VM & VSE & linux/390 Employment Web Page

[Dennis G. Wicks]

Greetings; (Posted to VMESA-L and VSE-L and LINUX-390)

- - Now in its fifth year! - - Now includes VSE and linux/390!

I have set up a public service web page at

http://www.eskimo.com/~wix/vm/

for posting positions available and wanted for VM, VSE and linux/390.

Please visit the web page for more information and feel free to
send me any info you would like to have posted.  Please make VM
or VSE or linux/390 the first word in the subject.
Questions and comments welcome!
(Text or html OK.  No java, gifs, etc. NO RESUMES or CVS!)

Please check the web pages for examples before sending your ad!

Good luck,
Dennis

VM & VSE & linux/390 Positions Available last updated May 8.
VM & VSE & linux/390 Positions Wanted last updated Apr 1.

patch problem

[Tim-Chr. Hanschen]

Hi,

I tried to patch kernel 2.4.17 with linux-2.4.17-s390.diff and got the
following:

patching file Documentation/Configure.help
patch:  Can't create file /tmp/pooFPYtf : File exists

I used: patch -p1 < linux-2.4.17-s390.diff.

The file does not exist! The directory tmp is completetly empty.

- Tim -

Re: Linux backup software

[Keldin]

Our company uses FDR/Upstream with StorageTek silos and Schuled with OPC-E
using batch jcl that gets submitted to start the backup process.  (for aix).
It works well.


- Original Message -
From: "John Summerfield" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, May 22, 2002 6:30 PM
Subject: Re: Linux backup software


> > On Wed, May 22, 2002 at 01:01:55PM -0700, Jon R. Doyle wrote:
> >
> > > Well you said Freeware, and the best fit it tar, in fact I think it is
> > > better than some commercial stuff, as long as you do not need so
> > > called "agents" like connectors to Oracle or something for Hot
> > > back-up/restore.  You can make some really neat scripts for tar using
> > > cron jobs, at jobs, and even rsync/ssh to do some extensive backup
> > > routines. Tar also can be called to compress and use tape devices.
> >
> > Um, suit yourself, but if you're going in that direction, I'd recommend
> > Amanda, which uses either tar or dump under the covers, but at least
> > gives you rudimentary tape management.
>
>
> I rather like afio.
>
>
> --
> Cheers
> John Summerfield
>
> Microsoft's most solid OS: http://www.geocities.com/rcwoolley/
>
> Note: mail delivered to me is deemed to be intended for me, for my
disposition.
>
> ==
> If you don't like being told you're wrong,
> be right!
>

Re: Changing network device

[Vic Cross]

On 23.05.2002 at 11:20:16, Sue Sivets <[EMAIL PROTECTED]> wrote:

> Can Suse and Marist use a 3172 network adaptor ?

The LCS driver should talk to anything that can talk LCS.  This will include
OSA, 3172 and 2216.

> Is there a way to change Suse and test the changes without losing the
working
> configuration if I mess
> things up again.

This has been covered in previous messages, but basically you can use your
system console to manually load the network driver and configure your
interface.

> Last but not least, can someone tell me all the "pieces" I need to change
> using Yast? I went to:
> 1. system administration>Integrate Hardware into system>configure networking
> device
> 2. system administration>Network Configuration>Network base configuration
> Did I miss something?

Ummm... I don't think YaST knows about the module parameters, so if you've
changed IP addresses etc using YaST, you would still have needed to update your
chandev.conf or modules.conf.

> To make things more difficult, we'll be switching everything over to an OSA
> gigibit adaptor in a month or
> so.

For this you will need QDIO.  The Linux Device Drivers manual from
DeveloperWorks will tell you about low level stuff to configure the modules, but
you should be able to use your distributions standard tools to configure IP.

Cheers,
Vic

--
Vic Cross  MACS  mailto:[EMAIL PROTECTED]
Networking, Linux, on zSeries and S/390

Changing network device

[Sue Sivets]

We will be changing cpu's in the next couple of weeks, and I'm using a network adaptor 
that is built into
the current cpu (MP3000)  for several Linux distributions. I have another adaptor in a 
3172 that I'd like
to change to. I need to change 3 distributions - Marist, Suse 7.0 and Suse 7.2. I've 
tried changing Suse
7.0 using Yast, and managed to make a mess of it -- twice. After rebooting, I no 
longer had a network
connection and I had to restore the volume from a backup. Both adaptors are LCS and 
ethernet. The current
configuration from ifconfig is:
eth0  Link encap:Ethernet  HWaddr 00:04:AC:8B:4B:52
  inet addr:192.168.150.244  Bcast:192.168.150.255  Mask:255.255.255.0
  UP BROADCAST RUNNING MULTICAST  MTU:1492  Metric:1
  RX packets:2876 errors:0 dropped:0 overruns:0 frame:0
  TX packets:194 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:100

The dns=192.168.75.252
The current adaptor is 0xfe0,1 and the new adaptor is 0xfe2,0 with an mtu = 1500

Can Suse and Marist use a 3172 network adaptor ?
Is there a way to change Suse and test the changes without losing the working 
configuration if I mess
things up again. If yes, would someone  tell me what commands I need to issue?
Last but not least, can someone tell me all the "pieces" I need to change using Yast? 
I went to:
1. system administration>Integrate Hardware into system>configure networking device
2. system administration>Network Configuration>Network base configuration
Did I miss something?
I understand that the network setup has changed with Suse 7.2, Can anyone help me with 
the information I
need to change and test this version

I am assuming that I need to change individual files in Marist. Can someone tell me 
which files I need to
change, or a way to find them?

I didn't think this was going to be nearly the problem it has turned out to be.
To make things more difficult, we'll be switching everything over to an OSA gigibit 
adaptor in a month or
so.

Thanks.

Sue

--
 Suzanne Sivets
 Systems Programmer
 Innovation Data Processing
 973-890-7300

Re: OpenSSH Problem

[Rick Troth]

Mike ...

Best to compare the problem account and host
to a working user@host pair.   Do you have SSH working anywhere?
So the first thing I would recommend is to back away from
this particular one and see if you can get that SSH client
working to some other Linux (or UNIX) host with any account there.

You mentioned "operator".
Check that his .ssh directory is owned by that ID.
One place SSH is picky  (and for which I am thankful
though it is a hassle)  is about ownership of the complete path
to the key files.   Now you were using password authentication,
so this may not apply.   (Why did you run 'ssh-keygen'
if you're using password authentication?)

Also check how the target account was created.
I create accounts by hand  (edit /etc/passwd,  create a
home directory for the user,  and so on)  and that requires
running  'pwconv'  at times for the shadow file to be in synch.

=?iso-8859-1?Q?Re: Sanity Check! Steps to apply Updates -- One more thought..=

[Vic Cross]

.chroot?=
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1

> The thought comes to mind that IF I were to follow my
> OS/390 maintenance procedures, I might consider
> cloning this system,  then from my production system issue a mount of cloned
> root to /SERVICE and chroot to /SERVICE to apply these RPM's.

This makes sense in the USS environment, where your current running system
drives the installation of the cloned system volumes[1].  In a Linux/Unix setup,
however, it's usually too easy to copy the disks to another machine and do the
install over there.

One way we can do this in a Penguin Colony: we can copy the DASD to new
minidisks, bring up a new 'upgrader' drone penguin on those minidisks and do the
upgrade.  Then, update the directory for the upgradee penguins to use the
upgraded minidisks.  Now, you can upgrade a swag of penguins simply with
shutdown-logoff-logon-IPL (once for each penguin).

Others have mentioned that it's just as effective to simply switch to
single-user mode and upgrade.  This is great, as long as you only have a small
number of penguins...

Cheers,
Vic Cross


[1] For those who don't do OS/390, maintenance of the Unix System Service (USS)
component of z/OS and OS/390 goes a bit like this: you make copies (clones) of
your system filesystems, mount these copies under a special mountpoint
(/service, according to current IBM convention), and tell the software
maintenance system (SMP/E) to update files under that mountpoint instead of root
(eg. /service/usr/bin instead of /usr/bin).  Then you IPL the system off your
new system volume, the configuration of which will mount the new filesystems.


--
Vic Cross  MACS  mailto:[EMAIL PROTECTED]
Networking, Linux, on zSeries and S/390

Re: OpenSSH Problem

[Post, Mark K]

Mike,

It's not enough for me, but I might just be slow.  One of the things that
came to mind is that you said you copied known_hosts from /root.  As root,
cd to /home/operator and do a
"chown -R operator.groupname *" command to make sure you don't have some
strange ownership left over.  If that doesnt' fix it (and I don't really
expect it to) you can stop the sshd process, and (as root) re-start it in
debug mode:
sshd -d -e
This will start sshd, which will _not_ fork and make itself into a daemon.
It will continue to run in the foreground, and process exactly one
connection.  You can increase the amount of debug information that gets put
out by specifying two or three (the max) "-d" parameters.

Go to another ssh window, and try to connect as "operator" as you have been:
ssh -v -l operator ip.address.of.system

Between the two of those, we'll at least get some hints as to what's not
working.

Mark Post

-Original Message-
From: Mike Spaniol [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, May 22, 2002 7:54 PM
To: [EMAIL PROTECTED]
Subject: OpenSSH Problem


Hello,

I've been trying to get SSH to work on my linux virtual machine to
logon to a remote host.  With the help of this list, I've now
got it working if I run from userid Root on my linux virtual machine.

I'm trying to get it to run now from a standard userid, operator.
I've added a user, operator, and ran the ssh-keygen command and
stored the identity and identity.pub files out in the
/home/operator/.ssh/ directory.  I also copied the
known_hosts file from /root/.ssh to /home/operator/.ssh directory.

The way it's working when I run it under root, is I issue the SSH
command with the -l option to pass it a userid, and I get prompted
from the remote host for a password, which the remote system
adminstrator supplied for me.  I get logged on and can run a
limited set of commands on the remote host.

