NFS problem with cloned hard disks

I have a computer that is attached to my network, and works just fine. I
cloned the hard disk, and installed this on a second computer, changing
the hostname and IP address. This to is on the network. I can telnet it,
and it can ping other stations, etc.

However, when I try to start the NFS client process using:

/etc/init.d/nfs start

I get an error as follows:

statd: Could not chdir: No such file or directory

I did a system trace, and have tracked this down to a difference in
behaviour between the first and second machine.

The system traces are identical up to a certain point.

Here is a section of the trace from the first (working) machine:

access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or
directory)
open("/etc/ld.so.cache", O_RDONLY) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=25907, ...}) = 0
mmap2(NULL, 25907, PROT_READ, MAP_PRIVATE, 3, 0) = 0xb7fda000
close(3) = 0
open("/lib/libc.so.6", O_RDONLY) = 3
read(3,
"\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0@a \1\0004\0\0\0"...,
512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=1192036, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0xb7fd9000
mmap2(NULL, 1197520, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3,
0) = 0xb7eb4000
mmap2(0xb7fd3000, 12288, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x11f) = 0xb7fd3000
mmap2(0xb7fd6000, 9680, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0xb7fd6000
close(3) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0xb7eb3000
set_thread_area({entry_number:-1 -> 6, base_addr:0xb7eb36c0,
limit:1048575, seg_32bit:1, contents:0, read_exec_only:0,
limit_in_pages:1, seg_not_present:0, useable:1}) = 0
mprotect(0xb7fd3000, 8192, PROT_READ) = 0
mprotect(0x8050000, 4096, PROT_READ) = 0
mprotect(0xb7ffb000, 4096, PROT_READ) = 0
munmap(0xb7fda000, 25907) = 0
getrlimit(RLIMIT_NOFILE, {rlim_cur=1024, rlim_max=1024}) = 0
pipe([3, 4]) = 0
clone(child_stack=0,
flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGC HLD,
child_tidptr=0xb7eb3708) = 2953
close(4) = 0
read(3, "\0", 1) = 1
exit_group(0) = ?

On the second machine, I get:

open("/lib/libc.so.6", O_RDONLY) = 3
read(3,
"\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0@a \1\0004\0\0\0"...,
512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=1192036, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0xb7fc9000
mmap2(NULL, 1197520, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3,
0) = 0xb7ea4000
mmap2(0xb7fc3000, 12288, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x11f) = 0xb7fc3000
mmap2(0xb7fc6000, 9680, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0xb7fc6000
close(3) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0xb7ea3000
set_thread_area({entry_number:-1 -> 6, base_addr:0xb7ea36c0,
limit:1048575, seg_32bit:1, contents:0, read_exec_only:0,
limit_in_pages:1, seg_not_present:0, useable:1}) = 0
mprotect(0xb7fc3000, 8192, PROT_READ) = 0
mprotect(0x8050000, 4096, PROT_READ) = 0
mprotect(0xb7feb000, 4096, PROT_READ) = 0
munmap(0xb7fca000, 25907) = 0
getrlimit(RLIMIT_NOFILE, {rlim_cur=1024, rlim_max=1024}) = 0
pipe([3, 4]) = 0
clone(statd: Could not chdir: No such file or directory
child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGC HLD,
child_tidptr=0xb7ea3708) = 3250
--- SIGCHLD (Child exited) @ 0 (0) ---
close(4) = 0
read(3, "", 1) = 0
exit_group(1) = ?

Notice that the clone call gives an error on the second machine.

What directory is is trying to chdir to?

This is a Gentoo based system initially built from stage 1.

Mark.

--
Mark Hobley,
393 Quinton Road West,
Quinton, BIRMINGHAM.
B32 1QE.

Maxwell Lol <(E-Mail Removed)> wrote:

> copy the /etc and /usr/lib directories, and do a recursive diff.

The drives were cloned. I will try swapping the drives between the two
machines to see if the problem moves with the drive or stays with the
machine.

> Also - execute
> sh -xv /etc/init.d/nfs start
> (I'm not sure this will reveal anything. It's been a while since I used NFS.).

This tells me:

+ opts=reload
+ restarting=no
+ exportfs=/usr/sbin/exportfs
+ mountd=/usr/sbin/rpc.mountd
+ nfsd=/usr/sbin/rpc.nfsd
+ smnotify=/usr/sbin/sm-notify

However the services do not appear to startup if I use that method.

> Make sure your NFS client has a mount point directory created.

The mount point exists on this machine

> Are you using automount? Or explicit mount configurations?

I am using the mount command as follows:

mount neptune:/volumes/vol3a/ /volumes/vol3a/
statd: Could not chdir: No such file or directory
mount.nfs: rpc.statd is not running but is required for remote locking
Either use "-o nolocks" to keep locks local, or start statd.

Out of interest I tried the -o nolocks option as follows:

mount -o nolocks neptune:/volumes/vol3a/ /volumes/vol3a/
Unsupported nfs mount option: nolocks

> Do you get different results using the NFS client only, or server only?

This is a client machine trying to connect to the server. I have not
configured this machine to run as a server, but I will test this.

> Are you running the port mapper?

The portmapper is running:

rpcinfo -p localhost
program vers proto port
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper

On the working client:

rpcinfo -p localhost
program vers proto port
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper
100021 1 tcp 43408 nlockmgr
100021 3 tcp 43408 nlockmgr
100021 4 tcp 43408 nlockmgr
100024 1 udp 34971 status
100024 1 tcp 57430 status

Regards,

Mark.

--
Mark Hobley,
393 Quinton Road West,
Quinton, BIRMINGHAM.
B32 1QE.

Mark Hobley <(E-Mail Removed)> wrote:

> I get an error as follows:
>
> statd: Could not chdir: No such file or directory

By using strace -tfv, I managed to identify the cause of this error.
This error occurs because the /var/lib/nfs directory did not exist. I
simply created the directory, and nfs became operational.

Regards,

Mark.

--
Mark Hobley,
393 Quinton Road West,
Quinton, BIRMINGHAM.
B32 1QE.