[Cooker] Enterprise kernel and CPU Usage in dual machine
Hello list, using RC2 kernel-enterprise in a dual processor machine, a companion complained about CPU usage (Look at top output attachment). He said it was not normal that too much memory was shared between user and system in one CPU. He argued that in other systems he had worked with (RedHat Linux and Sun) using the same programs, he did not have this behaviour. I do not have any idea about. Is this a bug? Is this normal? Should I use another kernel instead? Thanks. -- Toni Hermoso Pulido -- web: http://www.arrakis.es/~toniher http://drac.homelinux.org email: [EMAIL PROTECTED] [EMAIL PROTECTED] pgp_public_key: http://www.arrakis.es/~toniher/toniher.pgp *** Cífratelo, cífraselo *** -- Tasks: 53 total, 3 running, 50 sleeping, 0 stopped, 0 zombie Cpu0 : 56.2% user, 43.8% system, 0.0% nice, 0.0% idle Cpu1 : 100.0% user, 0.0% system, 0.0% nice, 0.0% idle Mem: 1031796k total, 600356k used, 431440k free,27472k buffers Swap: 6297440k total,0k used, 6297440k free, 378428k cached PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 2207 admin 14 0 20620 20m 2380 R 99.9 2.0 0:35.09 promd_opt_pc2.6 2209 admin 14 0 20620 20m 2380 R 94.0 2.0 0:34.77 promd_opt_pc2.6 2259 root 9 0 988 988 792 R 6.3 0.1 0:00.16 top 1 root 8 0 500 496 448 S 0.0 0.0 0:04.37 init 2 root 8 0 000 S 0.0 0.0 0:00.00 keventd 3 root 18 19 000 S 0.0 0.0 0:00.00 ksoftirqd_CPU0 4 root 19 19 000 S 0.0 0.0 0:00.01 ksoftirqd_CPU1 5 root 9 0 000 S 0.0 0.0 0:00.00 kswapd 6 root 9 0 000 S 0.0 0.0 0:00.00 bdflush 7 root 9 0 000 S 0.0 0.0 0:00.18 kupdated 8 root 9 0 000 S 0.0 0.0 0:00.00 kinoded 9 root -1 -20 000 S 0.0 0.0 0:00.00 mdrecoveryd 13 root 9 0 000 S 0.0 0.0 0:00.11 ahd_dv_0 14 root 9 0 000 S 0.0 0.0 0:00.00 ahd_dv_1 15 root 9 0 000 S 0.0 0.0 0:00.00 scsi_eh_0 16 root 9 0 000 S 0.0 0.0 0:00.00 scsi_eh_1 20 root 9 0 000 S 0.0 0.0 0:00.02 kjournald 129 root 8 0 1224 1224 752 S 0.0 0.1 0:00.16 devfsd 214 root 9 0 000 S 0.0 0.0 0:00.00 khubd 332 root 9 0 000 S 0.0 0.0 0:00.00 kjournald 333 root 9 0 000 S 0.0 0.0 0:00.14 kjournald 343 root 9 0 000 S 0.0 0.0 0:00.30 kjournald 742 rpc9 0 548 548 476 S 0.0 0.1 0:00.00 portmap 756 root 9 0 612 612 512 S 0.0 0.1 0:00.08 syslogd 764 root 9 0 1364 1360 444 S 0.0 0.1 0:00.05 klogd 807 root 9 0 696 696 624 S 0.0 0.1 0:00.00 rpc.statd 850 root 9 0 000 S 0.0 0.0 0:00.44 rpciod 851 root 9 0 000 S 0.0 0.0 0:00.00 lockd 877 daemon 9 0 508 508 452 S 0.0 0.0 0:00.00 atd 892 root 9 0 1088 1084 1068 S 0.0 0.1 0:00.01 saslauthd 893 root 9 0 1088 1084 1068 S 0.0 0.1 0:00.00 saslauthd 894 root 9 0 1088 1084 1068 S 0.0 0.1 0:00.00 saslauthd 895 root 9 0 1088 1084 1068 S 0.0 0.1 0:00.00 saslauthd 896 root 9 0 1088 1084 1068 S 0.0 0.1 0:00.00 saslauthd 915 root 9 0 1400 1400 1276 S 0.0 0.1 0:00.15 sshd 1057 root 9 0 1316 1316 1076 S 0.0 0.1 0:00.06 master 1075 postfix9 0 1420 1420 1172 S 0.0 0.1 0:00.03 nqmgr 1118 root 8 0 596 596 532 S 0.0 0.1 0:00.01 crond 1189 root 9 0 416 412 372 S 0.0 0.0 0:00.00 mingetty 1190 root 9 0 416 412 372 S 0.0 0.0 0:00.00 mingetty 1191 root 9 0 416 412 372 S 0.0 0.0 0:00.00 mingetty 1192 root 9 0 416 412 372 S 0.0 0.0 0:00.00 mingetty 1193 root 9 0 416 412 372 S 0.0 0.0 0:00.00 mingetty 1194 root 9 0 416 412 372 S 0.0 0.0 0:00.00 mingetty 1475 root 9 0 1944 1944 1684 S 0.0 0.2 0:01.20 sshd 1477 root 9 0 2016 2016 1148 S 0.0 0.2 0:00.07 bash 2017 postfix9 0 1388 1388 1152 S 0.0 0.1 0:00.01 pickup 2089 root 9 0 972 972 812 S 0.0 0.1 0:00.00 su 2090 admin 9 0 1816 1816 1080 S 0.0 0.2 0:00.02 tcsh 2162 admin 9 0 1712 1712 980 S 0.0 0.2 0:00.03 jmd_HB_sg1.csh 2208 admin 9 0 20620 20m 2380 S 0.0 2.0 0:00.00 promd_opt_pc2.6 2210 root 9 0 1944 1944 1684 S 0.0 0.2 0:00.04 sshd 2212 root 9 0 1976 1976 1120 S 0.0 0.2 0:00.10 bash signature.asc Description: Esta parte del mensaje =?ISO-8859-1?Q?est=E1?= f
