NOTICE
!!!NOTICE!!! !!!NOTICE!!! !!!NOTICE!!! !!!NOTICE!!! !!!NOTICE!!!!
! !
! Revision 1.29 represents the end of life for lslk. I don’t have time !
! to support it. Please don’t report bugs to me. I will politely !
! decline to work on them. !
! !
! Vic Abell <abe@purdue.edu>, July 11, 2001 !
! !
!!!NOTICE!!! !!!NOTICE!!! !!!NOTICE!!! !!!NOTICE!!! !!!NOTICE!!!!
NAME
lslk - list local locks
SYNOPSIS
lslk [ -abhnOvw ] [ -i i ] [ -k k ] [ -p p ] [ -S [t] ] [ paths ]
DESCRIPTION
Lslk revision 1.29 lists information about locks held on files with
local inodes on systems running the following UNIX dialects:
AIX 3.2.5, 4.1.4, 4.2[.1], and AIX 4.3[.[12]]
DEC OSF/1, Digital UNIX, and Tru64 UNIX [2345].[01] and 3.2
Linux
SCO OpenDesktop or OpenServer 3.0 and 5.0.[0245]
Sequent PTX 2.1.9, 4.2.1, 4.3, and 4.4
Solaris 2.[345], 2.5.1, 2.6, 7, and 8 (excluding Veritas
VxFS local files)
SunOS 4.1.3
The lock may belong to a process on the local system or to a process on
an NFS client system to which the local system is an NFS server.
Notes: Linux and PTX 2.1.9 lslk don’t report on locks held by NFS
clients; Solaris lslk won’t report locks held on local Veritas VxFS
files.
OPTIONS
In the absence of any options, lslk lists all locks associated with the
local files of the system.
When selection options are specified, the listing of all locks is
disabled, and the selection options are ORed together. Only locks
meeting any selection criterion are listed.
When the -a option is specified, the listing of all locks is disabled,
and the selection options are ANDed together. Only locks that meet all
selection criteria are listed.
-a This option causes list selection options to be ANDed, as
described above.
-b This option causes lslk to avoid kernel functions that might
block - lstat(2), readlink(2), and stat(2).
See the BLOCKS AND TIMEOUTS and AVOIDING KERNEL BLOCKS
sections for information on using this option.
-i i This option selects the listing of locks whose owning process
is on the Internet host whose name or network address is i.
Multiple addresses may be declared with multiple -i i options.
-k k This option specifies k as an alternate to the default kernel
name list file path. The default kernel name list file path
is listed in the -h help output.
It may be necessary to specify an alternate kernel name list
path when the file at the default path isn’t the booted kernel
-- e.g., the default is /vmunix, but the booted kernel file is
/vmunix.new.
Unless lslk accesses the correct kernel name list file, it may
derive incorrect addresses for symbols in kernel memory,
causing it to fail.
-n This option inhibits the conversion of network host names to
network addresses and the conversion of network addresses to
network host names.
This option may be useful when the host name to address
translation service (e.g., the Domain Name Server) is slow or
inoperative.
If you use this option on hosts whose kernel lock table
contains only host names - e.g., SCO or Solaris - or you use
this option and also select the listing of locks by an
Internet network address with the -i i option, lslk will not
be able to locate any locks with the specified network
address, nor will it be able to report network addresses in
its output. The -n option inhibits the necessary conversion
of kernel lock table host names to network addresses.
-O This option directs lslk to bypass the strategy it uses to
avoid being blocked by some kernel operations - i.e., doing
them in forked child processes. See the BLOCKS AND TIMEOUTS
and AVOIDING KERNEL BLOCKS sections for more information on
kernel operations that may block lslk.
-p p This option selects the listing of locks whose owning process
IDentification (PID) numbers are in the comma-separated list,
p.
-S [t] This option specifies an optional time-out seconds value for
kernel functions - lstat(2), readlink(2), and stat(2) - that
might otherwise deadlock. The minimum for t is two; the
default, fifteen; when no value is specified, the default is
used.
See the BLOCKS AND TIMEOUTS section for more information.
-v When this option is specified, lslk lists version information
- i.e., where, when and how it was constructed.
-w This option suppresses non-fatal warning messages.
paths This option specifies a list of file path names for which lslk
is to list lock information.
OUTPUT
Lslk lists the information for each lock on a separate line in the
following columns. (The columns are dynamically sized.)
SRC indicates the source of the process holding the lock.
If the source is a local process, its command name is
displayed, or ‘‘(unknown)’’ if the command name can’t be
determined.
If the source is a remote process, the host name or network
number where the remote process executes is displayed.
PID is the Process IDentification number of the process holding the
lock.
DEV is the device (major and minor numbers) on which the locked
file resides.
INUM is the inode number of the locked file.
SZ is the size of the locked file.
TY is the lock type:
r read;
rw read and write;
w write;
? unknown.
M is the mandatory state of the lock: 0 if none; 1 if set. (See
chmod(1)).
ST is the relative byte offset of the lock.
WH is the starting offset (‘‘whence’’) of the lock.
END is the ending offset of the lock.
LEN is the length of the lock.
NAME is the name of the locked file, if it was specified as a path
argument.
If there is no path argument name, then the mount point and
device paths of the file system on which the locked file
resides are displayed.
BLOCKS AND TIMEOUTS
Lslk can be blocked by some kernel functions that it uses - lstat(2),
readlink(2), and stat(2). These functions are stalled in the kernel,
for example, when the hosts where mounted NFS file systems reside
become inaccessible.
