> I did get the SuSE SLES 8, SP2 up at 12MB - Linux saw 8424 KB, so about 4M
> was LPAR overhead. And yes, it was slow during boot and trying to enter
> commands. But the swap size was 3044 KB, so it had to be doing a lot of
> swapping!
If you're really talking about a 12 MB LPAR, then I suspect t
> There is an obscure chandev parameter "memory_usage_in_k". Set this to
> a tiny value and qeth will use the minimum amount of memory for its
> inbound buffers (for example, try "qeth0,0xfd00,0xfd01,0xfd02,1")
But if your inbound buffer is not large enough to catch the data the OSA
received for y
10 * * * * root /usr/sbin/ntpdate -u condor >/dev/null 2>&1
> -Original Message-
> From: James Melin [SMTP:[EMAIL PROTECTED]
> Sent: Tuesday, June 24, 2003 12:35 PM
> To: [EMAIL PROTECTED]
> Subject: Cron sending e-mail
>
> I have this entry in /etc/crontab
>
> 10 * * * * root
Hi,
I am trying to follow instructions to setup a cross build environment
and running into a problem at step 7 "configure, compile and install glibc".
My current system is RH kernel 2.4.9. I downloaded binutils 2.12.90,
gcc 3.3, glibc 2.2.5 and kernel level 2.4.19.
All goes well until the 'make
Jim:
im looking for the Informix ESQL/C, its part of the Informix Client SDK,
but i cant find it for linux390, i found it for intel linux, aix, sco,
etc, but not for linux zseries.
thanks.
On Tue, 24 Jun 2003, Jim Elliott wrote:
> > hi all, we are planning to migrate some systems that actually
On Tue, 24 Jun 2003, Jim Sibley wrote:
> I'm not sure where you got the notion that I only want a router. What I am
> really looking for a minimum system that I can then add the apps that I
> need. Methinks you've made too many assumptions.
I think you didn't say what you wanted, but them maybe I
> hi all, we are planning to migrate some systems that actually are
> running on linux (intel) to linux390, the problem is that the db is
> informix, and the programs need the driver, the info about it:
> ESQL/C versions from 7.2x on should be OK. ESQL/C is now part of the
> Informix Client SDK.
I'm not sure where you got the notion that I only want a router. What I am
really looking for a minimum system that I can then add the apps that I
need. Methinks you've made too many assumptions.
Regards, Jim
Linux S/390-zSeries Support, SEEL, IBM Silicon Valley Labs
t/l 543-4021, 408-463-4021, [E
On Tue, 24 Jun 2003, Fargusson.Alan wrote:
> Try something like this:
> 10 * * * * root /usr/sbin/ntpdate -u condor > /tmp/Log 2>&1
No.
1. Logs in linux belong in /var/log
2. Each run clobbers the previous report.
3. Nobody will look at it.
I'm not entirely happy with my suggestion
On Tue, 24 Jun 2003, James Melin wrote:
> I have this entry in /etc/crontab
>
> 10 * * * * root /usr/sbin/ntpdate -u condor
>
>
> I am getting e-mails every hour that this ran. How do I suppress e-mails to
> root about this?
10 * * * * root /usr/sbin/ntpdate -u condor -s
Read the
On Tue, 24 Jun 2003, Jim Sibley wrote:
> In the thread on minimum memory size, a couple responders made comments
> like:
>
> > kreiserfsd
> No Reiserfs? don't need it.
>
> >From experience, this is an ill conceived statement. For data integrity,
> you need a logged file system in case of system fa
On Tue, 24 Jun 2003, James Peddycord wrote:
> I will give that a shot, but now I am worried about bigger things! PAV was
> turned on to ALL of our DASD last December. The linux instances that were
> already running seem to be OK, but if a problem arises I will be left with
> no support.
> All of m
hi all, we are planning to migrate some systems that actually are running
on linux (intel) to linux390, the problem is that the db is informix, and
the programs need the driver, the info about it:
ESQL/C versions from 7.2x on should be OK. ESQL/C is now part of the
Informix Client SDK.
somebody c
Try something like this:
10 * * * * root /usr/sbin/ntpdate -u condor > /tmp/Log 2>&1
-Original Message-
From: James Melin [mailto:[EMAIL PROTECTED]
Sent: Tuesday, June 24, 2003 10:35 AM
To: [EMAIL PROTECTED]
Subject: Cron sending e-mail
I have this entry in /etc/crontab
10
I have this entry in /etc/crontab
10 * * * * root /usr/sbin/ntpdate -u condor
I am getting e-mails every hour that this ran. How do I suppress e-mails to
root about this?
Jim,
> We would consider it VERY important.
thanks for your feedback.
>Since PAV is not going to be turned off (great z/OS performance
>improvements), Linux needs to be PAV tolerant, even if it does not take
>advantage of it.
