I recently built a CentOS from the 4.3 ISOs. Now I'm ready to build another 10 systems using kickstart. I took the anaconda-ks.cfg file from the first system I built manually, modified it slightly (see below), and made it available via http. I booted the new system from the first CentOS 4.3 CD and typed the following:
boot: linux ks=http://hostname/test-ks.cfg
Here are the changes I made to the kickstart file:
- Replaced "cdrom" with "url --url http://hostname" - Removed xconfig line (slightly different video HW) - Changed the host name in the network line - Uncommented the partition information
The system boots the initial kernel, gets an IP address, gets the kickstart file via http, but then dies with a "installation exited abnormally -- received signal 11" message. I've done this successfully in the past. Can a misconfigured kickstart file cause this problem? How can I debug this?
Alfred
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Wed, May 31, 2006 at 05:22:29PM -0400, Alfred von Campe wrote:
kickstart file via http, but then dies with a "installation exited abnormally -- received signal 11" message. I've done this successfully in the past. Can a misconfigured kickstart file cause this problem? How can I debug this?
In 90% of the cases, a signal 11 means faulty memory. In 9% of the cases, something wrong with the processor (usually overheating)
1%: Others reasons
Suggestion: Running memtest86
- -- Rodrigo Barbosa rodrigob@suespammers.org "Quid quid Latine dictum sit, altum viditur" "Be excellent to each other ..." - Bill & Ted (Wyld Stallyns)
On May 31, 2006, at 17:36, Rodrigo Barbosa wrote:
In 90% of the cases, a signal 11 means faulty memory. In 9% of the cases, something wrong with the processor (usually overheating)
1%: Others reasons
Suggestion: Running memtest86
I'll try that tomorrow when I'm back in the office. Or maybe I'll try one of the other systems. Or both.
Thanks, Alfred
On May 31, 2006, at 17:36, Rodrigo Barbosa wrote:
On Wed, May 31, 2006 at 05:22:29PM -0400, Alfred von Campe wrote:
kickstart file via http, but then dies with a "installation exited abnormally -- received signal 11" message. I've done this successfully in the past. Can a misconfigured kickstart file cause this problem? How can I debug this?
In 90% of the cases, a signal 11 means faulty memory. In 9% of the cases, something wrong with the processor (usually overheating)
1%: Others reasons
Suggestion: Running memtest86
I must be in the other 1% of reasons. I ran memtest86 and I also tried it on other systems. These are brand new IBM (Lenovo) ThinkCentre M52 systems with 4GB of RAM. On all systems it fails with the same error. So I ask again, can a misconfigured kickstart configuration file cause this problem? If not, how can I debug this? How can I capture any logs from the kickstart session? None of the virtual consoles appear to have a running shell where I can access the system.
On a related note, what's the consensus on enabling hyperthreading on a CentOS system? Good, bad, indifferent? The systems will be used as developer workstations.
Thanks, Alfred
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Thu, Jun 01, 2006 at 10:02:33AM -0400, Alfred von Campe wrote:
On a related note, what's the consensus on enabling hyperthreading on a CentOS system? Good, bad, indifferent? The systems will be used as developer workstations.
If it is a 100% workstation only (no mysql, postgres etc), then it is ok to use it. Otherwise, you are asking for trouble.
To be on the safe side, I simply disable it, no matter what.
- -- Rodrigo Barbosa rodrigob@suespammers.org "Quid quid Latine dictum sit, altum viditur" "Be excellent to each other ..." - Bill & Ted (Wyld Stallyns)
On Thu, Jun 01, 2006 at 12:22:11PM -0300, Rodrigo Barbosa wrote:
On a related note, what's the consensus on enabling hyperthreading on a CentOS system? Good, bad, indifferent? The systems will be used as developer workstations.
If it is a 100% workstation only (no mysql, postgres etc), then it is ok to use it. Otherwise, you are asking for trouble.
Any comments on this?
I have a x86-64 system with one Xeon(TM) CPU 3.00GHz running CentOS 4 with 2.6.9-34.0.1.ELsmp (hyperthreading) as a production PostgreSQL 8.x database. No probs so far.
Regards,
Any comments on this?
I have a x86-64 system with one Xeon(TM) CPU 3.00GHz running CentOS 4 with 2.6.9-34.0.1.ELsmp (hyperthreading) as a production PostgreSQL 8.x database. No probs so far.
Pretty much same setup here and no problems, always have HT enabled and never really had any problems. Mileage may vary depending on what you run on it basically as to whether its a good/bad thing in terms of performance.
