Greetings; (Posted to VMESA-L and VSE-L and LINUX-390)
- - Now in its sixth year! - - 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
You hit it right on the head. Somehow tar was gone from this instance of Linux. It's
funny, this is the one I use to clone, and I just went to another instance, found it
and copied it back.
Once tar was there, it worked great.
I really appreciate your help
Thanks
Gene
[EMAIL PROTECTED]
The Q NIC shows E04/E05/E06 and the original CHANDEV.CONF reads E03/E04/E05
because I changed from 3/4/5 to 4/5/6 to see if that would fix it and then I
did the Q NIC to respond to Mark.
Last night I took out the chandev= string and rebooted. The LCS driver
failed to come up but the QETH driver
I discovered something last week, and I don't know if most of you know about it or not
so I thought I would share.
I found a program called cygwin. The feature I like is that it lets you do Xwindows
stuff from Linux back to your windows desktop(Ie yast2 etc.) Up til now I had been
looking at
On Thu, 1 Apr 2004 09:01:06 -0500 Kern, Thomas said:
Last night I took out the chandev= string and rebooted. The LCS driver
failed to come up but the QETH driver worked fine. There does seem to be
some incompatibility between LCS and QETH. I doubt IBM ever tried both in
the same linux server.
I
It's funny, this is the one I use to clone, and I just went
to another instance, found it and copied it back.
Once tar was there, it worked great.
I really appreciate your help
Glad you are back in business now, though the missing binary IS
disconcerting. You should probably take inventory
Did you remove the complete chandev= statement, or just remove the
chandev= part of it? You need the rest of the statement, just not the
chandev= prefix on it.
Mark Post
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Kern,
Thomas
Sent: Thursday, April
I removed JUST the chandev= string preserving the rest of the line. The
LCS code attempted to initialize but failed saying it could not find any LCS
capable cards.
/Thomas Kern
/301-903-2211
-Original Message-
From: Post, Mark K [mailto:[EMAIL PROTECTED]
Sent: Thursday, April 01, 2004
Thomas,
Been there, done that... just recently. Change the comma after the noauto to a
semicolon(;).
Michael A. Geiger
Sr. Operating Systems Programmer
CommerceQuest, Inc.
5481 W. Waters Ave.
Tampa, FL 33634
Tel. 813.639.6516
-Original Message-
From: Kern, Thomas [mailto:[EMAIL
After changing 'CHANDEV=noauto,' to 'CHANDEV=noauto;' this is what I get.
cat /etc/chandev.conf
chandev=noauto;lcs0,0x39eo,0x39e1,0,0,1,1
add_parms,0x10,0xe00,0xe02,portname:VMLAN00
qeth1,0xe00,0xe01,0xe02,0,0
add_parms,0x10,0xe04,0xe06,portname:VMLAN02
qeth2,0xe04,0xe05,0xe06,0,0
Console log on
I don't think that you get this advantage on eckd. My understanding is that Reiser
uses the fact that SCSI and IDE disk use 512 byte sectors to get the small file space
savings. In other words: it does not read/merge/write, it just writes 512 bytes at a
time for small files, and 4K at a time
Thomas,
Remove the 'chandev=.
I think you want noauto;lcs0,0x39eo,0x39e1,0,0,1,1
Michael A. Geiger
Sr. Operating Systems Programmer
CommerceQuest, Inc.
5481 W. Waters Ave.
Tampa, FL 33634
Tel. 813.639.6516
-Original Message-
From: Kern, Thomas [mailto:[EMAIL PROTECTED]
Sent:
Hi,
On Thu, 2004-04-01 at 17:19, Jamie Lokier wrote:
Some documentation I'm looking at says MS_INVALIDATE updates the
mapped page to contain the current contents of the file. 2.6.4 seems
to do the reverse: update the file to contain the current content of
the mapped page. man msync agrees
Try this:
noauto
lcs0,0x39eo,0x39e1,0,0,1,1
qeth1,0xe00,0xe01,0xe02,0,0
add_parms,0x10,0xe00,0xe02,portname:VMLAN00
qeth2,0xe04,0xe05,0xe06,0,0
add_parms,0x10,0xe04,0xe06,portname:VMLAN02
Mark Post
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Kern,
All drivers fail on that one, Mark. This level of Liunx/LCS really wants the
'chandev='.