Linux is PAV tolerant. It just won't be able operating any alias devi
Ask IBM if there are memory leak issue with DB2/Tivoli-framework...
or you can take a look at /proc/pid/status on all your DB2 agents and
sum up all the storage used by those agents.
|-+>
| | "Fargusson.Alan" |
| |
>You might try turning off the DB2 agent. It looks like that is what the
OOM killer is trying to do.
The OOM killer is not very good at killing things, imho. It attacks the
processes with the most cpu+memory. After awhile, it will attach critical
system functions. (We found that out with a simila
>Back in the early, early days, we successfully IPLed 2.2.? on
>PenguinVM in 10MB.
>We didn't notice until we started investigating why it was running so
>slw.
I did get the SuSE SLES 8, SP2 up at 12MB - Linux saw 8424 KB, so about 4M
was LPAR overhead. And yes, it was slow during boot an
In the thread on minimum memory size, a couple responders made comments
like:
> kreiserfsd
No Reiserfs? don't need it.
>From experience, this is an ill conceived statement. For data integrity,
you need a logged file system in case of system failure (kernel panics,
stupid users, whatever) - ext3 o
You might try turning off the DB2 agent. It looks like that is what the OOM killer is
trying to do. It also looks like you have a lot of them running. Is that what you
want?
-Original Message-
From: Post, Mark K [mailto:[EMAIL PROTECTED]
Sent: Tuesday, June 24, 2003 7:12 AM
To: [EMAIL
On Fri, 20 Jun 2003 09:58:59 -0400, Michael MacIsaac <[EMAIL PROTECTED]>
wrote:
>My two cents:
>
>What would be good to know is the best way to exploit the famed zSeries
I/O
>capabilities. I have questions like:
>- How many PAV "paths" should I use for optimal I/O?
> - What does the IOCDS configur
Tim,
You need some tools that will tell you what process or processes are eating
up your storage over time. Arbitrarily turning off services to figure it
out could take you forever, and is not a good use of your time. (Although I
would say turning Tivoli off would be a good first choice.) This
I had to make my single Linux/390, running Oracle, 1 Gb. At least according
to our in-house Oracle people. Then they decided that a 1Gb Linux running on
an IFL on a z800 did not run as fast as it did on their 8-way Sun E4500 with
10Gb RAM. Duh!
I will give that a shot, but now I am worried about bigger things! PAV was
turned on to ALL of our DASD last December. The linux instances that were
already running seem to be OK, but if a problem arises I will be left with
no support.
All of my boot disks have PAV enabled!
This looks like somethin
Ingo,
We would consider it VERY important.
All of our DASD is PAV enabled. It was enabled last December. All of the
Linux instances that were already running have been OK, but I have to jump
thru hoops to get a new instance created. I'm also concerned about lack of
support should one of our produc
On Tue, 24 Jun 2003, Tzafrir Cohen wrote:
> On Tue, Jun 24, 2003 at 07:14:01AM +0800, John Summerfield wrote:
> > On Mon, 23 Jun 2003, Jim Sibley wrote:
> >
> > > init
>
> You really can't kill init for any practical purpose.
>
> > > migration_CP (1 per CP)
> > > kmcheck
> > > kvventd
> > > ksofti
On Tue, Jun 24, 2003 at 07:14:01AM +0800, John Summerfield wrote:
> On Mon, 23 Jun 2003, Jim Sibley wrote:
>
> > init
You really can't kill init for any practical purpose.
> > migration_CP (1 per CP)
> > kmcheck
> > kvventd
> > ksoftirqd_CP
> > kswapd
> > bdflush
> > kupdated
> > kinoded
> > mdre
Tim,
How long is it from IPL to memory saturation? It almost sounds like somebody
has a leak. What services other than Tivoli are running? Maybe try back off
the memory allocation again and eliminate some service from starting up. If
the machine comes up and is stable, there's your culprit. I
On Tue, 24 Jun 2003, Tim Hanschen wrote:
> Hi,
>
> I have a problem with some linux-guest runnung under z/VM 4.3.
>
> After a special time some linux-guests seems to not answer on any network requests,
> e.g. ssh, ftp, telnet but it is possible to ping the machine. Firstly I thought
> that t
Hi,
I have a problem with some linux-guest runnung under z/VM 4.3.
After a special time some linux-guests seems to not answer on any network requests,
e.g. ssh, ftp, telnet but it is possible to ping the machine. Firstly I thought
that the maschine is dead, but that I noticed, that the resp
While reliability is not the same as availability, availability depends on
it. The less reliable the component pieces are the more redundancy is
required to be available.It depends on what your target availability
is.
Joe Temple
[EMAIL PROTECTED]
845-435-6301 295/6301 cell 914-706-5211 h
Actually, you can tweak the amount of buffers qeth allocates.
There is an obscure chandev parameter "memory_usage_in_k". Set this to
a tiny value and qeth will use the minimum amount of memory for its
inbound buffers (for example, try "qeth0,0xfd00,0xfd01,0xfd02,1")
Best regards / Mit freundliche
33 matches
Mail list logo
