Date: Wed, 11 Aug 1999 17:34:08 -0700 (PDT) From: Matthew Dillon <dillon@apollo.backplane.com> To: A.Leidinger@WJPServer.CS.Uni-SB.de Cc: freebsd-current@FreeBSD.ORG Subject: Re: Crash with NFS V2 Message-ID: <199908120034.RAA79109@apollo.backplane.com> References: <199908111850.UAA00795@work.net.local>
next in thread | previous in thread | raw e-mail | index | archive | help
:
:Hi,
:
:I have a program which does something like userland-nfs (nfs v2 client).
:It is able to crash the machine. After compiling a debug kernel the
:only output I get after the crashdump is:
:
:---snip---
:(100) root@ttyp3 # gdb -kernel -se /sys/compile/WORK/kernel.debug -c /var/crash/vmcore.1
:IdlePTD 3743744
:initial pcb at 2cb960
:panic messages:
:---
:dmesg: kvm_read: invalid address (c02bcb50)
:---
:#0 boot (howto=Cannot access memory at address 0xc62e6b54.
:) at ../../kern/kern_shutdown.c:291
:291 dumppcb.pcb_cr3 = rcr3();
:(kgdb) bt
:#0 boot (howto=Cannot access memory at address 0xc62e6b54.
:) at ../../kern/kern_shutdown.c:291
:Cannot access memory at address 0xc62e6b4c.
:---snip---
:
: - System is -current from august 5th.
: - "mountd -2 -n"
: - "nfsd -u"
:
:My program does a nfs read call and wants a buffer
:(readres.readres_u.reply.data.data_val) of 32k to be filled.
: -> crashed from userland
:(the program and the nfs server are running on the same machine)
:
:The core and the debug kernel are available, feel free to give me some
:advice how to get more information out of it.
:
:-----
:There is also a strange behavior in nfs_readdir:
:
:I get a segfault with readdir (in libc, something xdr relatet if I
:remember correctly) if the buffer
:(readdirres.readdirres_u.reply.entries) is less than 8k+1
:[NFS_MAXDATA+1] (on Solaris 2.4 this buffer has to be 257
:[NFS_MAXNAMLEN+1] or greater).
:This one is testet with:
: - NFS-Server: FreeBSD 3.x, Userland-NFS-Client: Solaris 2.4
: - NFS-Server + Userland-NFS-Client (on the same machine): FreeBSD
: 4-current (some months ago)
:
:Bye,
:Alexander.
The problem regarding the readdir seg faults should be fixed in
both STABLE and CURRENT if you update your source tree.
I don't know about the first problem... maybe compile up your kernel
DDB and do a 'trace' when it panics - record the original panic
message as well as the trace.
-Matt
Matthew Dillon
<dillon@backplane.com>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199908120034.RAA79109>