This is the latest that allows at least the LCS driver to come up:
chandev=noauto
lcs0,0x39eo,0x39e1,0,0,1,1
add_parms,0x10,0xe00,0xe02,portname:VMLAN00
qeth1,0xe00,0xe01,0xe02,0,0
Hello all,
I probe to sharing my mdisk root in read only with others server but when load
Linux (IPL) I receive input output errors because the root dasd is not in
read write mode.
Have you any method or procedure to sharing between severals servers
Thanks
A+
--
Carlos ROMERO-MARTIN
Carlos,
I don't know of anyone who's really spent any time trying to do this. The
benefit you would gain from it (saving ~15MB per instance) really isn't that
big. Especially compared to the pain involved.
Mark Post
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED]
I don't know of anyone who's really spent any time trying to do this. The
benefit you would gain from it (saving ~15MB per instance) really isn't that
big. Especially compared to the pain involved.
Wouldn't one set of cache entires versus dozens or hundreds make a difference in a
large
Because of the timing of a meeting with IBM and SuSE and our own
maintenance schedule, we re-carved the memory on our z/800 to create a 2 gb
linux lpar, a 1 gb linux lpar and my 768 mb twiddling LPAR.
Then things happened and a schedule for IBM to come in and do a bunch of
stuff (loan us an IFL,
I am trying to add another disk to one of my SLES8 Linux images on z/VM.
I added the mdisk statement to the user direct, excuted the directxa,
and logged onto the guest.
after cms formatting the new mdisk, iI added the new disk address to
/boot/zipl/parmfile and ran zipl.
After rebooting, I
I think you added the device to the wrong file. I usually add it to /etc/zipl.conf,
then run zipl. I believe zipl copies zipl.conf to the /boot files (among other
things).
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of
Scorch Burnet
Sent: Thursday,
We're running z/VM 4.3 and 37 penguins happlily on 2GB right now. We will upgrade to
6GB this weekend, though.
Nature and nature's laws lay hid in night:
God said, 'Let Newton Be!' and all was light. - Alexander Pope
It did not last; the Devil howling 'Ho!
Let Einstein Be!' restored the status
As long as you're stingy on virtual machine memory allocations, gobs of
penguins can be squeezed into 2GB. 'Gobs', of course being one of those
scientific numeric expressions... :)
On Thu, 2004-04-01 at 14:42, James Melin wrote:
Because of the timing of a meeting with IBM and SuSE and our own
Umm, no, it does not. With SLES8, SUSE went to using an initrd. The HOWTO
I referenced in another reply talks about how to update that.
Mark Post
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Hall,
Ken (IDS ECCS)
Sent: Thursday, April 01, 2004 3:47
Take a look at this:
http://linuxvm.org/Info/HOWTOs/mkinitrd-notes.html
Mark Post
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Scorch
Burnet
Sent: Thursday, April 01, 2004 3:44 PM
To: [EMAIL PROTECTED]
Subject: new disk for linux
I am trying to add
Due to an addition error (not mine), I'm running about 25 penguins happily
in 1.5G + 160Meg of xstor. It was supposed to have been .5G of xstor.
That's the problem with sharing a box with z/OS - can't ever take it down!
We'll probably have a lot more once these penguins fly north (and east) to a
From what I've seen, and then some. Currently I have a 2gig lpar running
z/VM 4.3, and 5 images. The biggest one is a WebSphere Commerece Business
Edition V5.5 instance.
