Re: Redbook draft: The Virtualization Cookbook for IBM Z Volume 5: KVM

[Ray Mansell]

How can it possibly be 15+ years!! In any case, congratulations - it has
proved to be a great resource.

Ray

On 4/8/2020 06:35, Michael MacIsaac wrote:

Hello list,

Wow, The Virtualization Cookbook still alive and kicking. Nice work to all
involved!  I remember working on the first one more than 15 years ago.

- Sir Mike the Chef of Books

On Tue, Apr 7, 2020 at 9:42 AM Christian Borntraeger 
wrote:


There is now a redbook draft for a KVM cookbook:

http://www.redbooks.ibm.com/Redbooks.nsf/RedpieceAbstracts/sg248463.html?Open

As this is still in draft state review feedback is still possible:
redbo...@us.ibm.com

Christian

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or
visit
http://www2.marist.edu/htbin/wlvindex?LINUX-390



--
  -Mike MacIsaac

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www2.marist.edu/htbin/wlvindex?LINUX-390


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www2.marist.edu/htbin/wlvindex?LINUX-390

Re: Single user (rescue target) in SLES 12

[Ray Mansell]

On the subject of #CP VI VMSG... my brain/finger combination often had
problems trying to enter the command I wanted before the timeout
occurred. Eventually it dawned on me that I could enter '#CP' on its own
(or use PA1) to enter CP READ, after which I could take my time entering
the VI VMSG command correctly before restarting the virtual machine.

I'm sure this is old news to many of you, but I thought it worth
mentioning.

Ray

On 2017-06-26 15:26, Marcy Cortes wrote:

> One thing I haven't figured out in SLES 12 systemd system is how to boot into 
> single user mode (rescue.target)?
> On sles 11, I just did a #cp vi vmsg 1 1
>
> Anyone know?
>
> Marcy
>
> This message may contain confidential and/or privileged information. If you 
> are not the addressee or authorized to receive this for the addressee, you 
> must not use, copy, disclose, or take any action based on this message or any 
> information herein. If you have received this message in error, please advise 
> the sender immediately by reply e-mail and delete this message. Thank you for 
> your cooperation.
>
> --
> For LINUX-390 subscribe / signoff / archive access instructions,
> send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
> http://www.marist.edu/htbin/wlvindex?LINUX-390
> --
> For more information on Linux on System z, visit
> http://wiki.linuxvm.org/

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Debian CPU limitation?

[Ray Mansell]

Thank you...

Ray

On 3/25/2017 13:52, Philipp Kern wrote:

On 03/22/2017 01:06 PM, Christian Borntraeger wrote:

On 03/22/2017 12:51 PM, Ray Mansell wrote:

Thank you all for the information. It would appear we are stuck with 32
CPUs for the moment and, since I'm retiring at the end of next week, I
guess my replacement will have to decide how to proceed :-)

As a last duty, can you maybe open a bug against Debian to bump the
max number of CPUs?

This is now Debian bug #858731.

Kind regards
Philipp Kern

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Debian CPU limitation?

[Ray Mansell]

Thank you all for the information. It would appear we are stuck with 32
CPUs for the moment and, since I'm retiring at the end of next week, I
guess my replacement will have to decide how to proceed :-)

Thanks again,
Ray

On 3/21/2017 18:46, Mark Post wrote:

On 3/21/2017 at 06:32 PM, Ray Mansell  wrote:

cat /proc/cmdline
root=/dev/disk/by-path/ccw-0.0.9207-part3 BOOT_IMAGE=0

Can this be modified to allow >32 CPUs?

If Christian is right (and he almost certainly is), then the only thing that 
will allow you to have more than 32 CPUs is to recompile the kernel with 
CONFIG_NR_CPUS set to a higher value.  I just looked at one of my SLES12 test 
systems, and we have CONFIG_NR_CPUS=256 specified, so I'm guessing having a lot 
of headroom won't cause any problems.


Mark Post

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Debian CPU limitation?

[Ray Mansell]

On 3/21/2017 17:38, Mark Post wrote:

What does /proc/cmdline show for the kernel parms?


Mark Post


cat /proc/cmdline
root=/dev/disk/by-path/ccw-0.0.9207-part3 BOOT_IMAGE=0

Can this be modified to allow >32 CPUs?

Many thanks for your help,
Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Debian CPU limitation?

[Ray Mansell]

We've been running some beefy Debian servers, but of course one of them
was not beefy enough, and we were asked if we could increase the number
of CPUs from 32 to 48. Easy, I thought, until I rebooted the beefier
server only to discover it had restricted itself to 32 CPUs, even though
it knew 48 were present:

   cpu: 48 configured CPUs, 0 standby CPUs
   Brought up 32 CPUs

The output from uname is:

   uname -a
   Linux linux02 3.16.0-4-s390x #1 SMP Debian 3.16.7-ckt9-3
   (2015-04-23) s390x GNU/Linux

And it is running on a z/VM 6.3 host.

I've searched through many documents and publications, but haven't been
able to find anything that might help resolve this situation. Can anyone
provide some insight, please?

Thank you,
Ray


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: XMAS gift Ubuntu ooRexx 4.2 for s390x

[Ray Mansell]

Excellent!

root@ray:/tmp# rexx rexxcps.rex
- REXXCPS 2.1 -- Measuring REXX clauses/second -
 REXX version is: REXX-ooRexx_4.2.0(MT)_64-bit 6.04 16 Dec 2016
   System is: LINUX
   Averaging: 100 measures of 100 iterations

 Performance: 8017343 REXX clauses per second

That's on an EC12.

Thank you, Santa Dave...
Ray

On 12/24/2016 10:51, Dave Jones wrote:

I have created a Ubuntu ooRexx 4.2 .deb install file for the s390x
architecture, 64 bit. It has been tested on Ubuntu 16.04.1 LTS running as a 
guest of z/VM 6.3.
If anyone would like it, it can be found here:

http://www.vsoft-software.com/oorexx_4.2.0-1_s390x.deb

If anyone want the 32 bit version of ooRexx, drop me a note off list.
Happy Holidays.

