10-26-2005
Actually - what are you doing?
Unless you really have I/O problems - or disk thrashing - or some clearly defined problem, like every process is totally I/O bound, you should probably leave things alone.
It appears to me like you do not know what you are looking at, so your changing things could lead to big trouble.
Process space can also have file caches or buffers. Applications control caching (buffering) and that's where control of buffers usually should start, not at the kernel level. Look for applications that have called setvbuf() to turn off buffering, for example.
In general, disks have pretty big on-board caches, so you don't have to worry too much about the kernel cache if things are working reasonably well.
The write thing is the number of cache hits/misses - misses caused by writing new data to files; hits are a write updating an existing record in the cache, like a row in a database table.
Generally, the best thing to do is called load-leveling - placing really active directories on physically separate disks. This mostly applies to writes, but if there are lots of reports running you may also have a problem with reads bombing one disk, so move directories around in that case as well. The I/O request queue indicates the magnitude of this problem.
9 More Discussions You Might Find Interesting
1. UNIX for Dummies Questions & Answers
I executed the following command in the korn shell:
$ variable1="qwerty" ls | sort
and the shell executed the 'ls | sort' command.
I would have expected an error message from the shell, but instead of that the shell ran the 'ls | sort' command and didn't realize the variable assignement. ... (1 Reply)
Discussion started by: PhilippeCrokaer
1 Replies
2. AIX
hello
with a sar i have this result:
System configuration: lcpu=48 ent=4.00
14:06:37 %usr %sys %wio %idle physc %entc
14:06:39 26 9 3 62 1.63 40.7
14:06:41 26 9 3 63 1.58 39.4
14:06:43 ... (0 Replies)
Discussion started by: pascalbout
0 Replies
3. IP Networking
Hi Guys,
This is really really urgent. Am looking out for some quick answers.
I'm developing a DNS Resolver client that interprets DNS Query repsonses & pass on the needful to DNS applications.
When an ENUM query(modified to an nslookup naptr query) is issued & an NAPTR RR(Resource Record)... (1 Reply)
Discussion started by: smanu
1 Replies
4. UNIX for Dummies Questions & Answers
Hi,
I get the message NOTICE HTFS :No Space on dev hd (1/104), What does (1/104) mean? Is there any link, I can get material on understanding unix message log?
thanks. (4 Replies)
Discussion started by: scomrade
4 Replies
5. Solaris
Hi
What means Solaris 10 5/09 and Solaris 10 10/09, I mean the suffix 5/09 and 10/09 ?
thx for help. (1 Reply)
Discussion started by: presul
1 Replies
6. Shell Programming and Scripting
I'm writing a Korn script but am having trouble because the shell interprets the asterisk in this case. Can anyone tell me if there is a way to fix this so that grep takes in STDIN without the interpretation?
line="30 09 * * 1-4 /home/user01/bin/start"
echo "$line" | grep 'start' (16 Replies)
Discussion started by: sprucio
16 Replies
7. AIX
On my VIo I see the following for my disks:
$ lspath | grep hdisk6
Enabled hdisk6 fscsi0 200600a0b82193f7,4000000000000
Enabled hdisk6 fscsi0 200700a0b82193f7,4000000000000
Enabled hdisk6 fscsi2 200600a0b82193f8,4000000000000
Failed hdisk6 fscsi2 200700a0b82193f8,4000000000000
$ lspath |... (8 Replies)
Discussion started by: petervg
8 Replies
8. UNIX for Advanced & Expert Users
hi, anyone can help on this piece of truss output?
8094: 0.7028 write(4, 0x0043BE90, 236) = 236
8094: T S H \0\0\0EC020101\0\0\0\0\0\0\0\0\0 "02\0\0 303\0\0 I D
8094: \f %\0\0\0\0 2\0F67F\0\0\0\0 @06FFC99A ;
8094: L D6\0 303
8094: ... (6 Replies)
Discussion started by: ghostdog74
6 Replies
9. UNIX for Advanced & Expert Users
hi,
working on Solaris 10. need your help on ping behaviour that I encountered.
I ping from source to destination
-bash-3.2# ping -s -t 128 10.10.10.200
PING 10.10.10.200: 56 data bytes <===== stops here for 2 minutues before getting reply back
64 bytes from 10.10.10.200:... (9 Replies)
Discussion started by: ghostdog74
9 Replies
LEARN ABOUT REDHAT
pwrite
PREAD(2) Linux Programmer's Manual PREAD(2)
NAME
pread, pwrite - read from or write to a file descriptor at a given offset
SYNOPSIS
#define _XOPEN_SOURCE 500
#include <unistd.h>
ssize_t pread(int fd, void *buf, size_t count, off_t offset);
ssize_t pwrite(int fd, const void *buf, size_t count, off_t offset);
DESCRIPTION
pread() reads up to count bytes from file descriptor fd at offset offset (from the start of the file) into the buffer starting at buf. The
file offset is not changed.
pwrite() writes up to count bytes from the buffer starting at buf to the file descriptor fd at offset offset. The file offset is not
changed.
The file referenced by fd must be capable of seeking.
RETURN VALUE
On success, the number of bytes read or written is returned (zero indicates that nothing was written, in the case of pwrite, or end of
file, in the case of pread), or -1 on error, in which case errno is set to indicate the error.
ERRORS
pread can fail and set errno to any error specified for read(2) or lseek(2). pwrite can fail and set errno to any error specified for
write(2) or lseek(2).
CONFORMING TO
Unix98
HISTORY
The pread and pwrite system calls were added to Linux in version 2.1.60; the entries in the i386 system call table were added in 2.1.69.
The libc support (including emulation on older kernels without the system calls) was added in glibc 2.1.
SEE ALSO
read(2), write(2), lseek(2)
Linux 2.2.0-pre9 1999-01-21 PREAD(2)