Lslk attempts to break these blocks with timers and child processes,
but the techniques are not wholly reliable. When lslk does manage to
break a block, it will report the break with an error message. The
messages may be suppressed with the -w option.
The default timeout value may be displayed with the -h option, and it
may be changed with the -S [t] option. The minimum for t is two
seconds, but you should avoid small values, since slow system
responsiveness can cause short timeouts to expire unexpectedly and
perhaps stop lslk before it can produce any output.
When lslk has to break a block during its access of mounted file system
information, it normally continues, although with less information
available to display about open files.
Lslk can also be directed to avoid the protection of timers and child
processes when using the kernel functions that might block by
specifying the -O option. While this will allow lslk to start up with
less overhead, it exposes lslk completely to the kernel situations that
might block it. Use this option cautiously.
AVOIDING KERNEL BLOCKS
You can use the -b option to tell lslk to avoid using kernel functions
that would block. Some cautions apply.
First, using this option usually requires that your system supply
alternate device numbers in place of the device numbers that lslk would
normally obtain with the lstat(2) and stat(2) kernel functions. See
the ALTERNATE DEVICE NUMBERS section for more information on alternate
device numbers.
Second, you can’t specify the names of files you want lslk to locate
locks for unless they’re file system names. This is because lslk needs
to know the device and inode numbers of files listed with names in the
lslk options, and the -b option prevents lslk from obtaining them.
Moreover, since lslk only has device numbers for the file systems that
have alternates, its ability to locate locks on file systems depends
completely on the availability and accuracy of the alternates. If no
alternates are available, or if they’re incorrect, lslk won’t be able
to locate locks on the named file systems.
Third, if the names of your file system directories that lslk obtains
from your system’s mount table are symbolic links, lslk won’t be able
to resolve the links. This is because the -b option causes lslk to
avoid the kernel readlink(2) function it uses to resolve symbolic
links.
Finally, using the -b option causes lslk to issue warning messages when
it needs to use the kernel functions that the -b option directs it to
avoid. You can suppress these messages by specifying the -w option,
but if you do, you won’t see the alternate device numbers reported in
the warning messages.
ALTERNATE DEVICE NUMBERS
On some dialects, when lslk has to break a block because it can’t get
information about a mounted file system via the lstat(2) and stat(2)
kernel functions, or because you specified the -b option, lslk can
obtain some of the information it needs - the device number and
possibly the file system type - from the system mount table. When that
is possible, lslk will report the device number it obtained. (You can
suppress the report by specifying the -w option.)
You can assist this process if your mount table is supported with an
/etc/mtab or /etc/mnttab file that contains an options field by adding
a ‘‘dev=xxxx’’ field for mount points that do not have one in their
options strings.
The ‘‘xxxx’’ portion of the field is the hexadecimal value of the file
system’s device number. (Consult the st_dev field of the output of the
lstat(2) and stat(2) functions for the appropriate values for your file
systems.) Here’s an example from a Solaris 2.5 /etc/mnttab for a UFS
file system:
... ufs suid,rw,dev=80001f ...
Some dialects that do not use an ASCII /etc/mtab or /etc/mnttab file
for the mount table may still provide an alternative device number in
their internal mount tables. This includes AIX, DEC OSF/1, Digital
UNIX, and Tru64 UNIX. Lslk knows how to obtain the alternative device
number for these dialects and uses it when its attempt to lstat(2) or
stat(2) the file system is blocked.
If you’re not sure your dialect supplies alternate device numbers for
file systems from its mount table, use this lslk incantation to see if
it reports any alternate device numbers:
lslk -b
Look for standard error file warning messages that begin ‘‘assuming
"dev=xxxx" from ...’’.
DIAGNOSTICS
Errors are identified with messages on the standard error file.
Lslk returns a one (1) if an error was detected or if it couldn’t list
lock information for all the names that were specified.
EXAMPLES
To list all locks, use:
lslk
To list locks from the host ‘‘klaatu’’ in the local domain, use:
lslk -i klaatu
To list locks from the hosts helios.cc.purdue.edu and
vic1.cc.purdue.edu, use:
lslk -i helios.cc.purdue.edu -i vic1.cc.purdue.edu
To list locks held by processes 1234 and 56789, use:
lslk -p 1234,56789
To list all locks held by process 1234 on host klaatu, use:
lslk -p 1234 -a -i klaatu
PERMISSION
Lslk must have permission to access the system memory files - e.g.,
/dev/kmem and /dev/mem. Permission to do that is granted when the lslk
process is run from a root shell, or when its ‘‘setgid’’ group matches
the group (e.g., ‘‘sys’’) that can read the system memory files.
BUGS
Perhaps it should be possible to specify the host on which a search
target PID is located.
DEC OSF/1, Digital UNIX, and Tru64 UNIX lslk won’t find locks applied
with the flock(2) function. It will find locks applied with the
fcntl(2) and lockf(3) functions.
Linux lslk will not report locks held by NFS clients. It may have
difficulty reporting locks held on dynamic inodes (e.g., for the Win-95
file system type, smbfs).
Solaris lslk won’t find locks held on local Veritas VxFS files
by local or remote processes.
FILES
/dev/kmem kernel virtual memory device
/dev/mem physical memory device
AUTHORS
Lslk was written by Victor A. Abell <abe@purdue.edu> of the Purdue
University Computing Center (PUCC).
Chris Eleveld <chris@sector7.com> did the DEC OSF/1, Digital UNIX,
Linux, and Tru64 UNIX ports.
SEE ALSO
chmod(1), fcntl(2), fcntl(5), flock(3B), lockd(1M), lstat(2),
lockf(3C), readlink(2), stat(2).
Revision-1.29