Accesses to certain files on an NFS mounted file system make the system panic with a page fault in kernel mode. The file system is accessed through an AFS-NFS translator, AFS accesses from other systems work, NFS accesses from a SunOS 4.1.3 system hang, but can be interrupted. (There seems to be something wrong with that file system, accesses to files owned by one particular UID fail. But user mode accesses shouldn't crash the kernel ...) Panic messages: Fatal trap 12: page fault while in kernel mode fault virtual address = 0x28 fault code = supervisor read, page not present instruction pointer = 0x8:0xf0123eb2 code segment = base rx0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 877 (tar) interrupt mask = panic: page fault Fatal trap 12: page fault while in kernel mode fault virtual address = 0x28 fault code = supervisor read, page not present instruction pointer = 0x8:0xf0123eb2 code segment = base rx0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 152 (tar) interrupt mask = panic: page fault Fix: None I know of ... How-To-Repeat: mount rs2.rrz.uni-koeln.de:/afs/rrz/pub /ftp cat /ftp/tex/graphics/bm2font/* (Please don't NFS mount that file system, except for testing ...) The instruction pointer indicates a problem in vn_open(), it starts at 0xf0123aac and ends before 0xf0123f30 in my /kernel ...
State Changed From-To: open->feedback Stefan, please verify this has been fixed, if so, please close it.
Responsible Changed From-To: core->se awaiting feedback
State Changed From-To: feedback->closed Fixed by Doug Rabson, commit message was: dfr 95/07/07 04:01:32 Modified: sys/nfs nfs_bio.c nfs_vfsops.c Log: Use a consistent blocksize for sizing bufs to avoid panicing the bio system.