Date: Wed, 4 Aug 1999 23:40:06 -0700 (PDT) From: Matthew Dillon <dillon@apollo.backplane.com> To: Stephen Hocking-Senior Programmer PGS Tensor Perth <shocking@prth.pgs.com> Cc: current@freebsd.org Subject: Re: Interesting NFS hangs under current Message-ID: <199908050640.XAA33933@apollo.backplane.com> References: <199908050625.OAA20277@ariadne.tensor.pgs.com>
next in thread | previous in thread | raw e-mail | index | archive | help
:No sooner received than done....
:
:(kgdb) frame 18
:#18 0xc01ef2d6 in mmap (p=0xc5e49020, uap=0xc5f45f80) at ../../vm/vm_mmap.c:330
:330 error = vm_mmap(&p->p_vmspace->vm_map, &addr, size, prot, maxprot,
:(kgdb) print *p
:$2 = {p_procq = {tqe_next = 0xc0290ed0, tqe_prev = 0x0}, p_list = {
:...
: p_comm = "rpc.rstatd\000\000\000\000\000\000", p_pgrp = 0xc0a42ae0,
:...
: p_sysent = 0xc025bbc0, p_rtprio = {type = 1, prio = 0}, p_prison = 0x0,
: p_addr = 0xc5f44000, p_md = {md_regs = 0xc5f45fa8}, p_xstat = 0,
:---Type <return> to continue, or q <return> to quit---
: p_acflag = 2, p_ru = 0x0, p_nthreads = 0, p_aioinfo = 0x0, p_wakeup = 0,
: p_peers = 0x0, p_leader = 0xc5e49020, p_asleep = {as_priority = 0,
: as_timo = 0}, p_emuldata = 0x0}
:...
:(kgdb) print *uap
:$3 = {addr = 0xc07a8180 "", addr_ = 0xc0a41744 "À\206zÀ\001",
: len = 3229255360, len_ = 0xc0a41748 "\001", prot = 65537,
: prot_ = 0xc0a4174c "\001", flags = 1, flags_ = 0xc0a41750 "\200>¢ÀÄ#&À\002",
: fd = -1063108992, fd_ = 0xc0a41754 "Ä#&À\002", pad = -1071242300,
: pad_ = 0xc0a41758 "\002", pos = 17592186044418, pos_ = 0xc0a41760 ""}
:(kgdb)
Uh oh, this is rpc.statd.
I don't get it. How can rpc.statd be in an mmap() call at this point?
rpc.statd should have been running long before you even started the link.
Are you running rpc.statd from inetd.conf by any chance? If so, try
removing it from inetd.conf and run it manually.
:# ps -axl -N /sys/compile/bleep/kernel.debug -M /var/crash/vmcore.2
: UID PID PPID CPU PRI NI VSZ RSS WCHAN STAT TT TIME COMMAND
: 0 259 1 32 10 0 508 0 wait I #C1- 0:00.00 (sh)
: 88 276 259 0 2 0 11132 0 - R #C1- 0:00.00 (mysqld)
:...
The large number of processes in "R"un state indicates that the machine
is stuck in a supervisor loop in the mmap() code. i.e. interrupts have
woken up these processes but they have been unable to actually get any
cpu.
Nothing else seems to be an issue. I see you are running vinum. I do
not know if that is an issue.
: 0 4667 4659 14 -2 0 340 0 getblk I+ #C1 0:00.00 (install)
: 0 182 1 0 -1 0 208 0 nfsrcv I ?? 0:00.00 (nfsiod)
: 0 183 1 0 2 0 208 0 - R ?? 0:00.00 (nfsiod)
: 0 242 1 0 2 0 1452 0 - Rs ?? 0:00.00 (ppp)
: 0 351 1 1 -6 0 476 0 vinum Ds ?? 0:00.00 (vinum)
: 0 3486 292 0 2 0 1296 0 - R ?? 0:00.00 (sshd1)
: 0 4672 199 1502 105 0 868 0 - Rs ?? 0:00.00 (rpc.rstatd
:#
:...
-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?199908050640.XAA33933>