When I try it from my operator userid, I receive the following:

[operator@linuxtst .ssh]$ ssh -l  xxx.xxx.xxx.xxx
Permission denied, please try again.
Permission denied, please try again.
Permission denied (publickey,password,keyboard-interactive).
[operator@linuxtst .ssh]$

I never get prompted for the password.

Here's the last part of my standard error output when I ran
the command from root with the -v option:

  OpenSSH_3.1p1, SSH protocols 1.5/2.0, OpenSSL 0x0090603f
  debug1: Reading configuration data /usr/local/etc/ssh_config
  debug1: Rhosts Authentication disabled, originating port will not be
trusted.
  debug1: restore_uid
  debug1: ssh_connect: getuid 0 geteuid 0 anon 1
  .
  .
  .
  debug1: got SSH2_MSG_SERVICE_ACCEPT
  debug1: authentications that can continue:
  publickey,password,keyboard-interactive
  debug1: next auth method to try is publickey
  debug1: try privkey: /root/.ssh/id_rsa
  debug1: try privkey: /root/.ssh/id_dsa
  debug1: next auth method to try is keyboard-interactive
  debug1: authentications that can continue:
  publickey,password,keyboard-interactive
  debug1: next auth method to try is password

 at this point I get prompted for a password and enter it
>>>

  debug1: packet_send2: adding 64 (len 56 padlen 8 extra_pad 64)
  debug1: ssh-userauth2 successful: method password

Here's the last part of my standard error output when I ran
the command from operator with the -v option:

  OpenSSH_3.1p1, SSH protocols 1.5/2.0, OpenSSL 0x0090603f
  debug1: Reading configuration data /usr/local/etc/ssh_config
  debug1: Rhosts Authentication disabled, originating port will not be
trusted.
  debug1: restore_uid
  debug1: ssh_connect: getuid 503 geteuid 0 anon 1
  .
  .
  .
  debug1: got SSH2_MSG_SERVICE_ACCEPT
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  debug1: next auth method to try is publickey
  debug1: try privkey: /home/operator/.ssh/id_rsa
  debug1: try privkey: /home/operator/.ssh/id_dsa
  debug1: next auth method to try is keyboard-interactive
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  debug1: next auth method to try is password

 at this point I don't get the password prompt  >>>

  debug1: packet_send2: adding 64 (len 50 padlen 14 extra_pad 64)
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  Permission denied, please try again.
  debug1: packet_send2: adding 64 (len 50 padlen 14 extra_pad 64)
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  Permission denied, please try again.
  debug1: packet_send2: adding 64 (len 50 padlen 14 extra_pad 64)
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  debug1: no more auth methods to try
  Permission denied (publickey,password,keyboard-interactive).
  debug1: Calling cleanup 0x4271e0(0x0)

Is this enough information someone can tell me what I'm doing wrong?
It's like standard in is taking data fro

OpenSSH Problem

[Mike Spaniol]

Hello,

I've been trying to get SSH to work on my linux virtual machine to
logon to a remote host.  With the help of this list, I've now
got it working if I run from userid Root on my linux virtual machine.

I'm trying to get it to run now from a standard userid, operator.
I've added a user, operator, and ran the ssh-keygen command and
stored the identity and identity.pub files out in the
/home/operator/.ssh/ directory.  I also copied the
known_hosts file from /root/.ssh to /home/operator/.ssh directory.

The way it's working when I run it under root, is I issue the SSH
command with the -l option to pass it a userid, and I get prompted
from the remote host for a password, which the remote system
adminstrator supplied for me.  I get logged on and can run a
limited set of commands on the remote host.

When I try it from my operator userid, I receive the following:

[operator@linuxtst .ssh]$ ssh -l  xxx.xxx.xxx.xxx
Permission denied, please try again.
Permission denied, please try again.
Permission denied (publickey,password,keyboard-interactive).
[operator@linuxtst .ssh]$

I never get prompted for the password.

Here's the last part of my standard error output when I ran
the command from root with the -v option:

  OpenSSH_3.1p1, SSH protocols 1.5/2.0, OpenSSL 0x0090603f
  debug1: Reading configuration data /usr/local/etc/ssh_config
  debug1: Rhosts Authentication disabled, originating port will not be trusted.
  debug1: restore_uid
  debug1: ssh_connect: getuid 0 geteuid 0 anon 1
  .
  .
  .
  debug1: got SSH2_MSG_SERVICE_ACCEPT
  debug1: authentications that can continue:
  publickey,password,keyboard-interactive
  debug1: next auth method to try is publickey
  debug1: try privkey: /root/.ssh/id_rsa
  debug1: try privkey: /root/.ssh/id_dsa
  debug1: next auth method to try is keyboard-interactive
  debug1: authentications that can continue:
  publickey,password,keyboard-interactive
  debug1: next auth method to try is password

 at this point I get prompted for a password and enter it >>>

  debug1: packet_send2: adding 64 (len 56 padlen 8 extra_pad 64)
  debug1: ssh-userauth2 successful: method password

Here's the last part of my standard error output when I ran
the command from operator with the -v option:

  OpenSSH_3.1p1, SSH protocols 1.5/2.0, OpenSSL 0x0090603f
  debug1: Reading configuration data /usr/local/etc/ssh_config
  debug1: Rhosts Authentication disabled, originating port will not be trusted.
  debug1: restore_uid
  debug1: ssh_connect: getuid 503 geteuid 0 anon 1
  .
  .
  .
  debug1: got SSH2_MSG_SERVICE_ACCEPT
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  debug1: next auth method to try is publickey
  debug1: try privkey: /home/operator/.ssh/id_rsa
  debug1: try privkey: /home/operator/.ssh/id_dsa
  debug1: next auth method to try is keyboard-interactive
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  debug1: next auth method to try is password

 at this point I don't get the password prompt  >>>

  debug1: packet_send2: adding 64 (len 50 padlen 14 extra_pad 64)
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  Permission denied, please try again.
  debug1: packet_send2: adding 64 (len 50 padlen 14 extra_pad 64)
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  Permission denied, please try again.
  debug1: packet_send2: adding 64 (len 50 padlen 14 extra_pad 64)
  debug1: authentications that can continue:
  publickey,password,keyboard-interacti
  debug1: no more auth methods to try
  Permission denied (publickey,password,keyboard-interactive).
  debug1: Calling cleanup 0x4271e0(0x0)

Is this enough information someone can tell me what I'm doing wrong?
It's like standard in is taking data from someplace other than the
keyboard.

Thanks for your help.


Mike Spaniol
Blackwell North America, Inc.
6024 SW Jean Rd., Bldg. G
Lake Oswego, OR   97035
(503) 684-1140 Ext. 1231

Re: Using ram disks

[Rob van der Heij]

>I'll have to browse around in /usr/src/linux/Documentation more.
>It looks like there is a lot of good stuff in there.

Or the quick way: ;-)
 mke2fs /dev/ram0
 mkdir /mnt/mini
 mount /dev/ram0 /mnt/mini

Re: Linux backup software

[John Summerfield]

> On Wed, May 22, 2002 at 01:01:55PM -0700, Jon R. Doyle wrote:
>
> > Well you said Freeware, and the best fit it tar, in fact I think it is
> > better than some commercial stuff, as long as you do not need so
> > called "agents" like connectors to Oracle or something for Hot
> > back-up/restore.  You can make some really neat scripts for tar using
> > cron jobs, at jobs, and even rsync/ssh to do some extensive backup
> > routines. Tar also can be called to compress and use tape devices.
>
> Um, suit yourself, but if you're going in that direction, I'd recommend
> Amanda, which uses either tar or dump under the covers, but at least
> gives you rudimentary tape management.


I rather like afio.


--
Cheers
John Summerfield

Microsoft's most solid OS: http://www.geocities.com/rcwoolley/

Note: mail delivered to me is deemed to be intended for me, for my disposition.

==
If you don't like being told you're wrong,
be right!

Re: Sanity Check! Steps to apply Updates -- One more thought...chroot

[John Summerfield]

> The thought comes to mind that IF I were to follow my
> OS/390 maintenance procedures, I might consider
> cloning this system,  then from my production system issue a mount of cloned
> root
> to /SERVICE and chroot to /SERVICE to apply these RPM's.
>
> Is that workable, prudent etc. or does it cause too much pain (for reasons
> unknown to this rookie!).

Backing up the system can't hurt. Other than that, take it out of production
(along with any sharing its filesystems), go to single-user mode (or shut down
services not needed for the moment), upgrade and reboot.

Or clone, update the clone and switch. This is probably better if you're sharing
filesystems because those systems can be reipled at times more convenient to
them.



--
Cheers
John Summerfield

Microsoft's most solid OS: http://www.geocities.com/rcwoolley/

Note: mail delivered to me is deemed to be intended for me, for my disposition.

==
If you don't like being told you're wrong,
be right!

Re: Bastille

[Frank DeGilio]

Yeah - I set an environment variable (I can't remember which offhand) that
defines the system as a redhat system then I ran Interactive Bastille with
the --norequires parm.
You do get questions that don't apply but I found it somewhat educational
to run it that way

Frank J. De Gilio
Complex Solution Architect
IBM Design Center for e-transaction processing
Phone (845) 435-5033  (Tie 295-5033)
Cell Phone: (845) 546-1130


|-+>
| |   Thomas Emde  |
| | |
| |   Sent by: Linux on|
| |   390 Port |
| |   <[EMAIL PROTECTED]|
| |   IST.EDU> |
| ||
| ||
| |   05/21/2002 09:35 |
| |   AM   |
| |   Please respond to|
| |   Linux on 390 Port|
| ||
|-+>
  
>--|
  |
  |
  |   To:   [EMAIL PROTECTED]
  |
  |   cc:  
  |
  |   Subject:  Bastille   
  |
  |
  |
  |
  |
  
>--|




This distribution version: SE7.0 is not yet supported!

Is there any chance to get it running with this distribution?

Re: Red Hat 7.2 install - problem with fdasd

[Makhijani, Beena]

Thanks, using dasdfmt did solve the problem and I also had to IPL after
dasdfmt before I could run Anaconda.