Re: [Cooker] enterprise kernel
From: "Ray Carlino" <[EMAIL PROTECTED]> > I just bought 8 new Dell servers and need to put Mandrake 9.1 or 9.2 on > them .. > The problem is they have 6GB Ram and the enterprise kernel is only built > for 4GB. > So as a suggestion can you start building the kernel with 64GB support? > And second is there a config file of the enterprise kernel around that I > can use to rebuild with 64GB support? I want all of the same options > that are built in now but more memory.. > /boot/config*enterprise or /usr/src/linux/arch/i386/defconfig-enterprise Thomas
[Cooker] enterprise kernel
I just bought 8 new Dell servers and need to put Mandrake 9.1 or 9.2 on them .. The problem is they have 6GB Ram and the enterprise kernel is only built for 4GB. So as a suggestion can you start building the kernel with 64GB support? And second is there a config file of the enterprise kernel around that I can use to rebuild with 64GB support? I want all of the same options that are built in now but more memory.. Thanks -- Ray Carlino <[EMAIL PROTECTED]>
Re: [Cooker] Enterprise kernel and memory over 960Meg
Robert Fox wrote: I just got a new (old) machine with a P4 1.5Ghz and I bumped the memory to 1.25 Gig because 512M stick are so cheap (PC133 SD-RAM) This is the first time I've dealt with a machine with over 768M memory and I noticed that it only recognizes 960M of the 1.25Gig I know that the Enterprise kernel supports the larger memory - but I have heard that there are stability problems with it. I have seen recommendations to stick with the standard Mandrake Kernel and forget the memory over the 960M mark. Could someone explain WHY does one need the enterprise kernel (with additional security features) for large memory? Where can I get a detailed list of differences between the regular Mandrake kernel and the Enterprise version? Will the stock Mandrake kernel eventually support more memory? I know Linux handles memory better than Winblows, but WinXP has no problem seeing the 1.25G . . . . Many thanks in advance, R.Fox Hi, enterprise is not needed for higmem, I have lot of machines with 1,5-4 G RAM (mdk or rh) with "custom" kernels. Custom = standard distribution kernel recompiled with higmem support. Only bad thing WAS speed penalty of higmem, but with 2.4.20 is all OK (with IO patch enabled). Jiri Cerny
Re: [Cooker] Enterprise kernel and memory over 960Meg
IIRC the problems were due to an error in the xfs filesystem in combination with high memory. I made a patched version which should correct it, but did not post in on mandrake club, because nobody requested/voted for it. If anyone is experiencing problems, I can put it up somewhere. (alternatively, download the 2.4.19-16.7mdk src rpm from club and change the spec so that it builds the enterprise version only and build it. Danny On Tue, 14 Jan 2003, Chuck Shirley wrote: > On Tuesday 14 January 2003 03:45, Robert Fox wrote: > >I know that the Enterprise kernel supports the larger memory - but I > >have heard that there are stability problems with it. I have seen > >recommendations to stick with the standard Mandrake Kernel and forget > >the memory over the 960M mark. > > > I have only used the enterprise kernels on my system, and have had > no problems with kernel stability on my dual P-III 750 + 1024 MB > machine. > > > --C.S. >
Re: [Cooker] Enterprise kernel and memory over 960Meg
On Tuesday 14 January 2003 03:45, Robert Fox wrote: >I know that the Enterprise kernel supports the larger memory - but I >have heard that there are stability problems with it. I have seen >recommendations to stick with the standard Mandrake Kernel and forget >the memory over the 960M mark. > I have only used the enterprise kernels on my system, and have had no problems with kernel stability on my dual P-III 750 + 1024 MB machine. --C.S.
