guest that IPLs from SCSI.
Might have made sense for Novell to be more specific rather that raise
this as a required fix related to this kernel. If have a feeling it
might pull a lot of other service in which non-SCSI folks otherwise
would not care about.
Rob
--
Rob van der Heij
Velocity Softwar
On 7/24/06, Dominic Coulombe <[EMAIL PROTECTED]> wrote:
I'm sorry, but we managed to do live backup of our systems without any
problem. We restored a lot of backup and all were recoverable without any
problem. Even when data was stored on LVM volumes.
You've been lucky. That sometimes happen
On 7/24/06, Carsten Otte <[EMAIL PROTECTED]> wrote:
Well, that is what the device-mapper snapshot target is for:
a) make a snapshot, which is very fast because it does not actually
copy data
b) use dd to copy the snapshot to a backup dasd while writing to the
original disk
c) destroy the snapsho
On 7/24/06, Dominic Coulombe <[EMAIL PROTECTED]> wrote:
in *theory* you can do live backup of machines with journaled filesystems
(at least ext3) without any problem.
Stop dreaming. Not even in theory - at least not my theory.
Linux will repair filesystems from journal after recovering your
ay with
n>2, isn't it?
For n=2 that improvement was already with earlier releases. I have
read that discussion several times but the "nice improvement" was not
obvious to me.
Rob
--
Rob van der Heij
Velocity Software, In
But is dasda indeed online? Is the device driver module loaded?
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to
On 7/19/06, Carsten Otte <[EMAIL PROTECTED]> wrote:
> When the folks in Boeblingen don't try compilation for s390
> and don't test it on 32-bit anymore, we can expect some more problems
> in that area. I have seen such things already with dependencies on new
> hardware that can not be fully conf
On 7/18/06, Jim Elliott <[EMAIL PROTECTED]>
<[EMAIL PROTECTED]> wrote:
There will be NO further 31-bit (32-bit) distros (the s390
versions) for the mainframe. IBM announced months ago that from
now on only "patches" for the s390x (64-bit) would be made
I believe you were corrected on this alre
But when Oracle is starting
this processes, I have a slight feeling they will also check their
presence already and act upon problems.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LIN
On 7/11/06, Rick Troth <[EMAIL PROTECTED]> wrote:
Consider what Kris said about the '-i' flag on 'sudo'.
It appears there's no such flag in the sudo that I have with SuSE so I
can't tell.
I believe my approach with coding su for one command is similar in effect?
And even the auditing is as
ery useful to make you remember when you did
what, and sometimes why you did it. It also helps to educate
colleagues after you've been called in the middle of the night. I just
don't give much for restricting root access that way.
Rob
--
Rob van der Heij
Velocity
nd they source configuration files that you
may not all protect. IMHO letting people run this under sudo only
provides the illusion of security.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX
on VM (and use SCIF) or another web server on the Linux
server.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMA
bit Diag250.
IIRC the format used in the lastest Linux kernels use the new format
that only works on z/VM 5.2, whether your virtual machine is large or
small.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
---
On 6/29/06, Bill Carlson <[EMAIL PROTECTED]> wrote:
I can think of only one reason to use the console on a HEALTHLY system,
and that's assuming the system is secured correctly. And that would be to
verify the root password.
.bx on id=soap
In a former life we had inittab such that root would au
un the
real application with copies of real data sounds wiser.
And remember it's not always about how fast you can run, but how many
can run at the same time without slowing each other down. It's often
about throughput and cost rather than maximum speed.
Rob
--
Rob van der Heij
Veloci
On 6/28/06, Leland Lucius <[EMAIL PROTECTED]> wrote:
You'll also want to ensure you retain the modules in /lib/modules for the
previous version of the kernel. Again, this is handled when installing a new
RPM, but I'm not sure how the SP does it.
Just like the kernel. When you 'rpm -U' for the
and/or file
system that could deal with this, we would have a pretty fast disk
device for shared data.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / arc
On 6/22/06, John Summerfield <[EMAIL PROTECTED]> wrote:
I'd rather see a benchmark involving Oracle on the Z and the Java app on
the {pSeries,iSeries,Sun} and if you can find a decent Opteron box, that
too.
That would make sense.
And if it has to be all on zSeries, we learned that it does help
hat implements a wait by counting with very small steps). In that
case your measurement is not very useful this way.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff
ntation and run the test
again.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the m
Q3. MDC is probably eating a big chunk
of memory as well. I would expect one of them sitting in E3 now and
then.