Beena

-Original Message-
From: Rob van der Heij [mailto:[EMAIL PROTECTED]]
Sent: Friday, May 17, 2002 6:33 AM
To: [EMAIL PROTECTED]
Subject: Re: Red Hat 7.2 install - problem with fdasd


At 13:34 17-05-02, Alessandro Mucioli Travia wrote:

>Hello Beena, try to use dasdfmt before the fdasd against the volumes where
>the problem occurs.

I think the re-IPL addresses the situation where even fdasd will
fail after dasdfmt was done, unless you IPL in between.

Depending on the world you come from, people may think that it is
enough to only format cylinder 0 and label the volume. It is, but
only when you dasdfmt the disk before using fdasd.

Rob

Re: Linux backup software

[Neil Carson]

Thank you to everyone who replied.
...Neil

>>> [EMAIL PROTECTED] 05/22/02 04:16PM >>>
On Wed, May 22, 2002 at 01:01:55PM -0700, Jon R. Doyle wrote:

> Well you said Freeware, and the best fit it tar, in fact I think it is
> better than some commercial stuff, as long as you do not need so
> called "agents" like connectors to Oracle or something for Hot
> back-up/restore.  You can make some really neat scripts for tar using
> cron jobs, at jobs, and even rsync/ssh to do some extensive backup
> routines. Tar also can be called to compress and use tape devices.

Um, suit yourself, but if you're going in that direction, I'd recommend
Amanda, which uses either tar or dump under the covers, but at least
gives you rudimentary tape management.

Adam

Re: Linux backup software

[Adam Thornton]

On Wed, May 22, 2002 at 01:01:55PM -0700, Jon R. Doyle wrote:

> Well you said Freeware, and the best fit it tar, in fact I think it is
> better than some commercial stuff, as long as you do not need so
> called "agents" like connectors to Oracle or something for Hot
> back-up/restore.  You can make some really neat scripts for tar using
> cron jobs, at jobs, and even rsync/ssh to do some extensive backup
> routines. Tar also can be called to compress and use tape devices.

Um, suit yourself, but if you're going in that direction, I'd recommend
Amanda, which uses either tar or dump under the covers, but at least
gives you rudimentary tape management.

Adam

Re: Linux backup software

[Jim Elliott]

> I apologize if this has been asked before. What Linux software is
> available to perform logical backups of Linux S/390 files and
> directories? Ideally we would be looking for 'freeware or shareware'
> but if you know of any commercial ware that would be appreciated also.
> We would be looking for software that can backup on to 3490s and
> 3480s.

Neil: For an Open Source solution check out Amanda at
http://www.amanda.org

Regards, Jim

Re: kernel bug with samba

[Hall, Ken (ECSS)]

This is a known bug.  There have been two kernel upgrades since the pre-release 
version.  I assume that's what you're running, since the GA version had the first 
upgrade integrated.

If you have a maintenance agreement, you can get the new kernel packages from SuSE.  
If not, I guess you're out of luck, unless you want to download the 2.4.7 kernel from 
somewhere else, patch it, and
compile it.  Problem is, you need a level that's compatible with the IBM OCO modules.


> -Original Message-
> From: Tim-Chr. Hanschen [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, May 22, 2002 7:36 AM
> To: [EMAIL PROTECTED]
> Subject: [LINUX-390] kernel bug with samba
>
>
> Hi,
>
> I installed SLES 7.2 (kernel 2.4.5) on an IBM 2064. My samba
> version is
> 2.2.0a.
>
> Everything seems to work fine. I can map devices to my
> windows workstation
> and I can read files from the device.
> When I tried to write to the device the whole linux stop
> working. I got the
>  following error:
>
> illegal operation: 0001
> cpu:1
> Process smbd (pid: 465, stackpage=00FB9000)
> Kernel GPRS:
>     001b  0029b518
> 8005b520    005b9060  88c2
> 0003  0014  00fb8000  0003
> 0122e7bc  8005b480  8005b520  00fb9bb8
> Kernel ACRS:
>       
> 0001      
>       
>       
> Kernel BackChain  CallChain
>00fb9bb8   [<0005b520>]
>00fbb998   [<0005b612>]
>00fb9cf8   [<000658dc>]
>00fb9d58   [<000578ee>]
>00fb9db8   [<00057aec>]
>00fb9e28   [<0004674e>]
>00fb9ea8   [<00046c14>]
>00fb9f08   [<000130d0>]
> May 22 13:30:24 osa-rzl1 kernel: kernel BUG at fcntl.c:3971
>
> I rebooted the system and tried again the whole story. I am using
> lvm-0.9.1_beta7-14 for the dasds.
>
> TIA,
>   - Tim -
>

Re: Linux backup software

[Jon R. Doyle]

Well you said Freeware, and the best fit it tar, in fact I think it is
better than some commercial stuff, as long as you do not need so called
"agents" like connectors to Oracle or something for Hot back-up/restore.
You can make some really neat scripts for tar using cron jobs, at jobs,
and even rsync/ssh to do some extensive backup routines. Tar also can be
called to compress and use tape devices.


Regards,

Jon


Jon R. Doyle
Sendmail Inc.
6425 Christie Ave
Emeryville, Ca. 94608


   (o_
   (o_   (o_   //\
   (/)_  (\)_  V_/_



On Wed, 22 May 2002, Konkol, Josh wrote:

> We use FDR/Upstream.  It appears to work fine.
>
> Josh
>
> -Original Message-
> From: PAUL WILLIAMSON [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, May 22, 2002 10:10 AM
> To: [EMAIL PROTECTED]
> Subject: Re: Linux backup software
>
>
> I'm not sure if this will work, but Arkeia is pretty good, and is
> very good at managing tape libraries and such.  It is pretty
> involved, but once you get the hang of it, you'll really like it (IMHO).
>
> http://www.arkeia.com/
>
> Paul
>
> >>> [EMAIL PROTECTED] 05/22/02 10:33AM >>>
> I apologize if this has been asked before.
> What Linux software is available to perform logical backups of
> Linux S/390 files and directories?
> Ideally we would be looking for 'freeware or shareware' but if you know
> of any commercial ware that would be appreciated also.
> We would be looking for software that can backup on to 3490s and 3480s.
>
> Thanks in advance...Neil
>

Re: Linux backup software

[Konkol, Josh]

We use FDR/Upstream.  It appears to work fine.

Josh

-Original Message-
From: PAUL WILLIAMSON [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, May 22, 2002 10:10 AM
To: [EMAIL PROTECTED]
Subject: Re: Linux backup software


I'm not sure if this will work, but Arkeia is pretty good, and is
very good at managing tape libraries and such.  It is pretty
involved, but once you get the hang of it, you'll really like it (IMHO).

http://www.arkeia.com/

Paul

>>> [EMAIL PROTECTED] 05/22/02 10:33AM >>>
I apologize if this has been asked before.
What Linux software is available to perform logical backups of
Linux S/390 files and directories?
Ideally we would be looking for 'freeware or shareware' but if you know
of any commercial ware that would be appreciated also.
We would be looking for software that can backup on to 3490s and 3480s.

Thanks in advance...Neil

Re: Sanity Check! Steps to apply Updates -- One more thought...chroot

[Dave Myers]

The thought comes to mind that IF I were to follow my
OS/390 maintenance procedures, I might consider
cloning this system,  then from my production system issue a mount of cloned
root
to /SERVICE and chroot to /SERVICE to apply these RPM's.

Is that workable, prudent etc. or does it cause too much pain (for reasons
unknown to this rookie!).

Thanks,
Dave

Re: Sanity Check! Steps to apply Updates -- Followup Questions

[daniel . jarboe]

[EMAIL PROTECTED] wrote:

>
>3.  Observation:  Also read that RPM itself may need to be updated first,
>before running the other RPMs; however,
>in the S/390 updates I see not RPM for RPM, so I'm assuming I'll be ok.
>
>
You'll probably want to look at more than just the s390 updates.
 Unfortunately not all the current RH errata RPMs are available
precompiled for s390 yet, and this doesn't necessarily mean they aren't
applicable.  Check out http://rhn.redhat.com/errata/rh72-errata.html to
see what 7.2 errata updates are available as src.rpm's but not as
s390.rpm's.  If you have an older version installed already, download
the latest srpm and see if you can --rebuild it for s390.  You'll want
to do this before you bring your system down to single-user mode because
compiling these things can sometimes take a LONG time.  Not everything
will rebuild successfully (perhaps why they aren't available for s390
yet).  But many will and then you'll have a nice shiny new s390.rpm
ready to be installed.  Btw, if you try to rebuild the Perl 5.6.1
src.rpm, please let me know if you were successful (mine failed on the
study() test).

rpm -qi (product name without version) will summarize installed packages
including it's version number near the top.
rpm -qip (filename) is pretty much the same thing except it works with
rpm files instead of an installed package

As far as being live, I've found stuff to be pretty resilliant (to the
point of removing openssh and server without affecting my currently open
ssh sessions).  Better safe than sorry though.

~ Daniel

Re: Sanity Check please! Steps to apply Updates to a RH 7.2 system

[John Summerfield]

> > I always update my IA32 systems while they're live, but I think it can go a
> wry.
> > Think of something starting up while you're updating glibc - it could get s
> ome
> > old versions of libraries and some new.
>
> The application will be loading backward compatible libraries


glibc may not be the best example; I chose it simply because everything uses it.
I can imagine that a bug-fix in something might prevent complete compatibility.

Call me Doubting John if you wish. It's true I've never had a problem, but I've
never had a critical workload to worry over either.

>
> > Additionally, for some packages there can be configuration-file changes. As
>  so
> > often, it depends.
> >
> > Taking it out of service is always safe, but not always essential.
>
> I work on the rule of scheduling potential downtime but hoping its not
> needed. I've done 7.2->7.3 updates on x86 without downtime (barring kernel)

I did extensive upgrades on 4.2 too, remotely and without a problem.
--
Cheers
John Summerfield

Microsoft's most solid OS: http://www.geocities.com/rcwoolley/

Note: mail delivered to me is deemed to be intended for me, for my disposition.

