-----Original Message----- From: Bruce Edge Sent: Monday, December 15, 2003 4:11 PM To: 'Keith Owens' Subject: RE: [lkcd-general] This kernel supports KDB but LKCD was invoked directly, not via KDB....
> -----Original Message----- > From: Keith Owens [mailto:[EMAIL PROTECTED] > Sent: Monday, December 15, 2003 3:08 PM > To: Bruce Edge > Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED] > Subject: Re: [lkcd-general] This kernel supports KDB but LKCD was > invoked directly, not via KDB.... > > > On Mon, 15 Dec 2003 12:09:34 -0800, > Bruce Edge <[EMAIL PROTECTED]> wrote: > >This kernel supports KDB but LKCD was invoked directly, not via KDB. > >Falling back to the old and broken LKCD method of getting > data from all > >cpus, > >do not be surprised if LKCD hangs. > > > >when I dump. > > > >What should I do to force the dump through kdb instead of > calling lkcd > >directly? > > How are you requesting the dump in the first place? If you are > manually entering sysrq-c, replace that with enter kdb (control-A or > pause key), then type 'sr c'. I need more details of what steps you > are following. > > When I drop into kdb and do an 'sr c', I do not get that message: Entering kdb (current=0x40400000, pid 0) on processor 0 due to Keyboard Entry [0]kdb> sr c SysRq : Start a Crash Dump (If Configured) Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Warning: get_irqlock on cpu 0 while kdb is running, may hang Nmi Npx0 Dly P92 Sha0 Kbd0 Cmos Pci Dma0 PrtB Tim Exp Rfsh Geom Mem0 Mem1 Ebda Admgr1 Memmg ... box reboots I get it when I load a module that does a divide by zero: divide error: 0000 t3-cssdev t3-hbas-fcpt t3-css t3-hbas-fcpi t3-hbas-fcs t3-hbas-hw t3-fps-top t3-portctlr t3-fps-api t3-fps-ppe t3-dev_sfp t3-dev_voltage t3-dev_therm t3-dev_fan t3-dev_led t3-dev_ps t3-dev_eeprom t3-dev_lm87 t3-dev_wdt t3-dev_ich_smb t3-framectlr t3-cam t3-pasics t3-errmgr t3-hssl t3-ppe t3-voq t3-fmotel t3-egress t3-parser t3-pasicdma t3-pasicgbl t3-kconf t3-klib CPU: 2 EIP: 0010:[<c2d1cedd>] Tainted: P EFLAGS: 00010246 EIP is at k_portctlr_open+0x15/0x1a8 [t3-portctlr] eax: 0000000a ebx: 00000000 ecx: 0000000a edx: 00000000 esi: b8662640 edi: 00000000 ebp: 9e5b5f08 esp: 9e5b5ef0 ds: 0018 es: 0018 ss: 0018 Process servicemgr (pid: 963, stackpage=9e5b5000) Stack: 9e5b4000 b8662640 bfbccd40 00000001 00000001 00000001 9e5b5f24 4013afeb b4125860 b8662640 b8662640 bd496d40 b4125860 9e5b5f48 4017a7e6 b4125860 b8662640 b8662640 b4125860 ffffffe9 403e6280 bd496d48 9e5b5f64 40139bd3 Call Trace: [<4013afeb>] chrdev_open+0x67/0x9c [kernel] [<4017a7e6>] devfs_open+0x126/0x1c4 [kernel] [<40139bd3>] dentry_open+0xdf/0x190 [kernel] [<40139ae7>] filp_open+0x4f/0x5c [kernel] [<40139e2e>] sys_open+0x36/0xa4 [kernel] [<40108d37>] system_call+0x33/0x38 [kernel] Code: f7 fb 89 c1 a1 10 c3 d2 c2 8b 00 83 e0 0f 83 f8 03 76 23 83 This kernel supports KDB but LKCD was invoked directly, not via KDB. <--------------- Here Falling back to the old and broken LKCD method of getting data from all cpus, do not be surprised if LKCD hangs. unexpected IRQ trap at vector da unexpected IRQ trap at vector da unexpected IRQ trap at vector da dump: Dumping to device 0x1602 [ide1(22,2)] on CPU 2 ... dump: Compression value is 0x0, Writing dump header dump: Pass 1: Saving Reserved Pages: dump: Memory Bank[0]: 0 ... 7fffffff: ......... dump: Pass 2: Saving Remaining Referenced Pages: dump: Memory Bank[0]: 0 ... 7fffffff: .......................................................................................................................... dump: Pass 3: Saving Remaining Unreferenced Pages: dump: Memory Bank[0]: 0 ... 7fffffff: ..................................................................................................................................................................................................Nmi Npx0 Dly P92 Sha0 Kbd0 Cmos Pci Dma0 PrtB Tim Exp Rfsh Geom Mem0 Mem1 Ebda Admgr1 ...box reboots I cannot try using a sysrq-c, as the console is a serial port. This is on a two processor box, with hyperthreading enabled on both processors, if that matters. -Bruce Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe.