If you care to share some monitor data with me, we could validate
these assumptions.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitys
t;hcp" and "mount"
command under the covers to get hold of the disk. Very easy.
Rob--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructi
gure the virtual machine
with enough virtual processors. Whether the application is up to using
that is indeed another matter.
When DB2 is capable to use multiple CPUs and you think you are CPU
constrained, that may make sense if you don't need it on other virtual
machines.
Rob
--
Rob van der Heij
V
e what is going on. One can waste
an entire night trying to understand why file transfer would fail
after so many MB (eventually the data gets discarded and is seen as
corrupted disk).
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
sionals at Velocity Software's educational facility in
Mountain View, California (Silicon Valley).
"http://velocitysoftware.com/workshop.html"; for details.
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
-
rowed their unused z/OS-test LPAR for Linux (which had access to
all production MVS packs). That is probably not a good idea because
Linux will see all the packs the LPAR can access. The dasd= parameter
provides no protection.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitys
wipe out some MVS data (this is often not helpful in
getting management backing for your Linux project).
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive acc
ARs to provide relieve
for the 2G constraint with earlier z/VM releases running large Linux
guests. For those installations it could be an advantage to migrate
back to a single LPAR (and avoid the trouble with linking from
different LPARs as well).
Rob
--
Rob van der Heij
Velocity Software, Inc
h
one in
Boeblingen to allow the PARM option to specify additional options. But
each time I have the kernel source handy I forget to look whether it
was already published...
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftwa
IBM has no plans to make WAS CE available on Linux on System z.
I could see some value in potential hardware sales for IBM if peope
would be able to download and try WAS on System z.
Rob
--
For LINUX-390 subscribe / signoff /
if it burns more gas it
must be faster...
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the me
27;s (so no z/VM nor z/OS). Normally you'd use z/VM or z/OS
to capture PR/SM CPU usage numbers on LPAR level and use that for
capacity planning. If you only run Linux on the machine, this driver
allows you to capture that data from PR/SM.
Rob
--
Rob van der Heij
Velocity Software, Inc
http:/
easure instruction speed as the number of
instructions you can stick in between without slowing down the thing.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive
ood performance monitor you can even
understand what the problems are rather than just reject the platform
as is.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / ar
wered you here and came to the conclusion both
mini disk cache and CPU contention between the two virtual machines
explained the behaviour. Did you miss those posts?
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
Five years ago this class of publication would entertain us by
calculating that with 100 virtual machines, each Linux server would be
getting equivalent of a PIII at 7.5 MHz. At least I did not spot that
in the part that I read. We're making progress in educating the press.
:-)
Hey, five years a
Your new drivers may
pick up the wrong mode.
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the mes
the upgrade? The LVM commands may
reveal whether this physical device was indeed part of a volume group.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archiv
s - if it's in use
you wait for a few minutes and try again.
Don't know how NFS does this right.
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive acc
e new disk.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-3
ssing to do.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or
y have happened. When Linux was running with part
of the data in cache and other parts not, it may have picked up
inconsistent data from disk and used that to create files on its R/W
disks.
--
Rob van der Heij
Velocity Software, Inc
http:/
he sniffer was meant to analyze functional issues.
Though I have not seen any measurements of it, I would be surprised if
sniffing the traffic would not have a performance impact itself.
Rob
--
Rob van der Heij
Velocity Software,
ations store the real data in a separate database that can
be backed up and recovered with specific tools.
I have talked to a lot of people who don't backup their individual
servers but rely on an approach to rebuild them. Especially with Linux
on z/VM this offers a lot of advantages.
Rob
--
Rob
chine on VM
that would restart your Linux service again? You'd need such
communication also when you start your backup, don't you?
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LIN
u also added the correct IUCV statements to allow the FTP server to
talk to the stack?
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send
definitions similar. You could copy segments
using the stuff we inherited form IPF, and since the 190 and GCS disk
are shared, saving the NSS is not that hard.
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
terminated without any messages, and when I did set up a
GUI it finally showed me the error message (insufficient space on my
disk). Sigh.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-39
On 5/3/06, Feller, Paul <[EMAIL PROTECTED]> wrote:
What type of problems would you run into sharing the same SYSRES. I
Checkpoint, warmstart, and weird effects from the directory...