==
If you don't like being told you're wrong,
be right!

Re: Sanity Check! Steps to apply Updates -- Followup Questions

[Post, Mark K]

Dave,

Take the highest-numbered kernel that has an existing IBM OCO package.  You
will only need one.  You won't need the kernel-BOOT one, since that's used
for creating installation images.  Whether you install the other ones
depends on your needs.  Take a look at their contents (rpm -qlp
package.name.rpm) and see if you want them.  There will be some overlap, so
watch out for that.  If you decide to install the kernel-source one, you
won't need the kernel-headers one, since that is just a subset of the full
kernel source.

All of your file systems should still be mounted in single user mode.  If
not, do a "mount -a" command and they will be.  RPM will definitely run at
that point, but I imagine it would anyway.

Mark Post

-Original Message-
From: Dave Myers [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, May 22, 2002 1:54 PM
To: [EMAIL PROTECTED]
Subject: Re: Sanity Check! Steps to apply Updates -- Followup Questions


Thanks for the replies.
Couple of followup questions:

1. I noticed there are multiple levels of  kernel-** RPM's out there.
Do I just  update using the highest numbered ones?  (i.e. just
kernel-2.4.9-37...)
or do I have to do both the 2.4.9-31 and then the 2.4.9-37 ??

Also, does it matter what order I updated the various RPM's.
(i.e.  kernel-   kernel-BOOT,  kernel-doc,  kernel-headers,  kernel-source
etc.)


2. I read about single user mode.  Seems like not much is mounted or running
in single user mode.
Will RPM run OK in this mode?

3.  Observation:  Also read that RPM itself may need to be updated first,
before running the other RPMs; however,
in the S/390 updates I see not RPM for RPM, so I'm assuming I'll be ok.

Tia
Dave



In a message dated 5/22/2002 11:18:34 AM Mountain Daylight Time,
[EMAIL PROTECTED] writes:


> Dave,
>
> I haven't looked at all the updates on the Red Hat site, so I can't be
> sure,
> but you will probably want to take your system down to single user mode
> (telinit -s) before you install the RPMs.  For updates that replace core
> system components, this is usually considered best practice.  If there's a
> kernel update in there, and I think there is, you'll need to download the
> corresponding OCO modules from IBM to replace the one(s) you have now.
>
> Mark Post
>

Re: mprotect() doesn't not even work...

[Ferguson, Neale]

Is this the same bit of logic in 2.4.7:

good_area:
write = 0;
si_code = SEGV_ACCERR;

switch (error_code & 0xFF) {
case 0x04:/* write,
present*/
write = 1;
break;
case 0x10:   /* not
present*/
case 0x11:   /* not
present*/
if (!(vma->vm_flags & (VM_READ | VM_EXEC |
VM_WRITE)))
goto bad_area;
 Insert patch here?
break;
default:
   printk("code should be 4, 10 or 11 (%lX)
\n",error_code&0xFF);
   goto bad_area;
}


> -Original Message-
> Hi Jason,
> > I've tried the attached simple program on RedHat 2.4.9-17 and
> > the lastest 2.4.17 build from IBM, and both fail miserably:
>
> This is a bug in arch/{s390,s390x}/mm/fault.c that is present
> since the
> beginning. A small wonder that nobody noticed it until now...
> The following patch should take care of it:
>
> diff -urN linux-2.4.17/arch/s390/mm/fault.c
> linux-2.4.17-s390/arch/s390/mm/fault.c
> --- linux-2.4.17/arch/s390/mm/fault.c Wed May 22 20:18:55 2002
> +++ linux-2.4.17-s390/arch/s390/mm/fault.c  Wed May 22
> 20:14:10 2002
> @@ -222,6 +222,9 @@
> /* page not present, check vm flags */
> if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
>   goto bad_area;
> + } else {
> +   if (!(vma->vm_flags & VM_WRITE))
> + goto bad_area;
>   }

Re: mprotect() doesn't not even work...

[Martin Schwidefsky]

Hi Jason,
> I've tried the attached simple program on RedHat 2.4.9-17 and
> the lastest 2.4.17 build from IBM, and both fail miserably:

This is a bug in arch/{s390,s390x}/mm/fault.c that is present since the
beginning. A small wonder that nobody noticed it until now...
The following patch should take care of it:

diff -urN linux-2.4.17/arch/s390/mm/fault.c linux-2.4.17-s390/arch/s390/mm/fault.c
--- linux-2.4.17/arch/s390/mm/fault.c Wed May 22 20:18:55 2002
+++ linux-2.4.17-s390/arch/s390/mm/fault.c  Wed May 22 20:14:10 2002
@@ -222,6 +222,9 @@
/* page not present, check vm flags */
if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
  goto bad_area;
+ } else {
+   if (!(vma->vm_flags & VM_WRITE))
+ goto bad_area;
  }

 survive:
diff -urN linux-2.4.17/arch/s390x/mm/fault.c linux-2.4.17-s390/arch/s390x/mm/fault.c
--- linux-2.4.17/arch/s390x/mm/fault.cWed May 22 20:19:08 2002
+++ linux-2.4.17-s390/arch/s390x/mm/fault.c Wed May 22 20:14:10 2002
@@ -222,6 +222,9 @@
/* page not present, check vm flags */
if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
  goto bad_area;
+ } else {
+   if (!(vma->vm_flags & VM_WRITE))
+ goto bad_area;
  }

 survive:

blue skies,
   Martin

Linux/390 Design & Development, IBM Deutschland Entwicklung GmbH
Schönaicherstr. 220, D-71032 Böblingen, Telefon: 49 - (0)7031 - 16-2247
E-Mail: [EMAIL PROTECTED]

Re: mprotect() doesn't not even work...

[Ferguson, Neale]

It appears mprotect is setting the write protect bit of the storage key, but
this is being undone shortly after:

 -> 00031F1E' SSKE  B22B001300060CF0D203CC 1 <--- Write okay
when storage got by mmap
 -> 00400790" ST50102000 >> 40018000CC 1 <--- 1234
stored
V40018000  123406 L0CF0D000

 -> 0003A168' SSKE  B22B00020CF0D201CC 1 <--- Write
disabled
 -> 00031D36' SSKE  B22B001200060CF0D001CC 1 <--- Write
re-enabled
 -> 004007E4" ST50201000 >> 40018000CC 0 <--- 1235
stored
V40018000  123506 L0CF0D000

According to my System.map 0x31d36 is +22e in do_wp_page() which is in
mm/memory.c.

Re: Sanity Check! Steps to apply Updates -- Followup Questions

[Dave Myers]

Thanks for the replies.
Couple of followup questions:

1. I noticed there are multiple levels of  kernel-** RPM's out there.
Do I just  update using the highest numbered ones?  (i.e. just
kernel-2.4.9-37...)
or do I have to do both the 2.4.9-31 and then the 2.4.9-37 ??

Also, does it matter what order I updated the various RPM's.
(i.e.  kernel-   kernel-BOOT,  kernel-doc,  kernel-headers,  kernel-source
etc.)


2. I read about single user mode.  Seems like not much is mounted or running
in single user mode.
Will RPM run OK in this mode?

3.  Observation:  Also read that RPM itself may need to be updated first,
before running the other RPMs; however,
in the S/390 updates I see not RPM for RPM, so I'm assuming I'll be ok.

Tia
Dave



In a message dated 5/22/2002 11:18:34 AM Mountain Daylight Time,
[EMAIL PROTECTED] writes:


> Dave,
>
> I haven't looked at all the updates on the Red Hat site, so I can't be
> sure,
> but you will probably want to take your system down to single user mode
> (telinit -s) before you install the RPMs.  For updates that replace core
> system components, this is usually considered best practice.  If there's a
> kernel update in there, and I think there is, you'll need to download the
> corresponding OCO modules from IBM to replace the one(s) you have now.
>
> Mark Post
>

Re: Communicating with "Console" on HMC (SERVC instruction?)

[Romney White]

Brad:

I'm not aware of any published documentation that describes the
integrated console interface. I believe that the intellectual
property associated with this interface was released by virtue
of its disclosure in the GPL source code for the console driver
in Linux for S/390. However, I'm not aware of any associated
documentation.

Romney

On Tue, 21 May 2002 08:42:15 -0400 Snyder, Bradley (LNG) said:
>Does anyone know how to send messages to the "console" (i.e. operating
>system messages item) on the HMC?  Please DO NOT, and I repeat DO NOT
>respond to this with any GPL-licensed code examples.  I'm looking for a
>general description or a pointer in the right direction.  The closest I have
>come is the SERVC instruction, which is undocumented, and, as far as I can
>tell, not supported in GAS.
>
>Thanks,
>--Brad

Re: Sanity Check please! Steps to apply Updates to a RH 7.2 system

[Alan Cox]

> I always update my IA32 systems while they're live, but I think it can go awry.
> Think of something starting up while you're updating glibc - it could get some
> old versions of libraries and some new.

The application will be loading backward compatible libraries

> Additionally, for some packages there can be configuration-file changes. As so
> often, it depends.
>
> Taking it out of service is always safe, but not always essential.

I work on the rule of scheduling potential downtime but hoping its not
needed. I've done 7.2->7.3 updates on x86 without downtime (barring kernel)

Re: Sanity Check please! Steps to apply Updates to a RH 7.2 system

[John Summerfield]

> I have been tasked to upgrade a RH 7.2 system.
> Are these correct steps??
>
> Is it OK to run the RPM updates against a "live" system,
> assuming I have backed up first using OFFLINDR ?

I always update my IA32 systems while they're live, but I think it can go awry.
Think of something starting up while you're updating glibc - it could get some
old versions of libraries and some new.

Additionally, for some packages there can be configuration-file changes. As so
often, it depends.

Taking it out of service is always safe, but not always essential.



--
Cheers
John Summerfield

Microsoft's most solid OS: http://www.geocities.com/rcwoolley/

Note: mail delivered to me is deemed to be intended for me, for my disposition.