DJ

--
ITC-email-sig

*David Jones **|**Managing Director for zSystems Services **|**z/VM,
Linux, and Cloud
703.237.7370 (Office) **|**281.578.7544 (Cell)*

*Information Technology Company *

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: minor update to CMS Make including 'make', 'curl', and 'wget' for CMS

[Ray Mansell]

I see three versions internally... none of them written by Stuart
Feldman :-)

Ray

On 12/12/2016 09:58, Rick Troth wrote:


After creating this spin, I found there is an IBM version too. Looks
quite complete.


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Debian install using LVM

[Ray Mansell]

Martha...

I've been fighting the same thing all afternoon. Thought I was going
bonkers, since it all worked so well last week. For the moment I've
installed without LVM, but there's certainly a problem.

Ray

On 4/11/2016 19:00, Martha McConaghy wrote:

I've been fighting with a Debian install all day and wonder if anyone else
has run into the same problem. I wanted to use LVM for the root filesystem,
so created a /boot partition and an LVM for the rest.  Everything works fine
until it comes to the step where it tries to write out ZIPL.  That step fails.
I tried all sorts of combinations, but the only that actually works is
creating 1 partition for / with no LVM at all.  Something is not right.

The timestamp on the install files I'm using is April 2, so they are pretty
new.  The level of Debian is jessie.

Has anyone run into this?

Martha

Martha McConaghy
System Architect/Technical Lead
Marist College

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Ubuntu 16.04 LTS Beta for IBM LinuxONE and z Systems

[Ray Mansell]

After some offline troubleshooting (Viktor and I are both with IBM), it
turns out that the installation files on the CD image are slightly out
of date with respect to the development files. Having obtained the
relevant initrd and kernel images, I am now happily installing Ubuntu.

This is where you should retrieve the working files:
http://ports.ubuntu.com/dists/xenial/main/installer-s390x/current/images/generic/

There are also a couple of things you need to be aware of during the
installation:

 * After you select the installation language, the installer will ask
   you to select a download mirror. The only choice is to enter it
   manually, at which point you should enter: ports.ubuntu.com
 * After that, it will prompt you for the subdirectory, to which you
   should respond '/ubuntu-ports' (without the quotes, but *with* the
   leading slash)

From that point onwards, it looks a lot like the Debian installation
procedure. Well, so far... it hasn't yet completed.

Ray

On 3/31/2016 10:17, Viktor Mihajlovski wrote:

On 31.03.2016 15:57, Ray Mansell wrote:

I downloaded and expanded the CD image, and started to install it.
However, I'm now stuck looking for an FTP server from which to perform
the bulk of the installation.

On the "Select your location" screen, I choose USA.
The next screen "Choose a mirror..." supplies "enter information
manually" as the only choice.
Selecting that gets me to the "Choose a mirror..." screen, which invites
me to enter the mirror's hostname.
And that would be what? (pointing it to my local FTP server where I
downloaded the ISO doesn't help - it just fails with "Bad archive mirror")

Can you point me to the appropriate installation mirror, please?

You could try the following URL: http://ports.ubuntu.com/ubuntu-ports
[,,,]




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Ubuntu 16.04 LTS Beta for IBM LinuxONE and z Systems

[Ray Mansell]

I downloaded and expanded the CD image, and started to install it.
However, I'm now stuck looking for an FTP server from which to perform
the bulk of the installation.

On the "Select your location" screen, I choose USA.
The next screen "Choose a mirror..." supplies "enter information
manually" as the only choice.
Selecting that gets me to the "Choose a mirror..." screen, which invites
me to enter the mirror's hostname.
And that would be what? (pointing it to my local FTP server where I
downloaded the ISO doesn't help - it just fails with "Bad archive mirror")

Can you point me to the appropriate installation mirror, please?

Thank you,
Ray

On 3/31/2016 07:12, Dimitri John Ledkov wrote:

Hello,

Last week Ubuntu 16.04 LTS Beta was shipped, with final expected on
21st of April.

This release includes a port of Ubuntu to s390x architecture (zEC12 and up).

I hope you find this post relevant, and if you have any questions do
not hesitate to ask, I'll try to answer everything as best as I can.

= Downloads =
http://www.ubuntu.com/download/server/linuxone

= IRC =
#ubuntu-s390x on freenode

irc://irc.freenode.net/ubuntu-s390x

Say hello to xnox (me) upon joining =)

= Mailing List =
http://launchpad.net/~ubuntu-s390x

= Wiki =
https://wiki.ubuntu.com/S390X

--
Regards,

Dimitri.

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Migrate zLinux off zVM into standalone LPARs?

[Ray Mansell]

Thanks, Marcy - Perhaps I can offload all this effort onto our z/OS
admin. I rather like the sound of that...

Ray

On 11/17/2015 15:55, Marcy Cortes wrote:

Changing devices numbers is generally not a problem unless you have lots of 
dedicated devices.
SSI needs a bit of work as well if you have that.

DDR is very slow.
If you have z/OS, any of its utilities are way better.   DFDSS, zDMF (TDMF), or 
FDR all beat the pants off of DDR.
Just have to take down your VM & Linux first though.





--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Migrate zLinux off zVM into standalone LPARs?

[Ray Mansell]

Thank you Mark, and Alan, and Gregory...

I think our problem is going to be more one of boredom rather than
encountering insurmountable problems, as DDR plods its way through
several hundred volumes. Originally I thought we were going to have to
cope with changing device numbers on the new system, which I believe
would have posed a much more interesting problem, but I think I've
managed to persuade the storage admin that this simply isn't practical.
In any case, it looks as though I will have plenty to occupy my time
between now and the new year.

