All times are UTC-06:00




Post new topic  Reply to topic  [ 4 posts ] 
Author Message
PostPosted: Tue Oct 02, 2007 12:27 pm 
Offline

Joined: Tue Dec 26, 2006 5:13 pm
Posts: 37
Hi,

I'm using Debian 4.0r0 kernel 2.6.18.
Whenever gigabit (marvell) is stressed a lot ( with netcat or ftp), I get about 30Mb/s, which is nice.

Unfortunately, the system locks or even reboots after a few minutes.
I experienced the same prob on Suse10.2 and later, but not on Suse10.1.
Is there hope on later kernels or is the bug still present ?

( Note: I'm sure, it's no hardware problem, as the machine performs perfect over hours with suse10.1.. and all of my 3 Peg2's behave the same.)


Top
   
 Post subject:
PostPosted: Tue Oct 02, 2007 3:05 pm 
Offline

Joined: Sun May 08, 2005 8:46 pm
Posts: 559
Location: Paris
i had quite the same issue having 1,5 Gb of Ram in the Pegasos2. Removing 512Mb fixed it.
But i suppose you're on a different story.


Top
   
 Post subject:
PostPosted: Sun Oct 07, 2007 8:47 am 
Offline

Joined: Tue Dec 26, 2006 5:13 pm
Posts: 37
It's a kernel bug. I experienced the crash on Suse10.2, but not on Suse 10.1. (But also on Debian, Ubuntu and some others.)

As Suse 10.3 with it's very new kernel seems to work perfectly with gigabit again, I'll try 2.6.22 on debian to try whether it's fixed there, too.


Top
   
PostPosted: Mon Oct 08, 2007 8:22 am 
Offline

Joined: Mon Jan 30, 2006 7:24 am
Posts: 43
Location: Budapest, Hungary
Quote:
Unfortunately, the system locks or even reboots after a few minutes. I experienced the same prob on Suse10.2 and later, but not on Suse10.1.
Is there hope on later kernels or is the bug still present ?
I had the same on my Peg2 with 1GB RAM, running Debian Etch. I upgraded to 2.6.21.4, and that fixed it. But 2.6.18 kernel was highly unstable for me, even without the use of the GBit eth. I changed, and all problems gone:
Code:
charlie@stronghold:~$ uname -a
Linux stronghold 2.6.21.4 #3 Mon Jul 2 13:45:38 CEST 2007 ppc GNU/Linux
charlie@stronghold:~$ uptime
16:29:02 up 29 days, 2:32, 11 users, load average: 0.06, 0.06, 0.01
Previous record uptime with 2.6.18 (stock Debian 4.0r0 kernel) was 4 days:
Code:
charlie@stronghold:~$ uprecords | grep 2.6.21 | head -1
1 40 days, 18:46:28 | Linux 2.6.21.4 Thu Jul 26 18:54:27 2007
charlie@stronghold:~$ uprecords | grep 2.6.18 | head -1
6 4 days, 07:21:10 | Linux 2.6.18-4-powerpc Sat Jun 2 02:04:48 2007
... and usually it crashed within one day:
Code:
charlie@stronghold:~$ uprecords -m 10000 -B | grep 2.6.18 | grep "0 days" | head -5
14 0 days, 05:05:50 | Linux 2.6.18-4-powerpc Sun Jun 24 04:50:09 2007
15 0 days, 07:09:16 | Linux 2.6.18-4-powerpc Sat Jun 23 21:37:11 2007
16 0 days, 02:34:38 | Linux 2.6.18-4-powerpc Sat Jun 23 18:58:23 2007
17 0 days, 00:52:28 | Linux 2.6.18-4-powerpc Sat Jun 23 17:13:59 2007
20 0 days, 00:30:16 | Linux 2.6.18-4-powerpc Tue Jun 19 19:29:40 2007
Sarge was totally stable on the same machine. So it seems like the some kernel versions (most probably versions between 2.6.15 and 2.6.20?) has serious stability issues on a Pegasos 2...


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 4 posts ] 

All times are UTC-06:00


Who is online

Users browsing this forum: No registered users and 5 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
PowerDeveloper.org: Copyright © 2004-2012, Genesi USA, Inc. The Power Architecture and Power.org wordmarks and the Power and Power.org logos and related marks are trademarks and service marks licensed by Power.org.
All other names and trademarks used are property of their respective owners. Privacy Policy
Powered by phpBB® Forum Software © phpBB Group