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftwa
the other system too. Back in the days where we sometimes lost a spool
pack, the big NSS files mostly had some blocks on the lost pack and
were all purged. Having the NSS on a single pack improves your odds.
:ebox.
--
Rob van der Heij
Velocity Software, Inc
h
a
GUI, a better approach is to run the desktop on your workstation and
have the application show there (with ssh and X11 Forwarding that it
real easy).
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
27;s not
uncommon for installations to misconfigure this so that no response is
received and you're waiting for the timeout.
When logged on, try the 'host' command with your workstation
IP-address to check if that takes long too.
Rob
--
Rob van der Heij
Velocity Softw
On 5/1/06, Adam Thornton <[EMAIL PROTECTED]> wrote:
I guess you could run badblocks with various blocksizes and see which
one DOESN'T give you tons of errorsbut there's got to be a nicer
way.
dumpe2fs -h /dev/dasdb1
--
Rob van der Heij
Velocity Software, Inc
http://velo
r subsystem are that
share some of the logic inside the box). Typically that means avoiding
using a range of consecutive real device numbers for a single server.
That seems to have a much bigger impact.
Rob
--
Rob van der Heij
Velocity Software, Inc
h
3390-3, then you probably don't
want to combine them on a single 3390-9.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send
at the packet is
indeed being sent. So the question is how the other side of the
hipersockets is configured (see the threat about omproute etc earlier
this week).
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftwa
ime change because in dual-boot you never knew
whether another OS already had changed the RTC to DST.
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access ins
hat PR/SM does not have a CONCEAL option, but I don't
think we should avoid exploiting z/VM features just because PR/SM does
not have similar flexibility.
Rob
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
---
set up separate
things to monitor for servers sitting in CP READ (like HMF can do) or
watch the console with SCIF, or simply issue SEND CP xx IPL to any
Linux that appears to be unresponsive... :-(
--
Rob van der Heij
Velocity Software, Inc
h
On 4/28/06, Rob van der Heij <[EMAIL PROTECTED]> wrote:
[ something in Dutch - wrong finger - my apologies ]
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the m
s nam hwcfg-qeth-bus-ccw-0.0.4000 (merkwaardig genoeg onder ccw
ipv ccwgroup omdat je bij hardware detectie nog niet weet dat het een
device is dat meer dan 1 subchannel nodig heeft).
Hoop dat dit je iets verder helpt. Rob
--
Rob van der Heij
Velocity Software
ct for the cases where its
system clock failed to tick properly :-)
And now we have this clear, should we run UTC or GMT ?
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe
hanks again!
Be aware that the default does not include shared segments (in case
you have kernel in NSS or use DCSS). And getting the double quotes
in zipl.conf may take some experiments.
May you never find out whether this works :-)
--
Rob van der Heij
Velocity Software,
aused the dump, or at least one with exactly the same level of
kernel).
--
Rob van der Heij
Velocity Software, Inc
http://velocitysoftware.com/
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [E
.0 +0100 file64.tmp
29 2004-10-31 02:00:02.0 +0100 file65.tmp
Now as far as I recall the rmfpms gatherer is keeping the archive in
files that are named after the local time, so you'll probably lose
your data in 6 months. Since it's in the middle of the night, you
ngs in the /sys file system. The SuSE boot scripts
use control files in /etc/sysconfig/hardware and
/etc/sysconfig/network to define the devices. YaST will define these
files for you if you want.
Have a look at the current Device Drivers book to see how you define
them manually.
Rob
--
Rob van der He
ards are simply stuck together as-is to form one
string (so indenting cards with a blank helps assure you have
separators).
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instruction
hem onto each other
> to produce custom-tailored images. Obviously, they had it working on s390
> back then, as a bundled product at least.
>From what I know, Levanta have their 'mapfs' that was made GPL half a
year ago or so. I don't think I ever looked in detail a
that on this slow CPU the extra overhead was making a
huge difference for me, so much that I had to give up using it.
Will send Mark the PDF for the presentation...
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 su
. But I do agree it would be easier to configure with a
VSWITCH on an OSA that is shared with MVS.
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to
at the same time. But with LVM you
can quiesce the disk for a moment to allow the backups to be taken.
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to
ata to have related stuff on the same disk.
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
performance monitor could spot that moment. There's timer driven
write-out of dirty pages, and when Linux has not been writing data for
more than 30 seconds (or whatever it's set) we know it has no such
desire anymore. That might be the moment to flash the disks.
