Celsius V810 /hdi lost interupt
Hello * Goal: install a Celsius V810 with SLES9 / ( SP1 ) The WS has 8GB Mem, two SATA Disks of different size Problem: Due to high frequent errors: hdi lost interrupt dma_timer_expiry dma status=0x24 the OS does'n boot ok ( about the first runlevel scripts finished after 12 hours ) The WS was booted for testing also in failsafe mode, same Problem A Debian 64 Bit did't even boot ... Anybody any hints for this special Problem ? TIA Micha @Andy: sorry, please excuse my typo when sending the mail
Michael Arndt <M.Arndt@science-computing.de> writes:
Hello *
Goal: install a Celsius V810 with SLES9 / ( SP1 )
I would try ServicePack3 first - it has just been released. Are you using the latest BIOS version? The V810 had some BIOS issues which are fixed now.
The WS has 8GB Mem, two SATA Disks of different size
Does it work with less than 4 GB?
Problem:
Due to high frequent errors:
hdi lost interrupt
dma_timer_expiry dma status=0x24
the OS does'n boot ok ( about the first runlevel scripts finished after 12 hours ) The WS was booted for testing also in failsafe mode, same Problem A Debian 64 Bit did't even boot ...
Anybody any hints for this special Problem ?
Andreas -- Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126
Michael Arndt wrote:
Hello *
Goal: install a Celsius V810 with SLES9 / ( SP1 ) The WS has 8GB Mem, two SATA Disks of different size
Problem:
Due to high frequent errors:
hdi lost interrupt dma_timer_expiry dma status=0x24
the OS does'n boot ok ( about the first runlevel scripts finished after 12 hours ) The WS was booted for testing also in failsafe mode, same Problem A Debian 64 Bit did't even boot ...
Anybody any hints for this special Problem ?
TIA Micha
@Andy: sorry, please excuse my typo when sending the mail
Try booting & installing w/ only a small part of the RAM installed, like < 2 GB. There have been several threads over the years addressing this specific sort of thing, & that often helps. $0.02, no more, no less :-). -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton
hello * When looking more precisely at the Workstation in question i saw that is not the "good old v810" but the new Celsius V 830 experiences until now: "old" bios v1.02-> frequent lost interupts errors, but SLES9 installable in theory ... Only available actual bios 1.05.1818 -> the first suse install splash screen after selecting the installation method hangs ... no meaningful information accessible ... So anybody out there has a 64-bit Linux running on a V 830 ? If so, BIOS Settings would be of interest .. TIA Micha
Hi, * Michael Arndt <M.Arndt@science-computing.de> [060109 15:16]:
hello *
When looking more precisely at the Workstation in question i saw that is not the "good old v810" but the new Celsius V 830
Some more information about the machine would be appreciated.
experiences until now:
[...]
Only available actual bios 1.05.1818 -> the first suse install splash screen after selecting the installation method hangs ... no meaningful information accessible ...
What can be seen if the bootsplash is switched off? I suspect some broken memory configuration, so the BIOS configuration would be good as well.
TIA Micha
Stefan -- SUSE LINUX Products GmbH, Maxfeldstr. 5 Mail: sf@suse.de D-90409 Nuernberg Phone: +49-911-740 53 - 0 GPG 1024D/91614BBC B226 E3DA 37B0 2170 7403 D19C 18AF E579 9161 4BBC