==
If you don't like being told you're wrong,
be right!

Re: Adding a DASD volume

[Sergey Korzhevsky]

22.05.2002 08:06:27 "Felix Yoyok S. Aditias"  wrote:

>Hi all,
>
>I get message of server not found when searching in
>http://www.marist.edu:8000/htbin/wlvdbs?LINUX-VM
>
>I use suse linux 7.0 with kernel 2.2.16 ipl-ed from dasd.
>When I want to add a DASD volume, I try :
>insmod dasd dasd=2123
>but this message came out :
>insmod: dasd: no module by that name found

If you say about already installed system, dasd module already incorporated
into kernel.


>I tried to find module dasd.o but it's not there, except the source
>(dasd.c).
>What I want to do to add a DASD volume is :
>1. insmod dasd -> in order to register the volume in /proc/dasd/devices
>2. dasdfmt -> to format the volume

You must add address of this volume to parmfile  (dasd=...) and run silo.
Then, reIPL, dasdfmt, mkfs, etc...

>If the steps are not correct, which documentation should I refer to?

On CD1 you can read some pdfs.




WBR, Sergey

Re: Sanity Check please! Steps to apply Updates to a RH 7.2 syste m

[Post, Mark K]

Dave,

I haven't looked at all the updates on the Red Hat site, so I can't be sure,
but you will probably want to take your system down to single user mode
(telinit -s) before you install the RPMs.  For updates that replace core
system components, this is usually considered best practice.  If there's a
kernel update in there, and I think there is, you'll need to download the
corresponding OCO modules from IBM to replace the one(s) you have now.

Mark Post

-Original Message-
From: Dave Myers [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, May 22, 2002 12:19 PM
To: [EMAIL PROTECTED]
Subject: Sanity Check please! Steps to apply Updates to a RH 7.2 system


I have been tasked to upgrade a RH 7.2 system.
Are these correct steps??

Is it OK to run the RPM updates against a "live" system,
assuming I have backed up first using OFFLINDR ?

I'm not sure I really need to do step3 ???


TIA
Dave

1. Backup/Clone RH 7.2
2. Apply all RPM updates to RH 7.2 system from
ftp://ftp.redhat.com/pub/redhat/linux/updates/7.2/en/os/s390/";>ftp://f
tp.redhat.com/pub/redhat/linux/updates/7.2/en/os/s390/
3. Reapply the LCS OCO Driver to the newly updated system.
4. Rewrite boot sector with zipl util
5. Backup new RH 7.2  using the OFFLINDR util.
6. Test new RH 7.2

Re: Runlevel3 start up problems

[Post, Mark K]

John,

I just put an update on the linuxvm.org site about this.  It points to two
entries in the mailing list archives you might want to check out:
http://www.marist.edu/htbin/wlvtype?LINUX-VM.25655
http://www.marist.edu/htbin/wlvtype?LINUX-VM.25677

Mark Post

-Original Message-
From: Gustavson, John (ECSS) [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, May 22, 2002 8:37 AM
To: [EMAIL PROTECTED]
Subject: Runlevel3 start up problems


We are running SUSE 2.4.7 kernel.  We have had problems with the start-up
scripts in /etc/init.d(rc.d)/rc3.d for Apache, TOMCAT, and ucd-snmpd.  We
put several echo's of return codes in the scripts,
which all return 0, and it looks like the scripts are running fine, but the
processes do not continue.  We did use the no-hup for ucd-snmpd, but we do
not want to have to nohup every thing.  Apache
start-up problem does seem to be somewhat intermittent.  After we are fully
up, these same scripts can be run manually with out a problem.  It appears
to be only related to system boot/re-boot.  Does
anyone else have these problems at start-up?

Regards

John Gustavson
Enterprise Central Software Services (ECSS)
570 Washington Street - 2nd floor
New York, New York, 10080-6802

Telephone: 1-212-647-3793
Fax: 1-212-647-3321
Email: [EMAIL PROTECTED]

Re: Linux backup software

[Post, Mark K]

Neil,

In response to an off-list inquiry recently about this, I suggested taking a
look at these packages:
Amanda, as the Open Source alternative
Computer Associates' BrightStor
Innovation's FDRINSTANT and FDR/UPSTREAM
SecureAgent Software's SecureBackup
Tivoli's TSM client (you'd need to have a server somewhere already, or have
tape access on your system.  There is a server for VM, but they've neglected
it for years.)

Mark Post

-Original Message-
From: Neil Carson [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, May 22, 2002 10:33 AM
To: [EMAIL PROTECTED]
Subject: Linux backup software


I apologize if this has been asked before.
What Linux software is available to perform logical backups of
Linux S/390 files and directories?
Ideally we would be looking for 'freeware or shareware' but if you know
of any commercial ware that would be appreciated also.
We would be looking for software that can backup on to 3490s and 3480s.

Thanks in advance...Neil

Re: Strange problem [Suse Linux on S390 VM]

[Horst Hummel]

Donal,

the sense data you get are indicating that you might have problems with
your hardware (DASD / path to the DASD).
This is a so called 'Message to Operator' and the error recovery action
code means that the device driver should try an other path to the device
(if possible).

If you can't find any problem with your hardware, let me know more about
the used storage server, please.


Mit freundlichen Gruessen / Best regards
Horst Hummel
__
Linux for eServer Development (D/3303)
Schoenaicher Str. 220, 71032 Boeblingen, Germany
Internet:[EMAIL PROTECTED]
Phone: ext. +49-(0)7031/16-3643, int. *120-3643



|-+>
| |   Carsten Otte |
| ||
| |   05/22/02 10:11 AM|
| ||
|-+>
  
>--|
  |
  |
  |   To:   Horst Hummel/Germany/IBM   
  |
  |   cc:  
  |
  |   From: Carsten Otte/Germany/IBM@IBMDE 
  |
  |   Subject:  Re: Strange problem [Suse Linux on S390 VM]
  |
  |
  |
  |
  |
  
>--|





mit freundlichem Gruß / with kind regards
Carsten Otte

IBM Deutschland Entwicklung GmbH
Linux for eServer development - device driver team
Phone: +49/07031/16-4076
IBM internal phone: *120-4076
--
We are Linux.
Resistance indicates that you're missing the point!
- Forwarded by Carsten Otte/Germany/IBM on 05/22/02 10:11 AM -
|-+>
| |   "Post, Mark K"   |
| |   <[EMAIL PROTECTED]|
| |   m>   |
| |   Sent by: Linux on|
| |   390 Port |
| |   <[EMAIL PROTECTED]|
| |   IST.EDU> |
| ||
| ||
| |   05/18/02 11:30 PM|
| |   Please respond to|
| |   Linux on 390 Port|
| ||
|-+>
  
>--|
  |
  |
  |   To:   [EMAIL PROTECTED]
  |
  |   cc:  
  |
  |   Subject:  Re: Strange problem [Suse Linux on S390 VM]
  |
  |
  |
  |
  |
  
>--|



Donal,

This looks like a possible hardware problem.  I would have your IBM CE look
at it for you.

Mark Post

-Original Message-
From: Donal Hunt [mailto:[EMAIL PROTECTED]]
Sent: Saturday, May 18, 2002 4:26 PM
To: [EMAIL PROTECTED]
Subject: Strange problem [Suse Linux on S390 VM]


Hi there,

I was pointed in this direction for advice I'm having with a VM I'm
using.  I'm using the VM for a project machine running:

- Suse Linux 7.0
- Proftpd (standalone)
- OpenSSH
- Apache 1.3.22 (http and https)
- Mailman 2.1b2
- Postfix as the MTA (replacing the default sendmail install)
- Big Brother monitoring client (to try and keep track of problems).

Recently I've had a problem with ftp (was called from inetd before i
installed proftpd), telnet, SSH and Postfix dying on the machine.  The
processes are still running but they aren't usable. eg telnet 

Re: What PTF level am I at?

[Matt Lashley]

Thank you Dennis.  I'll try all of those and add them to my ever-growing
cheat sheet list.




Dennis
MusselwhiteTo: [EMAIL PROTECTED]
<[EMAIL PROTECTED]cc:
m.com> Subject: Re: What PTF level am I at?
Sent by: Linux
on 390 Port
<[EMAIL PROTECTED]
ARIST.EDU>


05/22/2002
09:56 AM
Please respond
to Linux on 390
Port





Matt,

You can get some idea of the system level using the CP command:

  CP QUERY CPLEVEL

That should (at least) reassure you that you are using z/VM Version 4
Release 2.0 and it
will list a service level that indicates which service tape has been
installed.

CP Guest LAN support on VM 4.2.0 was enabled by CP APAR VM62938 (and TCPIP
APAR PQ51738).  Use the following CP command to find out if VM62938 is
installed:

  CP QUERY VMLAN

If the response starts out with something like this:

VMLAN maintenance level:
  Latest Service: VM62938

you have the necessary CP support to define virtual HiperSockets networks
and adapters.

Regards,
Dennis Musselwhite ([EMAIL PROTECTED])
IBM Corporation -- z/VM Development -- CP Network Simulation

Re: has anyone seen this?

[Joe Jenkins]

You can buy one from thinkgeek.com

The Linux Fish:
http://www.thinkgeek.com/stuff/fun-stuff/2898.shtml


On Tue, 21 May 2002, Dunbar, Maggie wrote:

> Over the weekend I was driving somewhere, and I got behind a car that had
> one of those "fish" symbols on the back bumper - you know, the ones that
> usually have the Greek letters for Christ inside the fish, although there
> are also ones that say "Darwin" and have little feet on the fish.  Well,
> this one had LINUX inside the fish!  I had to look twice to convince myself!
>
>
> Maggie
>

Linux backup software

[Neil Carson]

I apologize if this has been asked before.
What Linux software is available to perform logical backups of
Linux S/390 files and directories?
Ideally we would be looking for 'freeware or shareware' but if you know
of any commercial ware that would be appreciated also.
We would be looking for software that can backup on to 3490s and 3480s.