Rob
--
Rob
image with the MD5SUM that's provided for the DVD image.
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LIN
o there (e.g. fix network
address, repair file system, etc).
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INF
n still get it into CP READ but not
because your terminal session was terminated.
/* LNXSEC EXEC Play secondary console for disconnected server */
/* Author: Rob van der Heij, 20 Feb 2006 */
arg uid .
if uid = '' then exit 24
signal on err
On 3/22/06, Bruce Hayden <[EMAIL PROTECTED]> wrote:
> What I have found is that the rpm sles-release-9-82.nn.s390(x).rpm is
> on the service pack CD but *not* part of the service applied by YOU.
> If you find that rpm on the Service Pack cd and update it, then the
> output from SPident should indi
n with the guest?
You need to collect performance data and see whether the server is
running (and if so, why it is not doing the work you need to get done)
or whether it is held back by something.
--
Rob van der Heij
Velocity Software, Inc
--
ll aready have used to the corporate
gun for abuse of someone else's trademark... :-)
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL
for this, you can buy from someone who did. I see that
http://www.cheeplinux.com/ explictly lists the s390 and s390x CD's
(you want to make sure not to get the obvious default) at GBP 10 per
set, but there may be places that are wiser w.r.t. shipping cost.
Ro
systems such as Windows."
Everyone is unique in some way, but...
"NEW ORLEANS, May 22, 1995 . . . IBM today announced at the IBM
Technical Interchange conference the first IBM PC server that can run
both PC and mainframe-based applications. The new server... "
--
R
stand when to retrieve what metrics so that you capture
enough detail with sufficient resolution, and still not generate
workload yourself by doing so.
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe
use a Linux server somewhere to write the
bootstrap on a volume...
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the
records will take on disk, so that you can predict whether it will
fit. Writing until it fills and bail out is not always an option.
When the curves have jumps, it becomes almost impossible to aim right.
In this case, you would be tempted to think 60+5 = 65 (plus a little
bit) but crossing the 64M
that (yeah, I know... but "it depends" does not
work in real life).
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED
99k
data, 64k init)
When the kernel has initialized, the 64k init is freed again,
resulting in 43080k or
cat /proc/meminfo | head -n 2
total:used:free: shared: buffers: cached:
Mem: 44113920 25088000 190259200 3051520 10993664
And the 44113920/1024 is 43080
--
Rob van de
Plus the (constant) size of the kernel and data area.
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINU
On 2/23/06, Ihno Krumreich <[EMAIL PROTECTED]> wrote:
> Yes. But as you said. Its a "Release Candidate" and this release candidate was
> declared gold master.
Considering the track record with SP3, starting with "bronze master"
would have been appropriate ;-)
Rob
---
about half of the links were dead or incorrect ;-)
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO
ars.
Rob
--
Rob van der Heij
--
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
1 Forwarding is
that an authentication token is generated and XAUTH (?) environment
variable is set up to authenticate through that. IIRC it is possible
to miss installing the SuSE package that allows the ssh session to do
this.
Rob
--
Rob van der Heij
Velocit
u could not try that?
Rob
--
Rob van der Heij
Velocity Software, Inc
--
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
hat is not good for CP to understand this server is
idle.
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX
e than tight security. Virtual machines running
Linux and connected to the outside world form a completely different
audience and you may want to adapt your configuration.
:eplug.
Rob
--
Rob van der Heij
Velocity Software, Inc
--
For
on the door :-)
Rob
--
Rob van der Heij
--
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
ntrol much better. When
the CP directory is the only control, then it is unwise to force the
adminstrator to define the same thing in two different places because
that causes stale authorisations.
Rob
--
Rob van der Heij
Velocity Software, Inc
--
logon (provided the user may
connect). Instead of logoff / logon you could have issued the COUPLE
300 ... yourself and witness Linux pick it up with witty remarks like
"you are lucky, someone plugged in the cable"
Rob
--
Rob van der He
the PLO would be that only the manual
COUPLE command is subject to the GRANT settings.
:erant.
Rob
--
Rob van der Heij rvdheij @ gmail.com
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive acce
ore of the
entries on the Google screen.
Rob
--
Rob van der Heij rvdheij @ gmail.com
Velocity Software, Inc
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] wit
801 - 900 of 1744 matches
Mail list logo