Thanks again,
Ray

On 11/17/2015 11:07, Mark Post wrote:

On 11/17/2015 at 10:12 AM, Ray Mansell  wrote:

Speaking of disk address changes... how does one cope with that? We have
a new DS8700 to replace our ancient DS8000s, and I'm wondering what is
the best way to migrate our LPAR Linux servers to the new box? I'm sure
this must have been done before, so I'd be very interested to learn how
others have managed this.

How you cope is going to be determined by how the file systems are referenced 
in /etc/fstab, and whether LVM is being used or not.  What does your setup look 
like?


Mark Post



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Migrate zLinux off zVM into standalone LPARs?

[Ray Mansell]

Speaking of disk address changes... how does one cope with that? We have
a new DS8700 to replace our ancient DS8000s, and I'm wondering what is
the best way to migrate our LPAR Linux servers to the new box? I'm sure
this must have been done before, so I'd be very interested to learn how
others have managed this.

Ray...

On 7/1/2015 11:51, Stewart, Lee wrote:
> Since you won't have VM to limit which and how many devices each of the 
> Linuxes sees, you may want to consider isolating them from each other via the 
> IOCP.  Also keep in mind that all your addresses will probably change as you 
> move from virtual addresses to real addresses.   And what the Linuxes have 
> created for volsers is probably not unique, and if your DASD is shared with 
> an MVS image, those duplicates will show up there...
>
> Lee Stewart ● VM System Support ● Visa ● Phone:  6(750)4601 - +1-303-389-4601 
> ● lstew...@visa.com
>
>

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: zKVM installation problem

[Ray Mansell]

On 10/6/2015 06:54, Philipp Kern wrote:

On 2015-10-05 20:26, Grzegorz Powiedziuk wrote:

I’ve seen weird python errors similar to these ones (unfortunately I
didn’t keep traces to compare) in some linux distribution during the
installation if I had a “dirty disk”.


This isn't a weird Python error. It's a straightforward coding error. 
Without having the code at hand I guess that s is only set in an "if" 
but who knows. It should at least have been initialized with 
something. At least you can just inspect the code given that it's Python.


Kind regards
Philipp Kern


Yes, I had looked at the source and indeed 's' is only set inside an 
'if', hence that particular error. However, as you will have seen in 
other responses, my actual problem appears to be with the Windows CDFS 
restricting (well, truncating) file names to 64 characters. I need to do 
some reading to find out if there's a way around that. In the meantime, 
a suitable set of 'rename' commands seems to have done the trick.


Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: zKVM installation problem

[Ray Mansell]

On 10/6/2015 06:27, Michael Mueller wrote:

On Mon, 5 Oct 2015 11:11:40 -0400
Ray Mansell  wrote:


I'm trying to install zKVM in a virtual machine, but no matter what
installation options I choose, I always get the following error:

2015-10-03 14:39:40,900 - controller.controller - INFO - InstallProgress
screen
2015-10-03 14:39:40,901 - controller.controller - INFO - Formatting disks...
2015-10-03 14:39:40,901 - controller.controller - INFO - Installing KVM
for IBM z into disk dasda...
2015-10-03 14:39:41,041 - model.installfunctions - INFO - Get repodata_file
2015-10-03 14:39:41,088 - model.installfunctions - CRITICAL - Failed
installSystem
2015-10-03 14:39:41,089 - model.installfunctions - CRITICAL -
EXCEPTION:
2015-10-03 14:39:41,089 - model.installfunctions - CRITICAL - local
variable 's' referenced before assignment
2015-10-03 14:39:41,089 - model.installfunctions - CRITICAL -
Stacktrace:Traceback (most recent call last):
File "/opt/ibm/kvmibm-installer/model/installfunctions.py", line 357,
in installSystem
  installPackages(rootDir, callback)
File "/opt/ibm/kvmibm-installer/model/installfunctions.py", line 749,
in installPackages
  repodata_file = getRepodataFile(repo, logger)
File "/opt/ibm/kvmibm-installer/model/installfunctions.py", line 548,
in getRepodataFile
  d = re.split('([\d\w]+-primary.sqlite.bz2)', s)
UnboundLocalError: local variable 's' referenced before assignment

2015-10-03 14:39:41,095 - controller.controller - CRITICAL - ZKVMError:
[['KVMIBMIN70500', 'Error while installing packages.'], ('INSTALLER',
'INSTALLSYSTEM', 'INSTALL_MSG')]

Help? Please?

Ray,

make sure you have copied the repodata folder from the installation image to
your ftp server. Should have the following files:

$ ls -Rl ./repodata
./repodata:
total 2284
-rw--- 1 mimu mimu 410546 Aug  6 12:11
1071bedb0b7dba6b8ed2fae387116e3f2b8160544d2abdad5411fbe8f737ff63-filelists.xml.gz
-rw--- 1 mimu mimu  14151 Aug  6 12:11
3a456231834ea9d32d6a7525ff38fe6246932c2a14d8df1b6cf75b810dc09545-KVMIBM-1.1.0-20150806-comps.xml
-rw--- 1 mimu mimu 517316 Aug  6 12:11
655cf5993b0a15383cfe5224bf2db4d95b3cba0539c3075e5fc57c3cd08ab40e-filelists.sqlite.bz2
-rw--- 1 mimu mimu 771729 Aug  6 12:11
836db89a0f5c266ea3b34d070a26cad34ad39db91a416081bcda615bb8605324-primary.sqlite.bz2
-rw--- 1 mimu mimu  76579 Aug  6 12:11
9fe8a5082cbb90e961d5e597ad6993c6a1f35035bf918b57b3fe1bba822550b8-other.xml.gz
-rw--- 1 mimu mimu  96919 Aug  6 12:11
b08023e34c172ac98447cf816543a2a02d0f9819d29cc40d375e02b37c8d6ce5-other.sqlite.bz2
-rw--- 1 mimu mimu   1706 Aug  6 12:11
d5e91197cddcab057566c403de6c1a71e3e56cec91bc95006c5dde1a69ffcba7-KVMIBM-1.1.0-20150806-comps.xml.gz
-rw--- 1 mimu mimu 398232 Aug  6 12:11
f410070319c7a1fbb90a37e20f85064390db0112a0ceb34f8c84ef4f866e5e4e-primary.xml.gz
-rw--- 1 mimu mimu   3740 Aug  6 12:11 repomd.xml
-r 1 mimu mimu   2599 Aug  6 12:16 TRANS.TBL

By moving them away I can reproduce the error message.