Thanks in advance...Neil

Tape Services Suite from UTS Global.

[Doug DeMers]

FYI (and apologies for the commercial plug);  thought this would
be of interest.  Here's a text version for the list.


For Immediate Release

UTS Global, LLC Announces Tape Services Suite for Mainframe Linux

Tape Services Suite (TSS) for Linux(r) on S/390(r) and zSeries(tm)

Santa Clara, CA - May 21, 2002 - UTS Global, LLC announced immediate
availability of their TSS product, a comprehensive suite of tape and
tape management service utilities for Linux(r) on the mainframe.   With
the TSS Tape Management Subsystem (TSS-TMS), mainframe Linux(r) users
can utilize 34x0/3590 tapes with the same levels of simplicity,
security and efficiency enjoyed by OS/390(r) systems.   TSS-TMS handles
device allocation, media access and disposition processing, and fully
supports standard tape label processing, including expiration date
handling.

Available optional TSS features include TSS-BAR (native Backup and
Restore), TSS-TSF (Tape Silo Feature for StorageTek(r) tape silos).
Available Q3 2002 is TSS-DTS (Distributed Tape Subsystem) to allow
client system access to mainframe tape devices.

"Our Tape Services Suite capitalizes on our 20 year track record of
providing robust, enterprise-ready software for Unix environments,"
said Keith Schwamkrug, CEO of UTS Global, LLC.  "Using TSS-TMS, Linux
enterprises can benefit from the cost effectiveness of tape media
without compromising the security or integrity of their business data."

UTS Global, LLC

Formed in 2000 as a spin-off from Amdahl(r) Corporation, UTS Global is
the worldwide leader in Unix(r) operating environments for mainframe
business computing. The company offers a wide range of products and
services designed to empower businesses through mission-critical system
software and support.

For more information on Tape Services Suite or UTS Global, LLC, contact:

Douglas DeMers
[EMAIL PROTECTED],
telephone  (408) 496-4230.
http://www.utsglobal.com/linux.html

UTS Global is a trademark of UTS Global, LLC. Linux is a registered
trademark of Linus Torvalds. UNIX is a registered trademark in the
United States and other countries, licensed exclusively (except in
Japan where such license is non-exclusive) through X/Open Company
Limited. ACSLS and StorageTek are trademarks of Storage Technology.
IBM, zSeries, OS/390 and z/OS are trademarks of IBM. Amdahl is a
trademark of Fujitsu IT Holdings.  All other trademarks belong to their
respective holders.

Douglas DeMers   | (408-496-4230)  | [EMAIL PROTECTED]
Linux Product Manager| UTS Global, LLC | [EMAIL PROTECTED]

Re: What PTF level am I at?

[Dennis Musselwhite]

Matt,

You can get some idea of the system level using the CP command:

  CP QUERY CPLEVEL

That should (at least) reassure you that you are using z/VM Version 4
Release 2.0 and it
will list a service level that indicates which service tape has been
installed.

CP Guest LAN support on VM 4.2.0 was enabled by CP APAR VM62938 (and TCPIP
APAR PQ51738).  Use the following CP command to find out if VM62938 is
installed:

  CP QUERY VMLAN

If the response starts out with something like this:

VMLAN maintenance level:
  Latest Service: VM62938

you have the necessary CP support to define virtual HiperSockets networks
and adapters.

Regards,
Dennis Musselwhite ([EMAIL PROTECTED])
IBM Corporation -- z/VM Development -- CP Network Simulation

Re: Using ram disks

[Post, Mark K]

Dennis,

Take a look at "Using the RAM disk block device with Linux" in
/usr/src/linux/Documentation/ramdisk.txt.

Mark Post

-Original Message-
From: Dennis G. Wicks [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, May 21, 2002 4:40 PM
To: [EMAIL PROTECTED]
Subject: Using ram disks


Greetings;

How does one go about getting access to the /dev/ram? disks?

I have tried formatting and making file systems and just mounting
but I can't seem to find the right combination.

Many TIA,
Dennis
--
Dennis G. Wicks Systems Programmer  [EMAIL PROTECTED]
Communications Data Group   Tel: (217)355-7117  Fax: (217)351-6994
102 S. Duncan Rd.
Champaign, IL  61822

Sanity Check please! Steps to apply Updates to a RH 7.2 system

[Dave Myers]

I have been tasked to upgrade a RH 7.2 system.
Are these correct steps??

Is it OK to run the RPM updates against a "live" system,
assuming I have backed up first using OFFLINDR ?

I'm not sure I really need to do step3 ???


TIA
Dave

1. Backup/Clone RH 7.2
2. Apply all RPM updates to RH 7.2 system from
ftp://ftp.redhat.com/pub/redhat/linux/updates/7.2/en/os/s390/";>ftp://ftp.redhat.com/pub/redhat/linux/updates/7.2/en/os/s390/
3. Reapply the LCS OCO Driver to the newly updated system.
4. Rewrite boot sector with zipl util
5. Backup new RH 7.2  using the OFFLINDR util.
6. Test new RH 7.2

[no subject]

[Pamela Christina IBM z/VM - Endicott]

 Subject:  Re: zSeries Technical Conference featuring z/OS, z/VM, and Linux
Hello,

I responded to a note that Bob sent to me directly.
(I read LINUX-390 in a daily roll-up).


Yes, there will be transportation (buses) for the delegates
between the hotels and IBM conference center (morning & evening).


Also, someone inquired about meals.   Lunches are provided
as part of the conference Tues, Wed, and Thurs.
(There may also be a conference dinner and possibly an
 evening reception, but I don't have the specifics on that yet.)


Regards,
Pam C

mprotect() doesn't not even work...

[Jason McMullan]

/ Kernel Problem Report & Test Case

Summary: mprotect() can't set a memory region as read-only

  In the course of our project development, we've noticed a major
flaw in the Linux 2.4.x kernels: in-process memory protection
just doesn't seem to work at all.

  I've tried the attached simple program on RedHat 2.4.9-17 and
the lastest 2.4.17 build from IBM, and both fail miserably:

On the s390 box, under both 2.4.x kernels, I get:

[root@s390devel root]# ./foo
Value of mem is 1234
Protect 0x40018000 with 0x5
Protect err=0 (0)
Value of mem is 1235


The expected output (ie, on an x86 box) is the following:

[root@x86devel root]# ./foo
Value of mem is 1234
Protect 0x40016000 with 0x5
Protect err=0 (0)
Segmentation fault (core dumped)

Note the "Segmentation fault" in the x86 case when I try to
write into the mprotect()ed memory...

*/
#include 
#include 
#include 
#include 
#ifndef PAGESIZE
#define PAGESIZE 4096
#endif

int *mem=NULL;

void doit(void)
{
  fprintf(stderr,"Value of mem is %x\n",*mem);
}

int main(int argc,char **argv)
{
  int err;

  mem=mmap(NULL,PAGESIZE,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANONYMOUS,-1,0);
  *mem=0x1234;
  doit();

  fprintf(stderr,"Protect %p with 0x%x\n",mem,PROT_READ|PROT_EXEC);
  err=mprotect(mem,PAGESIZE, PROT_READ|PROT_EXEC);
  fprintf(stderr,"Protect err=%d (%d)\n",err,errno);
  *mem=0x1235;
  doit();

  return 0;
}

/**
Jason McMullan, Senior Linux Consultant
Linuxcare, Inc. 412.432.6457 tel, 412.656.3519 cell
[EMAIL PROTECTED], http://www.linuxcare.com/
Linuxcare. Putting open source to work. */

Re: Even more strangeness...

[Ferguson, Neale]

Are you running under z/VM? If so, use #CP TR STOR INTO .
to trap overwriting of the area in question.

> -Original Message-
> The first mprotect() is to allow the jump tables to be
> rewritten by the
> run-time linker.  The second is to protect the text segment
> so that nobody
> overwrites the pointer tables.  Well, the pages are not protected, and
> somebody is overwriting the page under some very rare set of
> circumstances
> under select(), but I can't see who is doing it.  Using hardware
> watchpoints under gdb does not reveal the offending instruction, the
> watchpoint was not caught.  This is a bit of a mystery, as hardware
> watchpoints do work under gdb on 390, but the attempt to
> watch a particular
> address within the jump table did not result in any trap when
> the value
> changed.

Using CTCs as serial devices, rather than network interfaces?

[Adam Thornton]

I'm playing around with some of the Linux-HA stuff, and one of the
things I want to try is to heartbeat two images together.  Rather than
doing UDP, I'd like to use vCTCs as a straight serial interface.  How do
I use the CTC as if it were a tty?

Adam

Re: has anyone seen this?

[Scott Courtney]

On Tuesday 21 May 2002 09:52 am, Dunbar, Maggie wrote:
> Over the weekend I was driving somewhere, and I got behind a car that had
> one of those "fish" symbols on the back bumper - you know, the ones that
> usually have the Greek letters for Christ inside the fish, although there
> are also ones that say "Darwin" and have little feet on the fish.  Well,
> this one had LINUX inside the fish!  I had to look twice to convince
> myself!



You mainframers don't hang out at enough S-F conventions. Check this URL:

http://www.thinkgeek.com/stuff/fun-stuff/stickers-emblems.shtml

Now if we could just get them to mass-produce the sticker that says, "Never
trust a computer you can lift."



Scott

--
---+--
Scott Courtney | "I don't mind Microsoft making money. I mind them
[EMAIL PROTECTED]   | having a bad operating system."-- Linus Torvalds
http://www.4th.com/| ("The Rebel Code," NY Times, 21 February 1999)

Re: has anyone seen this?

[Daniel Casey]

One of my favorite "geek-stuff" websites:

http://www.thinkgeek.com/stuff/fun-stuff/2898.shtml





"Dunbar,
Maggie"   To: [EMAIL PROTECTED]
  Subject: has anyone seen this?
Sent by: Linux
on 390 Port
<[EMAIL PROTECTED]
ARIST.EDU>


05-21-2002
08:52
Please respond
to Linux on 390
Port






Over the weekend I was driving somewhere, and I got behind a car that had
one of those "fish" symbols on the back bumper - you know, the ones that
usually have the Greek letters for Christ inside the fish, although there
are also ones that say "Darwin" and have little feet on the fish.  Well,
this one had LINUX inside the fish!  I had to look twice to convince
myself!


