xfc00
IRQ: 9
SCBs: Used 3, HW 3, Page 3
Interrupts: 443501
Serial EEPROM: True
Extended Translation: Enabled
SCSI Bus Reset: Enabled
Ultra SCSI: Disabled
Target Disconnect: Disabled
I'm aware that Redhat 4.2 is now pretty old, and that there weren't many
supported scsi adapters.
Would it be advisable to upgrade to Redhat 6.1/6.2 ?.
I would have to clone the disks...I can't back them up because of the
scsi problem.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: -ljl- <[EMAIL PROTECTED]>
Subject: Re: dat driver problem
Date: Fri, 25 Aug 2000 15:36:13 GMT
In article <[EMAIL PROTECTED]>,
Jean-David Beyer-valinux <[EMAIL PROTECTED]> wrote:
> Andreas Moroder wrote:
...
> I could understand one, to get over the file mark.I do recall
> that Unix was very rudimentary and unconventional in handling
> magnetic tapes.
The described behavior is mutable, see my previous post. This
is one of Linux's feature; you have the source and the author
provides an easy way to change the device drivers behavior: BSD |
SYS V.
Thanks Kai.
--
Louis-ljl-{ Louis J. LaBash, Jr. }
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