Henk van Lingen wrote:
On Thu, Jun 01, 2006 at 12:22:11PM -0300, Rodrigo Barbosa wrote:
On a related note, what's the consensus on enabling hyperthreading on a CentOS system? Good, bad, indifferent? The systems will be used as developer workstations.
If it is a 100% workstation only (no mysql, postgres etc), then it is ok to use it. Otherwise, you are asking for trouble.
Any comments on this?
I have a x86-64 system with one Xeon(TM) CPU 3.00GHz running CentOS 4 with 2.6.9-34.0.1.ELsmp (hyperthreading) as a production PostgreSQL 8.x database. No probs so far.
Regards,
I run a dual xeon machine with HT enabled, and see no issues. Good machine.
On a related note, what's the consensus on enabling hyperthreading
on
a CentOS system? Good, bad, indifferent? The systems will be used as developer workstations.
If it is a 100% workstation only (no mysql, postgres etc), then it is ok to use it. Otherwise, you are asking for trouble.
Four x86_64 systems, all dual Xeon with hyperthreading enabled: one dedicated MySQL server, two web servers, one combined web + MySQL server, running CentOS 4.3 (2 systems) and Debian (2 systems) One x86 system, also dual Xeon with hyperthreading enabled, web server (Debian).
Everything has worked flawlessly so far.
Signal 11 is a bus error. I would try turning down the clock on your CPU. If this doesn't fix the problem and the memory isn't bad and the CPU isn't bad then I would go with the MotherBoard.
On 6/1/06, Alfred von Campe alfred@110.net wrote:
On May 31, 2006, at 17:36, Rodrigo Barbosa wrote:
On Wed, May 31, 2006 at 05:22:29PM -0400, Alfred von Campe wrote:
kickstart file via http, but then dies with a "installation exited abnormally -- received signal 11" message. I've done this successfully in the past. Can a misconfigured kickstart file cause this problem? How can I debug this?
In 90% of the cases, a signal 11 means faulty memory. In 9% of the cases, something wrong with the processor (usually overheating)
1%: Others reasons
Suggestion: Running memtest86
I must be in the other 1% of reasons. I ran memtest86 and I also tried it on other systems. These are brand new IBM (Lenovo) ThinkCentre M52 systems with 4GB of RAM. On all systems it fails with the same error. So I ask again, can a misconfigured kickstart configuration file cause this problem? If not, how can I debug this? How can I capture any logs from the kickstart session? None of the virtual consoles appear to have a running shell where I can access the system.
On a related note, what's the consensus on enabling hyperthreading on a CentOS system? Good, bad, indifferent? The systems will be used as developer workstations.
Thanks, Alfred
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Signal 11 is a bus error. I would try turning down the clock on your CPU. If this doesn't fix the problem and the memory isn't bad and the CPU isn't bad then I would go with the MotherBoard.
Well, I have only tested 3 (out of 10) systems, and they all get this error. I doubt the hardware is bad, as these are brand new systems and memtest86 doesn't report any problems, the CentOS 4.3 LiveCD boots fine, and (gasp) the pre-installed Windows XP boots fine.
I'm trying to find a way to debug this to find out the exact cause of this error. Maybe I'll start by putting the kickstart configuration file on a floppy (although I'm sure that it is being read OK from the web server)...
Alfred
On Jun 1, 2006, at 12:21, I wrote:
I'm trying to find a way to debug this to find out the exact cause of this error. Maybe I'll start by putting the kickstart configuration file on a floppy (although I'm sure that it is being read OK from the web server)...
OK, I have another data point. I changed the kickstart config file to use NFS instead of HTTP, and now it works! Go figure... Anyone have any guesses as to what could possibly cause this? Using install method "url --url http://hostname" produces a signal 11, but using "nfs --server=hostname --dir /var/www/html" does not.
Alfred
On Jun 1, 2006, at 10:02, I wrote:
I've resolved this issue, and here is the answer for anyone else who may encounter the same problem later.
So I ask again, can a misconfigured kickstart configuration file cause this problem?
The answer is a resounding yes. The following line in the kickstart configuration file causes the kickstart installation to fail with "installation exited abnormally -- received signal 11":
url --url http://hostname
Adding a slash at the end of the url makes the problem go away. I would say this is a bug in Anaconda. What's the best way to log this bug to make sure it gets fixed by the upstream provider?
Alfred
-
Alfred von Campe -
Henk van Lingen -
Ian mu -
JackHammer19 -
Joshua Gimer -
Rodrigo Barbosa -
Sam Drinkard