Michael



Aha! Thank you... I have the ISO image on a Windows machine, but the
tool I used to rip its contents truncated all the filenames to 64
characters, so obviously they could not be found. I must look for a new
ISO ripper.

Installation is now proceeding much more smoothly.

Thanks again,
Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: zKVM installation problem

[Ray Mansell]

On 10/5/2015 13:28, Mark Post wrote:

On 10/5/2015 at 11:11 AM, Ray Mansell  wrote:

Help? Please?

That looks like a problem IBM will need to fix.  One question, though; does 
your virtual machine provide all the system resources they ask for in the 
documentation?


Mark Post


I'm pretty sure it does, but I will certainly check again.

Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

zKVM installation problem

[Ray Mansell]

I'm trying to install zKVM in a virtual machine, but no matter what
installation options I choose, I always get the following error:

2015-10-03 14:39:40,900 - controller.controller - INFO - InstallProgress
screen
2015-10-03 14:39:40,901 - controller.controller - INFO - Formatting disks...
2015-10-03 14:39:40,901 - controller.controller - INFO - Installing KVM
for IBM z into disk dasda...
2015-10-03 14:39:41,041 - model.installfunctions - INFO - Get repodata_file
2015-10-03 14:39:41,088 - model.installfunctions - CRITICAL - Failed
installSystem
2015-10-03 14:39:41,089 - model.installfunctions - CRITICAL -
EXCEPTION:
2015-10-03 14:39:41,089 - model.installfunctions - CRITICAL - local
variable 's' referenced before assignment
2015-10-03 14:39:41,089 - model.installfunctions - CRITICAL -
Stacktrace:Traceback (most recent call last):
  File "/opt/ibm/kvmibm-installer/model/installfunctions.py", line 357,
in installSystem
installPackages(rootDir, callback)
  File "/opt/ibm/kvmibm-installer/model/installfunctions.py", line 749,
in installPackages
repodata_file = getRepodataFile(repo, logger)
  File "/opt/ibm/kvmibm-installer/model/installfunctions.py", line 548,
in getRepodataFile
d = re.split('([\d\w]+-primary.sqlite.bz2)', s)
UnboundLocalError: local variable 's' referenced before assignment

2015-10-03 14:39:41,095 - controller.controller - CRITICAL - ZKVMError:
[['KVMIBMIN70500', 'Error while installing packages.'], ('INSTALLER',
'INSTALLSYSTEM', 'INSTALL_MSG')]

Help? Please?

Ray...

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

[no subject]

[Ray Mansell]

This may be more successful:

sed -i 's:dev/sda1/scratch   ext3  defaults 1 2:dev/sda1/scratch   ext3 
 defaults 1 0:' /etc/fstab

Or... you could use something as simple as the 'cat' command:

cat /etc/fstab
cat > /etc/fstab.fixed

At this point you have the original fstab displayed on the screen, and
the cat command is waiting for input. Use copy and paste to copy the
original lines one at a time to the command area, and press enter,
taking care to modify the one you want to change. Terminate with ^D, and
you should have a fixed version of your file which you can activate by
suitable renaming.

Ray

On 6/11/2015 04:03, Berthold Gunreben wrote:

On Thu, 11 Jun 2015 07:21:50 +0530
Mainframe Mainframe  wrote:


Hello ,
  Thanks for reply .As I mentioned, fsck is running on
rebooting time and its not getting completed and having issue
mentioned earlier email chain.

So, to isolate this issue, I want to revert back

dev/sda1/scratch   ext3  defaults 1 2

to

/dev/sda1/scratch   ext3  defaults 1 0

under /etc/fstab

 From z/VM console and I am not good at using SID  editor commands. Can
anybody provide syntax for changing this line using SID editor
commands.

You are probably referring to sed, not to sid. I would use commands
like this:

cp /etc/fstab /etc/fstab.orig
sed -i 's:dev/sda1:/dev/sda1:' /etc/fstab

Berthold



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Expanding CMS Mini Disk - DDR or Copyfile? - still not expanding

[Ray Mansell]

I looked, and I still have my update to DMSFOR. All it needs is rework
to accommodate the intervening quarter-century's worth of changes to the
source :-)

On 2/3/2015 23:19, Alan Altmark wrote:

Hence my comment about revisiting DMSFOR and DMSAUD.   So instead of
trying to modify the disk such that is is large when subsequently
accessed, I went the route of tricking the RECOMP code.  But I still
believe that the allocation map was never intended to physically move, but
switches locations along with the DIRECTOR file each time the DOP is
"flopped."

But it's going to take someone with more patience than I have to perform
more experiments.   I would first write several files to the disk, THEN
expand it, THEN fill the disk.  My intuition is telling me that the newly
RECOMPed allocation map will gladly overlay those first files and create a
mess.

But as I say, some more experiments are needed.  Or Ray's code.  :-)

Alan Altmark

Senior Managing z/VM and Linux Consultant
Lab Services System z Delivery Practice
IBM Systems & Technology Group
ibm.com/systems/services/labservices
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Expanding CMS Mini Disk - DDR or Copyfile? - still not expanding

[Ray Mansell]

Back in those same mid-80s I added code to the FORMAT command to provide
an 'EXTEND' option. You could create a new, larger minidisk, the first
cylinders of which were a replica of the old one. Then you could simply
FORMAT the new minidisk, specifying the EXTEND option, and it would
magically grow in size to occupy all of the new space, with all of the
original files intact.

Now, I wonder where I stashed that code...