Maggie

Re: has anyone seen this?

[Rick Troth]

> this one had LINUX inside the fish!

Perhaps a messianic view of Linux?

Bastille

[Thomas Emde]

Hello,

I installed Bastille 1.3.0 under SuSE Linux/390 7.0 (2.2.16 Kernel). When I
enter "InteractiveBastille", the following is returned:


Using Curses user interface module.
Only displaying questions relevant to the current configuration.
This distribution version: SE7.0 is not yet supported!

Is there any chance to get it running with this distribution?


mit freundlichen Grüßen/with best regards
Thomas Emde

ScaleOn GmbH & Co. KG
Systems Engineering 1
Geb. B151, Raum 117
D-51368 Leverkusen
Telefon +49 214/30-67603
Telefax +49 214/30-24887
E-Mail  [EMAIL PROTECTED]
Internethttp://www.scaleon.de

Hercules 2.16.2 announcement

[Jay Maynard]

What's new in release 2.16.2

Release date: 20 May 2002
Fixed 3350 dasdtab entry (Greg Smith)
Fixed 370 interval timer error (Valery Pogonchenko)
Control panel attach command bug fix (David "Fish" Trout)

The latter is a fix for my screwup. Sorry about that, folks.

All users should install this version at their convenience; the attach
command bug causes Hercules to crash when issued. The other two bug fixes
affect 370-mode OS users. (I don't think there's an ESA-mode OS that
supports 3350s, is there?)

Re: kernel bug with samba

[Sergey Korzhevsky]

WBR, Sergey
- Forwarded by Sergey Korzhevsky/IBA on 22.05.02 18:51 -

daniel.jarboe@cu
stserv.com  To: [EMAIL PROTECTED]
Sent by: Linux  cc:
on 390 Port Subject: Re: Samba on Redhat 7.2
<[EMAIL PROTECTED]
RIST.EDU>


27.03.02 21:32
Please respond
to Linux on 390
Port






This fixed my similar problems:
 http://www.marist.edu/htbin/wlvtype?LINUX-VM.24614

~ Daniel

[EMAIL PROTECTED] wrote:

>Hi!
>
>I am currently running Samba version 2.2.1a-5 on my Redhat 7.2 system.  I
>can bring up Samba just fine and also connect to my share(s) from my
Windows
>2000 pc.  However, when I try to get to any of the files on my share it
>appears that the kernel blows up with an illegal operation:  0001.  I am
>also running kernel version 2.4.9-17.
>
>Any help would be greatly appreciated.
>
>Thanks!
>

Re: Runlevel3 start up problems

[Tim Verhoeven]

On Wed, 22 May 2002, Gustavson, John (ECSS) wrote:

> We are running SUSE 2.4.7 kernel.  We have had problems with the start-up scripts in 
>/etc/init.d(rc.d)/rc3.d for Apache, TOMCAT, and ucd-snmpd.  We put several echo's of 
>return codes in the scripts,
> which all return 0, and it looks like the scripts are running fine, but the 
>processes do not continue.  We did use the no-hup for ucd-snmpd, but we do not want 
>to have to nohup every thing.  Apache
> start-up problem does seem to be somewhat intermittent.  After we are fully up, 
>these same scripts can be run manually with out a problem.  It appears to be only 
>related to system boot/re-boot.  Does
> anyone else have these problems at start-up?

We are having simular problems. Only it's SuSE 7.0 and only Tomcat. But
the symtoms are the same. We worked aroundit by starting Tomcat using at
in the boot script.

We just installed SLES 7 and I didn't had a change to test is the problem
is still there.

Regards,
Tim

--
===
Tim Verhoeven
Linux & Open Source Specialist
GSM : 0496 / 693 453  + e-business solutions
Email : [EMAIL PROTECTED]   + consulting
URL : www.sin.khk.be/~dj/ + Server consolidation
===

Re: Runlevel3 start up problems

[Sergey Korzhevsky]

22.05.2002 15:37:08 "Gustavson, John (ECSS)"  wrote:

>We are running SUSE 2.4.7 kernel.  We have had problems with the start-up
>scripts in /etc/init.d(rc.d)/rc3.d for Apache, TOMCAT, and ucd-snmpd.  We
put
>several echo's of return codes in the scripts,
>which all return 0, and it looks like the scripts are running fine, but
the
>processes do not continue.  We did use the no-hup for ucd-snmpd, but we do
not
>want to have to nohup every thing.  Apache
>start-up problem does seem to be somewhat intermittent.  After we are
fully up,
>these same scripts can be run manually with out a problem.  It appears to
be
>only related to system boot/re-boot.  Does
>anyone else have these problems at start-up?

1. I had this problem with Apache+SSL. Unfortunatly, I can't connect to my
system now. But I remember, somewhere in config files you can increase
timeout for Apache startup. Look at startup script for apache.

2. I don't know, how run Tomcat without nohup.

WBR, Sergey

Re: Runlevel3 start up problems

[Yuval Turgeman]

Sergey Korzhevsky wrote:

>22.05.2002 15:37:08 "Gustavson, John (ECSS)"  wrote:
>
>
>
>>We are running SUSE 2.4.7 kernel.  We have had problems with the start-up
>>scripts in /etc/init.d(rc.d)/rc3.d for Apache, TOMCAT, and ucd-snmpd.  We
>>
>>
>put
>
>
>>several echo's of return codes in the scripts,
>>which all return 0, and it looks like the scripts are running fine, but
>>
>>
>the
>
>
>>processes do not continue.  We did use the no-hup for ucd-snmpd, but we do
>>
>>
>not
>
>
>>want to have to nohup every thing.  Apache
>>start-up problem does seem to be somewhat intermittent.  After we are
>>
>>
>fully up,
>
>
>>these same scripts can be run manually with out a problem.  It appears to
>>
>>
>be
>
>
>>only related to system boot/re-boot.  Does
>>anyone else have these problems at start-up?
>>
>>
I encountered this problem with my scripts also.  The way to solve this
prob, is to use nohup.  The problem is caused due to the "father" script
that runs all its "children" under /etc/rc?.d.  When the father script
finishes, it sends a SIGHUP to all its children.  If a child is not
properly configured to fork itself to the background (daemon), it will
exit with the SIGHUP.
The way to work around this, is to trap this signal.  You don't have to
use the /usr/bin/nohup binary (I don't like it either).  You can add
this line at the begining of your startup script (assuming /bin/bash):
trap "" SIGHUP
Enjoy.

--
Yuval Turgeman
Research Engineer, Aduva LTD.
Office +972-3-7534373, Mobile +972-54-308076

kernel bug with samba

[Tim-Chr. Hanschen]

Hi,

I installed SLES 7.2 (kernel 2.4.5) on an IBM 2064. My samba version is
2.2.0a.

Everything seems to work fine. I can map devices to my windows workstation
and I can read files from the device.
When I tried to write to the device the whole linux stop working. I got the
 following error:

illegal operation: 0001
cpu:1
Process smbd (pid: 465, stackpage=00FB9000)
Kernel GPRS:
    001b  0029b518
8005b520    005b9060  88c2
0003  0014  00fb8000  0003
0122e7bc  8005b480  8005b520  00fb9bb8
Kernel ACRS:
      
0001      
      
      
Kernel BackChain  CallChain
   00fb9bb8   [<0005b520>]
   00fbb998   [<0005b612>]
   00fb9cf8   [<000658dc>]
   00fb9d58   [<000578ee>]
   00fb9db8   [<00057aec>]
   00fb9e28   [<0004674e>]
   00fb9ea8   [<00046c14>]
   00fb9f08   [<000130d0>]
May 22 13:30:24 osa-rzl1 kernel: kernel BUG at fcntl.c:3971

I rebooted the system and tried again the whole story. I am using
lvm-0.9.1_beta7-14 for the dasds.

TIA,
  - Tim -

Using ram disks

[Dennis G. Wicks]

Greetings;

How does one go about getting access to the /dev/ram? disks?

I have tried formatting and making file systems and just mounting
but I can't seem to find the right combination.

Many TIA,
Dennis
--
Dennis G. Wicks Systems Programmer  [EMAIL PROTECTED]
Communications Data Group   Tel: (217)355-7117  Fax: (217)351-6994
102 S. Duncan Rd.
Champaign, IL  61822

Steps to apply all RPM Updates to RH 7.2 system?

[Dave Myers]

I am tasked to upgrade a RH 7.2 system.
Are these correct steps??

Is it OK to run the RPM updates against a "live" system,
assuming I have backed up first using OFFLINDR ?

I'm not sure I really need to do step3 ???


TIA
Dave

1. Backup/Clone RH 7.2
2. Apply all RPM updates to RH 7.2 system from
ftp://ftp.redhat.com/pub/redhat/linux/updates/7.2/en/os/s390/";>ftp://ftp.redhat.com/pub/redhat/linux/updates/7.2/en/os/s390/
3. Reapply the LCS OCO Driver to the newly updated system.
4. Rewrite boot sector with zipl util
5. Backup new RH 7.2  using the OFFLINDR util.
6. Test new RH 7.2

Re: Adding a DASD volume

[Rob van der Heij]

At 12:06 22-05-02 +0700, Felix Yoyok S. Aditias wrote:

I can get to http://www.marist.edu/htbin/wlvindex?linux-390 now
without problems, so maybe the port 8000 wasn't right?

>If the steps are not correct, which documentation should I refer to?

I think you're confusing a few steps. The 'insmod dasd' only works
in the ramdisk based starter system because there is no dasd driver
built into the kernel. The real system boots from disk and thus
needs the dasd driver to be built into the kernel. This is why you
don't find the dasd.o module in your real system.