At 03:42 PM 4/1/2004, you wrote:
Because of the timing of a meeting with IBM and SuSE and our own
maintenance schedule, we
Yeah. All of it is defined to be cental at the moment. What I will probably
recommend is that we combine the 3 GB and 2 GB LPARS into one and make the
768mb one extended stor and move level 1 swap into that.
Marcy Cortes
[EMAIL PROTECTED]
ellsfargo.com
Just a quick note that our Systems/ASM assembler (DASM) is
now available under z/VM with IFL engine prices.
For z/VM RACF management, you need an assembler to do RACF
updates... which is a special bid from IBM. We can help
reduce your costs by substituting Systems/ASM.
We have had some
Uh, yes it does.
Whether or not an initrd is used, the DASD parm comes from the boot parameter file
created by zipl. That's on the disk you boot from, which may or may not also contain
the root filesystem. (In my
case it does.)
I just changed /etc/zipl.conf on my test system, ran zipl, and
In response to an off-list inquiry as to how to enable TN3270 connections to
Linux/390 running in an LPAR, I prepared this sequence of steps. The setup
at the site used 2074 equipment to allow this sort of access for their other
mainframe operating systems, and they were tired of having to use the
-Original Message-
From: Rich Smrcina [mailto:[EMAIL PROTECTED]
Sent: Thursday, April 01, 2004 2:52 PM
To: [EMAIL PROTECTED]
Subject: Re: Recommended memory for a VM LPAR
As long as you're stingy on virtual machine memory
allocations, gobs of
penguins can be squeezed into 2GB.
I always thought that an oodle was a gob of gobs.
-Original Message-
From: McKown, John [mailto:[EMAIL PROTECTED]
Sent: Thursday, April 01, 2004 1:38 PM
To: [EMAIL PROTECTED]
Subject: Re: Recommended memory for a VM LPAR
-Original Message-
From: Rich Smrcina [mailto:[EMAIL
It depends on which side of the pond you are on!
Fargusson.Alan
[EMAIL PROTECTED] To: [EMAIL PROTECTED]
tb.ca.gov cc:
Sent by: Linux on Subject: Re: Recommended memory for a
VM
Sort of like the relationship between a google and a googleplex.
On Thu, 2004-04-01 at 15:58, Fargusson.Alan wrote:
I always thought that an oodle was a gob of gobs.
-Original Message-
From: McKown, John [mailto:[EMAIL PROTECTED]
Sent: Thursday, April 01, 2004 1:38 PM
To: [EMAIL
On SLES8 systems (default, out of the box), the kernel modules and the
script that invokes them are all in the initrd. At the time the DASD driver
gets loaded by that script, there is no way for it to be able to access the
contents of anything on disk. If you're not using an initrd, then of
Thank you all for your responses.
It appears that changing /etc/zipl.conf instead of /boot/zipl/parmfile
was the correct way.
Since I am out here by myself, with old redbooks, it is good to know
taht there are people who are knowledgeable and willing to share their
experiences.
Thanks again to you
The recommendations I was given when I asked a while back was at least 25%
xstor, perhaps up to 50%. This is because of the heirarchical nature of the
VM paging system.
Marcy Cortes
Wells Fargo Services Company
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On
On Thu, 1 Apr 2004, Phil Payne wrote:
Wouldn't one set of cache entries versus dozens or hundreds make a
difference in a large environment?
I suppose, but for the root filesystem there's generally too much
system-unique stuff in there. Keeping that stuff unique while making the
filesystem
Vic,
150MB? I said about 15MB. I think I can fit a whole system into 150MB. :)
Mark Post
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Vic
Cross
Sent: Thursday, April 01, 2004 7:05 PM
To: [EMAIL PROTECTED]
Subject: Re: Sharing root mdisk with SLES
On Thu, 1 Apr 2004, Post, Mark K wrote:
150MB? I said about 15MB. I think I can fit a whole system into 150MB. :)
Yep, I saw that in the ohnoseconds after sending my reply. My point got
an order of magnitude stronger, though! ;)
Cheers,
Vic
41 matches
Mail list logo