Ray

On 2/2/2015 19:19, Alan Altmark wrote:

On Monday, 02/02/2015 at 05:55 EST, Rick Troth
 wrote:

Evidently, FORMAT makes a note of the number of cylinders it block
formatted (the "low level" phase) first time around. Guessing that is
the purpose of ADTMCYL. I should have known. Sorry, Joe.

So you can (RECOMP smaller and you can (RECOMP larger, but no larger
than some pre-detected size.

You can't RECOMP anything larger than the originally formatted size.   You
can RECOMP lower and back again, but no larger.  And, yes, ADTMCYL holds
the number of formatted cylinders..

The protection is in place because of the allocation map.  Unlike other
files on the disk, it is in a fixed location since it is written during
FORMAT and its size never changes, as that size is based on the size of
the disk.  But if you add more cylinders, you add more blocks.  And as you
do that, the allocation map has to grow.  But it can't grow.  It's
surrounded by other data.

I have to say, it's kind of fun to lean back in my rocking chair and see
my programmer id in FORMAT (DMSFOR) from back in the mid-80s.  :-)

Alan Altmark

Senior Managing z/VM and Linux Consultant
Lab Services System z Delivery Practice
IBM Systems & Technology Group
ibm.com/systems/services/labservices
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Installing RHEL 7

[Ray Mansell]

Thanks, Dan - that did the trick.

Ray

On 11/6/2014 10:06, Dan Horák wrote:

On Thu, 6 Nov 2014 09:46:47 -0500
Ray Mansell  wrote:


I'm having pretty much the same problem described by Mike, except
that my generic.prm looks like this:

ro ramdisk_size=4 cio_ignore=all,!condev
inst.repo=ftp://9.2.98.16/e/redhat/rhel7

Looking at the FTP server, there is no sign that any attempt was made
to access it.

Any ideas? I must be doing something wrong, but it's eluding me so
far.

There is no interactive s390 specific part in the installation where
you could enter network and disk info as opposed to RHEL <= 6. They
all need to be specified in advance in the generic.prm or in the old
style CMSCONF file.

Please read
https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Installation_Guide/chap-installer-booting-ipl-s390.html#sect-customizing-generic-prm-s390


Dan

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Installing RHEL 7

[Ray Mansell]

I'm having pretty much the same problem described by Mike, except that 
my generic.prm looks like this:


ro ramdisk_size=4 cio_ignore=all,!condev
inst.repo=ftp://9.2.98.16/e/redhat/rhel7

Looking at the FTP server, there is no sign that any attempt was made to 
access it.


Any ideas? I must be doing something wrong, but it's eluding me so far.

Ray

On 9/16/2014 15:41, Michael MacIsaac wrote:

Hello list,

1) Has anyone been able to install RHEL 7 on z?
2) Is there any documentation specific to System z?

I was able to download the two ISO images from rhn.redhat.com.  I did not
see any documentation on the Web nor on the mounted ISO (perhaps I'm
missing it).

I tried the following steps, but got errors similar to what others have
reported in this forum:

Thanks.

 -Mike M.

-) Mount ISO image on an NFS server and export it
-) Copy kernel and RAMdisk from images/ directory to CMS disk
-) Update parmfile in the images/ directory with new first line:
# cat generic.prm
ro ramdisk_size=4 cio_ignore=all,!condev
-) Log on to Linux virtual machine
-) DEF STOR 1G
-) IPL CMS
-) run the RHEL7 EXEC - system starts booting:
...
   7.280120 systemd-udevd: renamed network interface eth0 to enccw0.0.0600
...

*really, systemd, I will miss the old sysVinit ... (sniff, sniff :))really,
enccw0.0.0600 is better than eth0?  (not a good start systemd :))*
...
   OKMounted Configuration File System.
   OKReached target System Initialization.
   OKStarted Show Plymouth Boot Screen.
   OKReached target Paths.
   OKReached target Basic System.
...
*install stops here for a couple of minutes, then:*
...
dracut-initqueueÝ706¨: Warning: Could not boot.
   OKStarted Show Plymouth Boot Screen.
   OKReached target Paths.
   OKReached target Basic System.
dracut-initqueue: Warning: Could not boot.
dracut-initqueue: Warning: /dev/root does not exist
...
generating /run/initramfs/rdsosreport.txt

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Need an Easy way to Identify whether on 1st or 2nd Level System

[Ray Mansell]

Along the same lines, I set the status area to reverse video for my
second-level systems.

Ray

On 4/10/2014 15:01, Ronald van der Laan wrote:

Robert,

You could use the STATUS command in the SYSTEM CONFIG to change the usual
CP RUNNING,  CP READ, etc into something else.
We once had messages like CP CRAWLING and so on a very slow test lpar

Ronald van der Laan




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: AUTO: Umberto Silvestri is prepared for DELETION (FREEZE)

[Ray Mansell]

I looked him up in the IBM internal phonebook - he isn't there. So I
sent him email (inside IBM), and in reply I received:

Umberto Silvestri is prepared for DELETION (FREEZE)

When I looked more carefully, the reply had actually been sent by some
sort of agent server in IBM Germany, so it would seem that is the source
of the problem. In any case, deleting him from the list seems to have
been the optimal solution.

Ray...

On 3/14/2014 22:01, A. Harry Williams wrote:

On Fri, 14 Mar 2014 16:48:46 -0300 Mauro Souza said:

Maybe Umberto is frozen in a cryogenic storage tank somewhere and they are
freezing all his assets and deleting everything related to him on Internet,
so when they reanimate him, in hundreds of years, he can have a fresh
start.Or he has been fired by IBM, his account has been frozen and will be
deleted. No way to know... :)

Mauro
http://mauro.limeiratem.com - registered Linux User: 294521
Scripture is both history, and a love letter from God.


I've deleted Umberto Silvestri from the list.

/ahw



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Fedora 19 for IBM System z 64bit official release

[Ray Mansell]

No, I had not seen that bugzilla entry, but after reading it I defined a 
second CPU on my virtual machine, and the installation ran to 
completion. So thank you for that. However, my user now tells me that 
he's getting seg faults in yum, and that rpm simply doesn't work.