Re: [Cooker] Enterprise kernel and memory over 960Meg
hi: i have many problems with enterprised kernel. (highmem enabled). sometimes some kernel version maybe ok with ur machine , if u r lucky!! > I just got a new (old) machine with a P4 1.5Ghz and I bumped the memory > to 1.25 Gig because 512M stick are so cheap (PC133 SD-RAM) > > This is the first time I've dealt with a machine with over 768M memory > and I noticed that it only recognizes 960M of the 1.25Gig > > I know that the Enterprise kernel supports the larger memory - but I > have heard that there are stability problems with it. I have seen > recommendations to stick with the standard Mandrake Kernel and forget > the memory over the 960M mark. > > Could someone explain WHY does one need the enterprise kernel (with > additional security features) for large memory? Where can I get a > detailed list of differences between the regular Mandrake kernel and > the Enterprise version? Will the stock Mandrake kernel eventually > support more memory? > > I know Linux handles memory better than Winblows, but WinXP has no > problem seeing the 1.25G . . . . > > Many thanks in advance, > R.Fox
[Cooker] Enterprise kernel and memory over 960Meg
I just got a new (old) machine with a P4 1.5Ghz and I bumped the memory to 1.25 Gig because 512M stick are so cheap (PC133 SD-RAM) This is the first time I've dealt with a machine with over 768M memory and I noticed that it only recognizes 960M of the 1.25Gig I know that the Enterprise kernel supports the larger memory - but I have heard that there are stability problems with it. I have seen recommendations to stick with the standard Mandrake Kernel and forget the memory over the 960M mark. Could someone explain WHY does one need the enterprise kernel (with additional security features) for large memory? Where can I get a detailed list of differences between the regular Mandrake kernel and the Enterprise version? Will the stock Mandrake kernel eventually support more memory? I know Linux handles memory better than Winblows, but WinXP has no problem seeing the 1.25G . . . . Many thanks in advance, R.Fox
Re: [Cooker] enterprise kernel lockup - with trace
Brent Hasty wrote: On Friday 08 November 2002 03:53 pm, Bryan Whitehead wrote: what is the key sequence to show "Memory" "ShowReg" "Show State"? huey.jpl.nasa.gov/~driver/crash.txt This has the full kernel boot sequence upto and including the hard lock. After the machine locked up I Alt-SysRq'ed "Memory" "ShowReg" and "Show State". Sync will not work, it will just be stuck. This is on a machine running 2.4.19-16mdkenterprise with 2GB of RAM. Memtest shows all memory is good. This is the best I can do for logs... If this can't find the problem Me and Todd tried many differnt options being passsed to the kernel... and nothing worked. Any idea's? Hit Alt-SysRq-H for the help menu at a Linux console. -- Bryan Whitehead SysAdmin - JPL - Interferometry Systems and Technology Phone: 818 354 2903 [EMAIL PROTECTED]
Re: [Cooker] enterprise kernel lockup - with trace
On Friday 08 November 2002 03:53 pm, Bryan Whitehead wrote: what is the key sequence to show "Memory" "ShowReg" "Show State"? > huey.jpl.nasa.gov/~driver/crash.txt > > This has the full kernel boot sequence upto and including the hard lock. > > After the machine locked up I Alt-SysRq'ed "Memory" "ShowReg" and "Show > State". > > Sync will not work, it will just be stuck. > > This is on a machine running 2.4.19-16mdkenterprise with 2GB of RAM. > Memtest shows all memory is good. > > This is the best I can do for logs... If this can't find the problem > > Me and Todd tried many differnt options being passsed to the kernel... > and nothing worked. > > Any idea's?
[Cooker] enterprise kernel lockup - with trace
huey.jpl.nasa.gov/~driver/crash.txt This has the full kernel boot sequence upto and including the hard lock. After the machine locked up I Alt-SysRq'ed "Memory" "ShowReg" and "Show State". Sync will not work, it will just be stuck. This is on a machine running 2.4.19-16mdkenterprise with 2GB of RAM. Memtest shows all memory is good. This is the best I can do for logs... If this can't find the problem Me and Todd tried many differnt options being passsed to the kernel... and nothing worked. Any idea's? -- Bryan Whitehead SysAdmin - JPL - Interferometry Systems and Technology Phone: 818 354 2903 [EMAIL PROTECTED]
[Cooker] Enterprise kernel
Just a note that you must use Xfree 4.0x and not 3.x with this kernel and 2gb of ram. Always get a can't read /dev/mem failure error -- Ray