Hi Kristof, Your jumpstart question is easier to find experts on this alias: install-discuss at opensolaris.org Regards, Crisson On Wed, 2008-06-04 at 12:13 -0700, kristof wrote: > While early tests all looked fine, I'm now strugling in getting opensolaris > build 89 installed via our jumpstart server. (with patched bge kenel modules) > > Iperf results: 934-941 Mbits/second , both in 32&64 bit mode, both vlan/non > vlan traffic > > But Now I'm having a couple of problems: > > 1. When I replace the 32bit bge driver in the miniroot with the patched one, > and after having updated /etc/driver_aliases in the miniroot to associate bge > with pci14e4,166a > > -> jumpstart is dog slow, using snoop I see continious nfs retransmits, and > install hangs after a few hourss (at 60 % of unpacking the flar), while > normally jumpstart only takes +/- 15min > > 2. When I use my old build 89 miniroot (the one where I added the bcme > driver), jumpstart install is just fine, but, When I try to do the following > via the finish script: > > check if 14e4,166a is found via prtconf > if yes, mv /kernel/drv/bge /kernel/drv/bge_ori > mv /kernel/drv/amd64/bge /kernel/drv/amd64/bge_ori > copy new files > modify /etc/driver_aliases > touch /reconfigure > rm /etc/dhcp.bcme (because nic is been detected by bcme driver as bcme) > touch /etc/dhcp.bge > > After the jumpstart the system panics at least 1 time on 2 boots. > > I'm not sure if the problem is related to the driver, or maybe I'm doing > something wrong in test 1 & test 2 > > Do I have to do something else, when I replace the bge driver in the miniroot? > > The panic is causing a crash dump: which output always look the same > > ]# mdb -k unix.0 vmcore.0 > Loading modules: [ unix genunix specfs dtrace cpu.generic uppc pcplusmp > scsi_vhci ufs ip hook neti sctp arp usba fctl nca md lofs zfs cpc random > crypto fcip nfs ptm nsctl ] > > ::status > debugging crash dump vmcore.0 (64-bit) from halfRack-appliance > operating system: 5.11 snv_89 (i86pc) > panic message: BAD TRAP: type=e (#pf Page fault) rp=ffffff000fff5b80 addr=a > occurred in module "tmpfs" due to a NULL pointer dereference > dump content: kernel pages only > > $c > tdirtrunc+0x3a(ffffff02d4504b68) > tmp_unmount+0x107(ffffff02d5100c88, 0, ffffff02d4765d88) > fsop_unmount+0x1e(ffffff02d5100c88, 0, ffffff02d4765d88) > dounmount+0x5c(ffffff02d5100c88, 0, ffffff02d4765d88) > vfs_unmountall+0x95() > kadmin+0x4ab(2, 1, 0, ffffff02d4765d88) > uadmin+0xb7(2, 1, 0) > sys_syscall32+0x101() > > $? > %rax = 0x0000000000000004 %r9 = 0xffffff02dab18b00 > %rbx = 0x0000000000000000 %r10 = 0x0000000000000001 > %rcx = 0x0000000000000005 %r11 = 0xffffff000fff5a70 > %rdx = 0xffffff02d5784800 %r12 = 0x0000000000000051 > %rsi = 0xffffffffffffffaf %r13 = 0xffffff02d4504b68 > %rdi = 0xfffffffffbd010e0 tmp_kmemspace %r14 = 0x0000000000000002 > %r8 = 0x0000000000000004 %r15 = 0xffffffffc005d338 > bge_priv_prop > > %rip = 0xfffffffff78c685a tdirtrunc+0x3a > %rbp = 0xffffff000fff5cc0 > %rsp = 0xffffff000fff5c70 > %rflags = 0x00010202 > id=0 vip=0 vif=0 ac=0 vm=0 rf=1 nt=0 iopl=0x0 > status=<of,df,IF,tf,sf,zf,af,pf,cf> > > %cs = 0x0030 %ds = 0x004b %es = 0x004b > %trapno = 0xe %fs = 0x0000 %gs = 0x01c3 > %err = 0x0 > > > > > This message posted from opensolaris.org > _______________________________________________ > networking-discuss mailing list > networking-discuss at opensolaris.org -- ************************************* * Guang-Hao, Crisson Hu * * China ERI, Sun Microsystems, Inc. * * Tel: +86-10-62673095 * *************************************
[install-discuss] [networking-discuss] bcme not working with clearview
guanghao hu - Sun Microsystems - Beijing China Thu, 05 Jun 2008 15:10:06 +0800