I'm leaving for holiday tomorrow, so I guess I'll start chasing this 
again when I return. I must have done something very wrong if basic 
things like yum and rpm generate seg faults!


Thanks again for your help,
Ray

On 8/1/2013 03:30, Dan Horák wrote:

On Wed, 31 Jul 2013 12:44:21 -0400
Ray Mansell  wrote:


I eventually got all the way to the initial VNC installer screen, which
promptly failed with a python error: "python: cannot join thread before
it is started ". So I decided to try Fedora 18 instead, but that insists

I think you saw https://bugzilla.redhat.com/show_bug.cgi?id=970858


on coming up in layer2 mode, and thus fails, regardless of what I tell
it (i.e. LAYER2=0). I even tried LAYER2=NO and, just to be perverse,
LAYER2=1, but with no perceptible difference.

please try switching from the old network config to the dracut based
one, eg.
rd.znet=qeth,0.0.0800,0.0.0801,0.0.0802,layer2=0,portname=FOOBAR,portno=0
ip=192.168.100.100::192.168.100.1:24:fedora.example.com:eth0:none

as shown in the older release notes, "man dracut.cmdline" will describe
all the available options (same content on x86)


Dan



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Fedora 19 for IBM System z 64bit official release

[Ray Mansell]

Thanks, Richard, but it's not clear to me how I'd even start
investigating thread-related things in Fedora. I know I'm naive, but I
kind of expected this just to work, since it is, after all, a rather
basic function (installing the system).

Ray

On 7/31/2013 13:45, Richard J Moore wrote:

Ray, if you still get problems you can investigate problems using the
debugging options for dracut (rdshell and rdinitdebug). See
https://fedoraproject.org/wiki/How_to_debug_Dracut_problems

Richard





--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Fedora 19 for IBM System z 64bit official release

[Ray Mansell]

I eventually got all the way to the initial VNC installer screen, which
promptly failed with a python error: "python: cannot join thread before
it is started ". So I decided to try Fedora 18 instead, but that insists
on coming up in layer2 mode, and thus fails, regardless of what I tell
it (i.e. LAYER2=0). I even tried LAYER2=NO and, just to be perverse,
LAYER2=1, but with no perceptible difference.

Any suggestions? Or should I just give up and leave early for my vacation?

Ray

On 7/31/2013 11:03, Neale Ferguson wrote:

s3...@lists.fedoraproject.org


On 7/31/13 10:54 AM, "Ray Mansell"  wrote:


Thanks, Neale. That's a big improvement, so now I can play a bit more.
Can you point me to the s390x fedora forum you mentioned?




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Fedora 19 for IBM System z 64bit official release

[Ray Mansell]

Thanks, Neale. That's a big improvement, so now I can play a bit more.
Can you point me to the s390x fedora forum you mentioned?

Ray

On 7/31/2013 10:29, Neale Ferguson wrote:

In the s390x fedora forum Dan advised another:

"you should replace the root= value with the URL of the installer image,
see
http://fedoraproject.org/wiki/Architectures/s390x/18 for an example"


On 7/31/13 10:22 AM, "Ray Mansell"  wrote:


root=/dev/ram0

On 7/31/2013 10:11, Neale Ferguson wrote:

What does your root= parameter line look like?

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Fedora 19 for IBM System z 64bit official release

[Ray Mansell]

root=/dev/ram0

On 7/31/2013 10:11, Neale Ferguson wrote:

What does your root= parameter line look like?


On 7/31/13 10:00 AM, "Ray Mansell"  wrote:


I've spent several frustrating hours trying in vain to install Fedora,
both in text mode and using VNC, but it never actually gets as far as
inviting me to connect to complete the installation. The network is
active (I can ping the server), but that's all. I'm sure I must be doing
something wrong, but with no apparent documentation to guide me, it's
hard to work out what it might be.

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Fedora 19 for IBM System z 64bit official release

[Ray Mansell]

I've spent several frustrating hours trying in vain to install Fedora, 
both in text mode and using VNC, but it never actually gets as far as 
inviting me to connect to complete the installation. The network is 
active (I can ping the server), but that's all. I'm sure I must be doing 
something wrong, but with no apparent documentation to guide me, it's 
hard to work out what it might be.


These are the last few messages I see on my host 3270 session:

   [ [32m  OK   [0m] Started udev Coldplug all Devices.
 Starting dracut initqueue hook...
 Starting Show Plymouth Boot Screen...
   [ [32m  OK   [0m] Reached target System Initialization.
   [5.470279] dasdconf.sh Warning: 0.0.4105 is already online, not
   configuring
   [ [32m  OK   [0m] Started Show Plymouth Boot Screen.
   [ [32m  OK   [0m] Reached target Paths.
   [ [32m  OK   [0m] Reached target Basic System.
   dracut-initqueue[484]: Warning: Could not boot.
   [ [32m  OK   [0m] Started Show Plymouth Boot Screen.
   [ [32m  OK   [0m] Reached target Paths.
   [ [32m  OK   [0m] Reached target Basic System.
   dracut-initqueue[484]: Warning: Could not boot.
   dracut-initqueue[484]: Warning: /dev/ram0 does not exist
 Starting Dracut Emergency Shell...
   Warning: /dev/ram0 does not exist

   Generating "/run/initramfs/sosreport.txt"


   Entering emergency mode. Exit the shell to continue.
   Type "journalctl" to view system logs.
   You might want to save "/run/initramfs/sosreport.txt" to a USB stick
   or /boot
   after mounting them and attach it to a bug report.

If you can help me get past this, I would be very happy!

Thank you,
Ray Mansell

On 7/16/2013 08:33, Dan Horák wrote:

Hello everyone,

Fedora 19 GA release for the IBM System z is here. This time two weeks
later than the primary mainly because me being on vacation. It's hard
for me to take time off work when there are always some deadlines in
sight :-) And again we are closer to primary when we count the number
of available packages.




