Soft lockup (and reboot): IPMI
Hello!
While playing with "openipmigui", the server just rebooted with the following last message:
BUG: soft lockup detected on CPU#0! <c0103416> show_trace+0xd/0xf <c01034e5> dump_stack+0x15/0x17 <c01382b0> softlockup_tick+0x9d/0xae <c012190a> run_local_timers+0x12/0x14 <c0121748> update_process_times+0x3c/0x61 <c010e1d9> smp_apic_timer_interrupt+0x57/0x5f <c010307c> apic_timer_interrupt+0x1c/0x24 <f8aa8783> ipmi_thread+0x43/0x71 [ipmi_si] <c0129e62> kthread+0x78/0xa0 <c0100e31> kernel_thread_helper+0x5/0xb
Any idea on what went wrong?
BYtE
Philipp
Signature · 5 lines
Dipl.-Inform. [email protected] Abteilung Systemsoftware und verteilte Systeme, Fk. II Carl von Ossietzky Universitaet Oldenburg, 26111 Oldenburg, Germany http://www.svs.informatik.uni-oldenburg.de/contact/pmhahn/ Telefon: +49 441 798-2866 Telefax: +49 441 798-2756
Quoted 17 lines · Philipp Matthias Hahn
Hello!
While playing with "openipmigui", the server just rebooted with the following last message:
BUG: soft lockup detected on CPU#0! <c0103416> show_trace+0xd/0xf <c01034e5> dump_stack+0x15/0x17 <c01382b0> softlockup_tick+0x9d/0xae <c012190a> run_local_timers+0x12/0x14 <c0121748> update_process_times+0x3c/0x61 <c010e1d9> smp_apic_timer_interrupt+0x57/0x5f <c010307c> apic_timer_interrupt+0x1c/0x24 <f8aa8783> ipmi_thread+0x43/0x71 [ipmi_si] <c0129e62> kthread+0x78/0xa0 <c0100e31> kernel_thread_helper+0x5/0xb
Quoted 1 line
Any idea on what went wrong?
What kernel please? If 2.6.17 or higher, this is new. If < 2.6.17, this is most likely due to the udelay(1) call in ipmi_thread() which in 2.6.17 was replaced with a schedule().
Thanks,
Matt
Signature · 4 lines
Matt Domsch Software Architect Dell Linux Solutions linux.dell.com & www.dell.com/linux Linux on Dell mailing lists @ http://lists.us.dell.com
Quoted 17 lines · Matt Domsch
Quoted 15 lines · Philipp Matthias Hahn
While playing with "openipmigui", the server just rebooted with the following last message:
BUG: soft lockup detected on CPU#0! <c0103416> show_trace+0xd/0xf <c01034e5> dump_stack+0x15/0x17 <c01382b0> softlockup_tick+0x9d/0xae <c012190a> run_local_timers+0x12/0x14 <c0121748> update_process_times+0x3c/0x61 <c010e1d9> smp_apic_timer_interrupt+0x57/0x5f <c010307c> apic_timer_interrupt+0x1c/0x24 <f8aa8783> ipmi_thread+0x43/0x71 [ipmi_si] <c0129e62> kthread+0x78/0xa0 <c0100e31> kernel_thread_helper+0x5/0xb
Quoted 5 lines
Quoted 1 line
Any idea on what went wrong?
What kernel please? If 2.6.17 or higher, this is new. If < 2.6.17, this is most likely due to the udelay(1) call in
ipmi_thread()which in 2.6.17 was replaced with aschedule().
Sorry for omitting that information, it's a plain 2.6.17.8 kernel.
model name : Intel(R) Xeon(TM) CPU 2.66GHz
ipmi message handler version 39.0
IPMI Watchdog: driver initialized
ipmi device interface
IPMI System Interface driver.ipmi_si: Trying SMBIOS-specified KCS state machine at I/O address 0xca2, slave address 0x24, irq 0
ipmi: Found new BMC (man_id: 0x002880, prod_id: 0x0000, dev_id: 0x00)
IPMI KCS interface initialized
BYtE
Philipp
Signature · 5 lines
Dipl.-Inform. [email protected] Abteilung Systemsoftware und verteilte Systeme, Fk. II Carl von Ossietzky Universitaet Oldenburg, 26111 Oldenburg, Germany http://www.svs.informatik.uni-oldenburg.de/contact/pmhahn/ Telefon: +49 441 798-2866 Telefax: +49 441 798-2756
Quoted 31 lines · Philipp Matthias Hahn
Quoted 18 lines · Matt Domsch
Quoted 15 lines · Philipp Matthias Hahn
While playing with "openipmigui", the server just rebooted with the following last message:
BUG: soft lockup detected on CPU#0! <c0103416> show_trace+0xd/0xf <c01034e5> dump_stack+0x15/0x17 <c01382b0> softlockup_tick+0x9d/0xae <c012190a> run_local_timers+0x12/0x14 <c0121748> update_process_times+0x3c/0x61 <c010e1d9> smp_apic_timer_interrupt+0x57/0x5f <c010307c> apic_timer_interrupt+0x1c/0x24 <f8aa8783> ipmi_thread+0x43/0x71 [ipmi_si] <c0129e62> kthread+0x78/0xa0 <c0100e31> kernel_thread_helper+0x5/0xbQuoted 6 lines
Quoted 2 lines
Any idea on what went wrong?
What kernel please? If 2.6.17 or higher, this is new. If < 2.6.17, this is most likely due to the udelay(1) call in
ipmi_thread()which in 2.6.17 was replaced with aschedule().Sorry for omitting that information, it's a plain 2.6.17.8 kernel.
model name : Intel(R) Xeon(TM) CPU 2.66GHz
Quoted 9 lines
ipmi message handler version 39.0
IPMI Watchdog: driver initialized
ipmi device interface
IPMI System Interface driver.ipmi_si: Trying SMBIOS-specified KCS state machine at I/O address0xca2, slave address0x24, irq 0
ipmi: Found new BMC (man_id:0x002880,prod_id:0x0000,dev_id:0x00)
IPMI KCS interface initialized
The only way I can see this would happen is if the BMC is misbehaving somehow, like leaving the "ready" bit set for a long time (not clearing it when it is supposed to) and always signalling that there is some event ready. I guess if the BMC was so fast that it was alway ready that could cause this problem, but that doesn't really seem very likely as userland would have to run to feed it to do that.
This is apparently a Fujitsu device, right? The fact that the product id is 0x0000 leads me to believe that this device is not completely cooked yet.
-Corey