Hi, We have got a Fujitsu-Siemens Celsius V830 runnung under SuSE 10. When running the Gaussian application it crashes at link changes (forking a new executable but retaining the same PID). The error cannot be reproduced, it may or may not occur and not in a specific place in the application. It seems it is related to a kernel bug. The machine details are: Fujitsu-Siemens Celsius V830 ----Gigabyte GA-2CEWH ----NVIDIA GeForce 6200 TC 128MB Dual DVI-I ----DVD-ROM 16x48x ATAPI ----2xOpteron 275 2.2GHz (dual core) ----8x2GB DDR333 rg ECC (Infineon CL2.5 ECC Reg) ----2xHDD SATA II 250GB 7.2k (WD Caviar, 2500) BIOS is 1.05.1818, NUMA enabled, but the behavior is the same when NUMA is disabled in BIOS. Sometimes, but not always, the following type of message appears in the syslog: Kernel BUG at "fs/exec.c":788 invalid operand: 0000 [1] SMP CPU 1 Modules linked in: ipt_pkttype ipt_LOG ipt_limit cpufreq_ondemand cpufreq_usersp ace cpufreq_powersave powernow_k8 freq_table snd_pcm_oss snd_mixer_oss button sn d_seq battery snd_seq_device ac nvidia af_packet edd ide_cd cdrom snd_intel8x0 s nd_ac97_codec tg3 snd_ac97_bus snd_pcm pci_hotplug ohci1394 ieee1394 snd_timer f orcedeth ohci_hcd snd ehci_hcd usbcore i2c_nforce2 i2c_core soundcore generic sn d_page_alloc parport_pc lp parport ip6t_REJECT ipt_REJECT ipt_state iptable_mang le iptable_nat iptable_filter ip6table_mangle ip_conntrack ip_tables ip6table_fi lter ip6_tables ipv6 dm_mod raid0 reiserfs fan thermal processor sg sata_nv liba ta amd74xx sd_mod scsi_mod ide_disk ide_core Pid: 10403, comm: l502.exe Tainted: P U 2.6.13-15.7-smp RIP: 0010:[<ffffffff80198368>] <ffffffff80198368>{flush_old_exec+1608} RSP: 0018:ffff810438b29ca8 EFLAGS: 00010283 RAX: ffff81007fe2d7d0 RBX: 0000000000000001 RCX: ffff81007fe2d640 RDX: ffff81007fe2d640 RSI: ffff810258a98680 RDI: ffff810037d77ac0 RBP: ffff810440e0e840 R08: ffff810438b28000 R09: 0000000000000001 R10: 000000000000003f R11: ffffffff804861c8 R12: 0000000000000003 R13: ffff810258af6100 R14: ffff810258af6908 R15: ffff810258b52200 FS: 00002aaab1db20a0(0000) GS:ffffffff8050e880(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000508e30 CR3: 000000042de73000 CR4: 00000000000006e0 Process l502.exe (pid: 10403, threadinfo ffff810438b28000, task ffff81007fe2d640) Stack: ffff81044fd3cad0 ffff810258a98680 0000000000000080 0000000000000001 ffff81044e71b1f0 ffff810258b52200 000000009fd0f8c0 ffffffff8018c876 ffff810000000000 ffff810258b52200 Call Trace:<ffffffff8018c876>{vfs_read+342} <ffffffff801bb9e6>{load_elf_binary+1254} <ffffffff8016be16>{buffered_rmqueue+534} <ffffffff801bb500>{load_elf_binary+0} <ffffffff80198903>{search_binary_handler+195} <ffffffff80198ca3>{do_execve+467} <ffffffff8010ed7e>{system_call+126} <ffffffff8010cfb4>{sys_execve+68} <ffffffff8010f19e>{stub_execve+106} Code: 0f 0b a3 c9 06 38 80 ff ff ff ff c2 14 03 65 48 8b 04 25 00 RIP <ffffffff80198368>{flush_old_exec+1608} RSP <ffff810438b29ca8> Any help is welcome. Best wishes, Ödön -- ------------------------------------- Odon Farkas Department of Organic Chemistry, Eotvos Lorand University, P.O.Box 32 Budapest 112.,H-1518 E-mail: farkas@chem.elte.hu URL: http://organ.elte.hu/farkas Phone: (36)-(1)-209-0555, ext. 1325 (36)-(1)-372-2570 (36)-(30)-255-3111 FAX: (36)-(1)-209-0602 (36)-(1)-372-2620 -------------------------------------
On Wednesday 18 January 2006 09:33, Odon Farkas wrote:
Kernel BUG at "fs/exec.c":788
The BUG is wrong and can just be removed because its test is by itself racy. I can do that for the next update kernel. Don't know when that will happen though. You might be able to get an experimental untested kernel rpm with this fix earlier (in a day or two) from ftp://ftp.suse.com/pub/projects/kernel/kotd/10.0-x86_64/ -Andi
Dear Andi, It seems that commenting out line 788 in fs/exec.c and recompiling the kernel solved the problem. Best wishes, Odon Ödön Farkas Research associate professor Deparment of Organic Chemistry, Institute of Chemistry, Eötvös Loránd University, Budapest Address: 1/A Pázmány Péter sétány, H-117 Budapest, Hungary Phone: +36-1-372-2570 Cell phone: +36-30-255-3111 Fax: +36-1-372-2620 URL: http://organ.elte.hu/farkas -----Original Message----- From: Andi Kleen [mailto:ak@suse.de] Sent: Thursday, January 19, 2006 6:27 AM To: suse-amd64@suse.com Cc: Odon Farkas Subject: Re: [suse-amd64] Celsius V830 Kernel BUG at "fs/exec.c":788 On Wednesday 18 January 2006 09:33, Odon Farkas wrote:
Kernel BUG at "fs/exec.c":788
The BUG is wrong and can just be removed because its test is by itself racy. I can do that for the next update kernel. Don't know when that will happen though. You might be able to get an experimental untested kernel rpm with this fix earlier (in a day or two) from ftp://ftp.suse.com/pub/projects/kernel/kotd/10.0-x86_64/ -Andi
Hi, We have got a Fujitsu-Siemens Celsius V830 runnung under SuSE 10. When running the Gaussian application it crashes at link changes (forking a new executable but retaining the same PID). The error cannot be reproduced, it may or may not occur and not in a specific place in the application. It seems it is related to a kernel bug. The machine details are: Fujitsu-Siemens Celsius V830 ----Gigabyte GA-2CEWH ----NVIDIA GeForce 6200 TC 128MB Dual DVI-I ----DVD-ROM 16x48x ATAPI ----2xOpteron 275 2.2GHz (dual core) ----8x2GB DDR333 rg ECC (Infineon CL2.5 ECC Reg) ----2xHDD SATA II 250GB 7.2k (WD Caviar, 2500) BIOS is 1.05.1818, NUMA enabled, but the behavior is the same when NUMA is disabled in BIOS. Sometimes, but not always, the following type of message appears in the syslog: Kernel BUG at "fs/exec.c":788 invalid operand: 0000 [1] SMP CPU 1 Modules linked in: ipt_pkttype ipt_LOG ipt_limit cpufreq_ondemand cpufreq_usersp ace cpufreq_powersave powernow_k8 freq_table snd_pcm_oss snd_mixer_oss button sn d_seq battery snd_seq_device ac nvidia af_packet edd ide_cd cdrom snd_intel8x0 s nd_ac97_codec tg3 snd_ac97_bus snd_pcm pci_hotplug ohci1394 ieee1394 snd_timer f orcedeth ohci_hcd snd ehci_hcd usbcore i2c_nforce2 i2c_core soundcore generic sn d_page_alloc parport_pc lp parport ip6t_REJECT ipt_REJECT ipt_state iptable_mang le iptable_nat iptable_filter ip6table_mangle ip_conntrack ip_tables ip6table_fi lter ip6_tables ipv6 dm_mod raid0 reiserfs fan thermal processor sg sata_nv liba ta amd74xx sd_mod scsi_mod ide_disk ide_core Pid: 10403, comm: l502.exe Tainted: P U 2.6.13-15.7-smp RIP: 0010:[<ffffffff80198368>] <ffffffff80198368>{flush_old_exec+1608} RSP: 0018:ffff810438b29ca8 EFLAGS: 00010283 RAX: ffff81007fe2d7d0 RBX: 0000000000000001 RCX: ffff81007fe2d640 RDX: ffff81007fe2d640 RSI: ffff810258a98680 RDI: ffff810037d77ac0 RBP: ffff810440e0e840 R08: ffff810438b28000 R09: 0000000000000001 R10: 000000000000003f R11: ffffffff804861c8 R12: 0000000000000003 R13: ffff810258af6100 R14: ffff810258af6908 R15: ffff810258b52200 FS: 00002aaab1db20a0(0000) GS:ffffffff8050e880(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000508e30 CR3: 000000042de73000 CR4: 00000000000006e0 Process l502.exe (pid: 10403, threadinfo ffff810438b28000, task ffff81007fe2d640) Stack: ffff81044fd3cad0 ffff810258a98680 0000000000000080 0000000000000001 ffff81044e71b1f0 ffff810258b52200 000000009fd0f8c0 ffffffff8018c876 ffff810000000000 ffff810258b52200 Call Trace:<ffffffff8018c876>{vfs_read+342} <ffffffff801bb9e6>{load_elf_binary+1254} <ffffffff8016be16>{buffered_rmqueue+534} <ffffffff801bb500>{load_elf_binary+0} <ffffffff80198903>{search_binary_handler+195} <ffffffff80198ca3>{do_execve+467} <ffffffff8010ed7e>{system_call+126} <ffffffff8010cfb4>{sys_execve+68} <ffffffff8010f19e>{stub_execve+106} Code: 0f 0b a3 c9 06 38 80 ff ff ff ff c2 14 03 65 48 8b 04 25 00 RIP <ffffffff80198368>{flush_old_exec+1608} RSP <ffff810438b29ca8> Any help is welcome. Best wishes, Ödön -- ------------------------------------- Odon Farkas Department of Organic Chemistry, Eotvos Lorand University, P.O.Box 32 Budapest 112.,H-1518 E-mail: farkas@chem.elte.hu URL: http://organ.elte.hu/farkas Phone: (36)-(1)-209-0555, ext. 1325 (36)-(1)-372-2570 (36)-(30)-255-3111 FAX: (36)-(1)-209-0602 (36)-(1)-372-2620 -------------------------------------
participants (8)
-
Andi Kleen -
Andreas Jaeger -
Farkas Ödön -
Michael Arndt -
Odon Farkas
-
Stefan Fent -
William A. Mahaffey III -
Ödön Farkas