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Ten years of Linux on the mainframe

[Ray Mansell]

You joined seven months earlier than I did - so between us we have more
than 7 decades of VM experience.

Yikes!

Ray

On 6/11/2010 19:38, Jim Elliott wrote:

And I should have noted, today is the 37th anniversary of my
joining IBM. Most of that has been spent working on VM and
more recently Linux (with a short excursion on OS/2 Warp). For
years I was told that working on VM would be a dead-end to my
IBM career. Linux has happily proved that wrong.



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: SLES10 SP2 question

[Ray Mansell]

Thank you, David and Mark.

As it turns out, I *had* done the right thing (changed the boot parm in
zipl.conf), but due to a finger check had managed to specify the wrong
partition. Your responses caused me to double-check my work and thus
discover my error.

Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

SLES10 SP2 question

[Ray Mansell]

I need to move a SLES10 SP2 server from one DASD volume to another
(since we just installed a new storage box). However, after the move I
discovered that the server's boot parms included the following:

root=/dev/disk/by-id/ccw-IBMetc

Naturally, the id of the new DASD device is not the same as the
original, and the server refuses to boot. If I boot the server from its
original home, is there a way to change the root= parm to be something
rather less specific so that it will then boot in its new home?

Many thanks,
Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Dynamic configuration changes

[Ray Mansell]

Mark Perry wrote:

Rare - No, infrequent more like. Of course it depends on planning :-)


Thanks to all who replied... your responses have pretty much confirmed
what I suspected, and the quote above ("it depends on planning") is also
most apt.

Thanks again,
Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Dynamic configuration changes

[Ray Mansell]

Mark Perry wrote:

Yes Ray, when adding extra DASD to a non-stop server, and of course a
lot more in a testing environment, we even add and remove CPUs for
different tests.

Mark


Thanks, Mark... but this is a relatively rare event, yes? I should have
been a little more explicit, in that I'm trying to determine if such
dynamic configuration changes might be something that happens often. In
CMS applications, for example, it is commonplace to LINK/DETACH
minidisks often during a session. I'm pretty sure such configuration
changes are much rarer in Linux, is that correct?

Thanks again,
Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Dynamic configuration changes

[Ray Mansell]

I have a feeling this is something that rarely happens, but how common
is it for Linux servers running in a virtual machine to change their
configuration on the fly? For example, although it is obviously possible
to LINK or DETACH virtual devices while the server is running, does it
ever happen in practice?

Many thanks...
Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Make utility

[Ray Mansell]

Kim Goldenberg wrote:

Not usually so hard. often the Linux three-step:

1. ./cofigure
2. make
3. (as root) make install

Kim

Exactly what I was looking for - thank you, Kim!

Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Make utility

[Ray Mansell]

Eli Dow wrote:

Hi Ray,

Red Hat Enterprise Linux AS release 4 (Nahant Update 5) contains make
3.8 as does my sles10 sp1 system here in the lab.  I imagine it might
need to be built from source if you need that particular version :)


First, I apologise for somehow managing to hang this off an existing
thread - my fault.

Thank you both for your replies. SuSE SLES10 certainly has make 3.80,
but I need make 3.81, which is significantly different. I have
downloaded the 3.81 source, but it isn't obvious (to me) how to build
it, so I was hoping for a ready-built solution :-)  If that isn't
available, I guess I'll have to bite the bullet and attempt to build it
myself.

Thanks again,
Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Make utility

[Ray Mansell]

Is MAKE 3.81 available for mainframe Linux, and if so, where might I
find it?

Thank you...
Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: LPAR for z890

[Ray Mansell]

David Kreuter wrote:

Well, I wouldn't go down the road of comparing one linux in an lpar to one 
linux virtual machine in an lpar running z/VM.  The strength of running z/VM, 
well, one of the main ones, is in its ability to run many linux critters 
simultaneously. I'd be very careful about what such a comparison intends to 
show, and will show.

I've heard various numbers lately 'bout vm overhead.  In my experience on z9 
class machines, I'm putting it in the less than 5% range, including CP overhead 
AND CMS overhead (CMS servers doing systems work). Other numbers I've seen put 
it higher, but I don't.

David


I would echo what David says, and would also add that, although you can
share processors with LPAR, you can *not* share memory. Thus, the memory
you (by definition) dedicate to your LPAR is no longer available to
other LPARs.

So be aware that even though, as someone has already pointed out, you
may well see a performance improvement, nothing is free, and something
else on your system will suffer. If it is important for this particular
Linux to have certain performance characteristics, then perhaps you
could achieve that by using some of z/VM's performance knobs and dials
instead?

Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Linux and Railroad Diagrams

[Ray Mansell]

Your Name wrote:

Note that the examples provided by IBM's online books use only -+|<>
characters, making the diagrams somewhat harder to read than if extended
box characters were used.  The original diagrams in the Pascal book used
quarter- and semi-circular joins, making the diagrams more
"railroad-like" than the diagrams in use these days

... which is why, many years ago, I wrote RRT XEDIT which renders the
HELP diagrams using graphic characters where possible. *Much* more
readable, and prettier too.

Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: How do I find ...

[Ray Mansell]

SPident

LJ Mace wrote:

What fix paxk oof sles I'm on?? I know I'm on sp3 but
let say I fall and bump my head tonight and tomorrow
someone asks me "What fix pack of sles are we on?"
I could try a unmae -a ,if I rememebered it but that
doesn't show it.

thanks
Mace




--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Changing DASD addresses

[Ray Mansell]

Mark Post wrote:

No.  The _first_ device number presented to the kernel becomes dasda.  The
second becomes dasdb.

Well, wouldn't you just know it? My colleagues decided to completely
reinstall Linux on the new system, rather than trying to adjust the
device addresses!! (Where's their sense of adventure?)

In any case, thank you all again for a most informative thread. I have
learned many new things, and have been delving into initrd and other
places I had never visited before, so it was far from a waste of time.