When the dasd driver is built into the kernel, the dasd= parameter
is taken from the kernel command line (written on disk with silo).
With the 2.2.16 kernel you can not dynamically add disks to your
dasd driver, so you must put the additional disks in the parmfile
(after the current disks), run silo, and reboot.

Rob

Re: Timer Patch

[Marcy Cortes]

At 10:56 20-05-02 -0400, Mcphillips, James wrote:
>Here is what I received from SuSE support. For k_timer and the OCO's in
>k_deflt

I put this on a couple of guests yesterday.  I noticed today
that the cpu time used by an idle Linux has been reduced from
about .23% of a cpu (as told by VMRTM) to .03%.   Wonderful!
(This was on a 9672-R26 z/VM V4).

Thanks SuSE!  Thanks IBM!

Marcy Cortes
Wells Fargo Services Co

Re: Yet Another Salvo in the Sun vs. IBM War of Words

[Phil Payne]

> This will probably be the last one of these that _I_ report, anyway.
>
> http://www.linuxgram.com/article.pl?sid=02/04/16/1931245

We have the making of a deal here.  It will probably be the last one that _I_ read.

Mark's not quite right, though:

'And by the way, Khan says, "IBM was forced to offer the OS because it was mandated in 
the IBM
antitrust lawsuit years ago." Since he has chosen to go back so far, I feel obliged to 
point
out that until that point we had provided an OS for free to customers and with source 
code.
And yes, we did charge other hardware vendors for it. But that was the then business 
model of
nearly some 35-years ago. Talk about sins of the father.'

Actually, IBM didn't charge other vendors for it.  Early System/360 operating systems 
were
public domain - not just free of charge but also uncopyrighted - anyone could take one 
and
turn it into a chargeable or non-chargeable product at will.  I have a copy of the 
original
letter to Dr Amdahl's lawyers confirming this: "IBM asserts no rights over its Systems 
Control
Programming".

That is the environment in which IBM rose to world dominance, and perhaps one reason 
why they
find Linux so attractive today - with its more modern analogue of the old public domain
concept.

--
  Phil Payne
  http://www.isham-research.com
  +44 7785 302 803
  +49 173 6242039

Communicating with "Console" on HMC (SERVC instruction?)

[Snyder, Bradley (LNG)]

Does anyone know how to send messages to the "console" (i.e. operating
system messages item) on the HMC?  Please DO NOT, and I repeat DO NOT
respond to this with any GPL-licensed code examples.  I'm looking for a
general description or a pointer in the right direction.  The closest I have
come is the SERVC instruction, which is undocumented, and, as far as I can
tell, not supported in GAS.

Thanks,
--Brad

Even more strangeness...

[Jason McMullan]

We are getting a very bizarre error on a binary compiled with gcc-2.95.3.
This program has a main event loop which calls select(), and sometimes it
comes back from the select() and then the following subroutine call
branches into neverland.

Here is a piece of the code, disassembled from the on-disk binary image:

00450c74 :
  450c74:   90 6f f0 18 stm %r6,%r15,24(%r15)
  450c78:   a7 d5 00 1c bras%r13,450cb0 
  450c7c:   00 44 d4 d8 .long   0x0044d4d8
  450c80:   00 44 d9 98 .long   0x0044d998
  450c84:   00 44 da a8 .long   0x0044daa8
  450c88:   00 44 db 68 .long   0x0044db68
  450c8c:   00 45 22 84 .long   0x00452284
  450c90:   00 0f 42 40 .long   0x000f4240
  450c94:   00 40 91 c0 .long   0x004091c0
  450c98:   00 40 89 60 .long   0x00408960
  450c9c:   00 44 dc c4 .long   0x0044dcc4
  450ca0:   00 44 df dc .long   0x0044dfdc
  450ca4:   00 44 e2 14 .long   0x0044e214
  450ca8:   00 44 e1 54 .long   0x0044e154
  450cac:   00 45 23 50 .long   0x00452350
  450cb0:   18 1f   lr  %r1,%r15

Basically, we enter the function, save some registers, then jump around
this object which I'm calling the "constants table".  This table consists
of:
entries 1-5:  function entry points with static linkage within the module
entry 6:  function with external linkage from another module
entry 7:  a scalar constant, 100, used within the function
entry 8:  a double-indirection to select() in glibc with external linkage
entry 9:  as above, but for usleep()
entries 10-13:  more function entry points with static linkage
entry 14: function with external linkage from another module


When the program loads, the dynamic linker rewrites these addresses for the
virtual address space under which the binary is executing.  The table
becomes:

44 8e f3 68
44 8e f8 28
44 8e f9 38
44 8e f9 f8
44 8f 41 14
00 0f 42 40
40 ?? 7d 48
40 2c 78 f0
44 8e fb 54
44 8e fe 6c
44 8f 00 a4
44 8e ff e4
44 8f 41 e0

Note that one byte is unknown, gdb masked it away when it tried to
disassemble the constants as valid machine code.  Anyway, this is the way
the code appears during execution.  I can CTRL-C it several times in the
debugger, and it always looks like this.

As the program runs, and passes through this event loop hundreds or
thousands of times, there comes a time when something strange happens.
Suddenly, gdb reports a segmentation violation, as we tried to branch to
address 0x00021b54.  This fails, our binary is mapped in at addresses
starting around 0x448ce000, and there is no valid mapping to that address.
Looking at the constants table, it has been changed:

00 02 13 68
00 02 18 28
00 02 19 38
00 02 19 f8
00 00 00 00
00 0f 42 40
00 00 00 00
00 00 00 00
00 02 1b 54
00 02 1e 6c
00 02 20 a4
00 02 1f e4
00 00 00 00

Looking at the backtrace, it fails the first time we try to branch from the
constants table following a select().  It appears that, very rarely, within
select(), something with a detailed knowledge of the binary format decides
to rewrite the jump addresses, but not the scalar constant 0x000f4240.
Pointers to functions with static linkage within the same module are set to
different values (but with the same spacings as in the original code).
Pointers to functions with external linkage, whether in the binary or in
external libraries, are zeroed out.  The 0x00021b54 now occupies the
position in the constants table which used to be held by the address of the
next function to be called after select().  The memory on either side of
the constants table is not touched.

This code is single-threaded, and not particularly magical.  It is running
as a fork()-ed child of another process, though.

Note that all of the kernels we have tested have a bug in mprotect(), it
returns success while failing to set any protection on the page(s).  At
run-time, these pages are being mprotect()-ed:
mprotect(0x448ce000, 172032, PROT_READ|PROT_WRITE) = 0
mprotect(0x448ce000, 172032, PROT_READ|PROT_EXEC) = 0

The first mprotect() is to allow the jump tables to be rewritten by the
run-time linker.  The second is to protect the text segment so that nobody
overwrites the pointer tables.  Well, the pages are not protected, and
somebody is overwriting the page under some very rare set of circumstances
under select(), but I can't see who is doing it.  Using hardware
watchpoints under gdb does not reveal the offending instruction, the
watchpoint was not caught.  This is a bit of a mystery, as hardware
watchpoints do work under gdb on 390, but the attempt to watch a particular
address within the jump table did not result in any trap when the value
changed.


So, has anybody see

Re: Linux backup software

[PAUL WILLIAMSON]

I'm not sure if this will work, but Arkeia is pretty good, and is 
very good at managing tape libraries and such.  It is pretty 
involved, but once you get the hang of it, you'll really like it (IMHO).

http://www.arkeia.com/

Paul

>>> [EMAIL PROTECTED] 05/22/02 10:33AM >>>
I apologize if this has been asked before.
What Linux software is available to perform logical backups of
Linux S/390 files and directories?
Ideally we would be looking for 'freeware or shareware' but if you know
of any commercial ware that would be appreciated also.
We would be looking for software that can backup on to 3490s and 3480s.

Thanks in advance...Neil

Re: Using ram disks

[Dennis G. Wicks]

Thanks Mark!

I did a man -k on everything I could think of that might
relate to ram disk. I usually also do a find but I
thought that all other things "ram" were in /dev!

I'll have to browse around in /usr/src/linux/Documentation more.
It looks like there is a lot of good stuff in there.

Dennis




"Post, Mark K"
  cc:
Sent by: LinuxSubject: Re: Using ram disks
on 390 Port
<[EMAIL PROTECTED]
ARIST.EDU>


05/21/02 04:15
PM
Please respond
to Linux on 390
Port






Dennis,

Take a look at "Using the RAM disk block device with Linux" in
/usr/src/linux/Documentation/ramdisk.txt.

Mark Post

-Original Message-
From: Dennis G. Wicks [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, May 21, 2002 4:40 PM
To: [EMAIL PROTECTED]
Subject: Using ram disks


Greetings;

How does one go about getting access to the /dev/ram? disks?

I have tried formatting and making file systems and just mounting
but I can't seem to find the right combination.

Many TIA,
Dennis
--
Dennis G. Wicks Systems Programmer  [EMAIL PROTECTED]
Communications Data Group   Tel: (217)355-7117  Fax: (217)351-6994
102 S. Duncan Rd.
Champaign, IL  61822

Runlevel3 start up problems

[Gustavson, John (ECSS)]

We are running SUSE 2.4.7 kernel.  We have had problems with the start-up scripts in 
/etc/init.d(rc.d)/rc3.d for Apache, TOMCAT, and ucd-snmpd.  We put several echo's of 
return codes in the scripts,
which all return 0, and it looks like the scripts are running fine, but the processes 
do not continue.  We did use the no-hup for ucd-snmpd, but we do not want to have to 
nohup every thing.  Apache
start-up problem does seem to be somewhat intermittent.  After we are fully up, these 
same scripts can be run manually with out a problem.  It appears to be only related to 
system boot/re-boot.  Does
anyone else have these problems at start-up?

Regards

John Gustavson
Enterprise Central Software Services (ECSS)
570 Washington Street - 2nd floor
New York, New York, 10080-6802

Telephone: 1-212-647-3793
Fax: 1-212-647-3321
Email: [EMAIL PROTECTED]