Thanks again,
Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Changing DASD addresses

[Ray Mansell]

Romanowski, John (OFT) wrote:

A brute force method for SLES 9  would be to edit the initial ramdisk's
/linuxrc script to insert lines like
/sbin/dasd_configure 0.0.16zz 1 0



Thank you, John... I will try that right now.

Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Changing DASD addresses

[Ray Mansell]

Kyle Smith wrote:

SUSE creates files in /etc/sysconfig/hardware for each DASD or qeth
device
and uses hotplug/coldplug to bring them up on boot.  If the files get
renamed & the contents updated I would assume the changes would get
picked
up, although YaST may keep it's own cache somewhere.


Yes - I knew about those, but was curious also about how the devices are
mapped to dasda, dasdb, etc., since that information is not included in
those files.

Klaus gave me a couple of other ideas to try, so I'm experimenting with
those now.

Many thanks for all the hints so far...

Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Changing DASD addresses

[Ray Mansell]

Klaus Bergmann wrote:

Change the "dasd=" parameter in /etc/zipl.conf, then execute "zipl -V", and
you are prepared for the new addresses.



Thank you, Klaus. However, zipl.conf does not contain such a line, at
least on the system in question, so where else might Linux be obtaining
such information?

Ray

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Changing DASD addresses

[Ray Mansell]

A colleague had a test Linux server running in a partition on one
machine, but has now been moved to a different machine where the same
DASD devices are mapped to a different address range (16xx as opposed to
15xx). Needless to say, booting his Linux server doesn't result in a
great deal of success.

Assuming he is able, temporarily, to revert to the original addresses
and bring his system up, is there some way to tell Linux about the new
addresses?

Many thanks,
Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: New DASD

[Ray Mansell]

It's hugely more difficult to map (and keep consistent over time)
virtual addresses onto real ones, one to one, than it is to determine on
which real volume a particular guest's minidisk resides.

Ray Mansell

Mike Lovins wrote:

The main reason I asked the question was because my Z/VM specialist does not 
want to keep track of virtual address. He said that when a virtual drive 0750 
has a problem it is harder to trackdown which one is really having a problem. I 
thank he doesn't need to worry about that because the device will call IBM or 
Hatachi or whom ever and tell them which device is having the problem...


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: New DASD

[Ray Mansell]

But then if I modelled it on the LINK statement, I'd wind up back to
front again :-)

Ray Mansell

Marcy Cortes wrote:

If you think of it as device statements in the dir are usually SOMETHING
VADDR, then remembering that the virtual address is first makes more
sense.


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: New DASD

[Ray Mansell]

You are not alone. I *always* have to check DIRM HELP before adding a
dedicate statement. On the other hand, it is similar to, for example,
the MVC instruction in that the 'target' appears before the 'source'.

Ray Mansell

Bates, Bob wrote:

Careful on the DEDICATE statement. I never understood why but the addresses 
seem backwards to me. The virtual address is first and the real address second. 
So the DEDICATE would be

DEDICATE 991 10FC

Bob



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: collaborative memory management on Linux weekly news

[Ray Mansell]

Barton Robinson wrote:

(So how do you validate that CMM2 is working?)


Use the CP QUERY MEMASSIST command (alone, for a class G user, or
together with FOR /userid/ for a class B user). If the response shows
/ACTIVE,/ then Linux is using CMM2; if the response shows /INACTIVE,/ it
is not.

For example:

q memassist
ALL USERS SET - ON

USER  SETTING   STATUS
MANSELL   ONACTIVE

Ray Mansell



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Tracing question

[Ray Mansell]

Gentlemen,

Many thanks... I think I have it now.  Much appreciated...

Ray

Richard Hitt wrote:

Hi, Ray

gdb will give you an instruction trace.  Use the gdb command "display/i
$pc" and then use either "si" or "ni" to step instructionwise through
your program.


Neale Ferguson wrote:

Use gdb to stop the program where you are interested. Get its address. Quit
gdb. Go to the VM console:



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Tracing question

[Ray Mansell]

Thank you both for the responses, but this isn't quite what I'm after. I
really do need a CP instruction trace of a given program running in
Linux, and as far as I can tell, neither gdb nor ptrace will give me this.

Thanks again,
Ray

Alan Cox wrote:

Ar Llu, 2006-08-21 am 10:23 -0400, ysgrifennodd Ray Mansell:

The low level interface is ptrace (2) (see man 2 ptrace).



McKown, John wrote:

GDB - The GNU Debugger. It use the "ptrace" function in Linux to do
these things. It is also a source level debugger if you compiled the
program with the "-g" switch.



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Tracing question

[Ray Mansell]

Please forgive the naivety of this question, but my knowledge of Linux
is severely limited.

Back in the good old days of VM and CMS, it was easy to load a program,
locate it in storage, set a few CP trace traps within it, and then start
it running. How can I do the same thing in Linux? Specifically, I'd like
to be able to trace the entire execution of a given program running
under Linux, but I have less than a clue as to how to do that.

Many thanks...
Ray Mansell

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Unexpectedly closing VM console breaks Linux server under z/VM

[Ray Mansell]

The length of time after which CP will force a disconnected user that
has an outstanding read is configurable in the SYSTEM CONFIG file (in
the Features section, it is the Disconnect_Timeout setting, which can be
set to OFF to disable the function entirely). The default value is
fifteen minutes, as I recall.

Ray Mansell

Dominic Coulombe wrote:

Hi,

my observations are that you absolutely need to disconnect from a running
linux machine.

If you logoff, your machine is killed.

If you close the TN3270 or it crashes, you have to reconnect quickly
to your
running session, then you can disconnect.  You might need to enter
"begin"
when reconnecting to the session, but it depends of your profile.

I don't know what is the length of the timer, but you need to reconnect
quickly.  As soon as you reconnect to the TN3270 session, your Linux
guest
will continue its operations.

Regards.



--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390