LinuxSA Mailing list archives
Index:
[thread]
[date]
[subject]
[author]
[stats]
From: Andrew Halliday <andrew@recalldesign.com>
To : <linuxsa@linuxsa.org.au>
Date: Wed, 13 Dec 2000 11:59:40 +1030
samba speed
Hi, I was wondering if anyone could help me with this samba speed problem.
BACKGROUND:
I am using Samba Version 2.0.7. in a box on our lan whose sole purpose is
to serve mp3s using a samba share so that people dont waste space on their
workstations. It is on the external network (behind nothing) which means
that it has to pass through a router to get to the lan, but i cant see this
as being a problem.
Box specs:
p1-120
24mb ram
debian (recent release - cant remember the versio number)
default kernel
10mb ethernet card (Realtek)
The problems are that mp3s skip. Chronically. They play fine for the first
5 seconds and then skip for half a second every 2-3 seconds (caching?!).
This is only with one user on the box. Annoying. Initially I thought it
was my machine (client), since all my workmates in my cubicle have much
faster computer (mine: 200, theirs: 500+++). However, someone just walked
up to me with a 700 and complained of the same skipping problem.
So I went and read the man page for smb.conf and read speed.txt and made the
following additions to my smb.conf file:
oplock = True
level2 oplocks = true
fake oplocks = yes
share modes = off
read prediction = yes
and also changed:
socket options = TCP_NODELAY IPTOS_LOWDELAY
from
socket options = TCP_NODELAY
According to speed.txt, these changes should have astronomically improved
speed. I copied a file and it did appear to improve speed, but only
marginally. More specifically, it took 4minutes 32sec to copy a 3.52mb file
... which is an absolute nightmare compared to the speeds i have seen samba
work at before ... ie as if it were like a local hard disk (almost) !!!
So WHATS BOTTLENECKING?!
The client? (mostlikely)- given that it works okay on some clients, in which
case how do you set options for that - i havent seen any smb options in
windows ...
The network card? - but i thought it would be able to handle ***1*** mp3
stream!
RAM? - ie an mp3 plays fine for the first few seconds ... suggesting some
form of caching issue?
Thoughts anyone?
AndrewH
PS Here are some PS and TOP listings during and after the copy
DURING
------
F S UID PID PPID C PRI NI ADDR SZ WCHAN TTY TIME CMD
100 S 0 1 0 0 60 0 - 303 select ? 00:00:11 init
040 S 0 2 1 0 60 0 - 0 bdflus ? 00:00:00 kflushd
040 S 0 3 1 0 60 0 - 0 kupdat ? 00:00:01 kupdate
040 S 0 4 1 0 60 0 - 0 kpiod ? 00:00:00 kpiod
040 S 0 5 1 0 60 0 - 0 kswapd ? 00:00:38 kswapd
140 S 0 137 1 0 60 0 - 322 select ? 00:00:01 syslogd
140 S 0 139 1 0 60 0 - 403 syslog ? 00:00:01 klogd
000 S 31 171 1 0 60 0 - 1389 select ? 00:00:00
postmaster
140 S 0 185 1 0 61 0 - 653 select ? 00:00:43 sshd
000 S 0 196 1 0 60 0 - 299 read_c tty2 00:00:00 getty
000 S 0 197 1 0 60 0 - 299 read_c tty3 00:00:00 getty
000 S 0 198 1 0 60 0 - 299 read_c tty4 00:00:00 getty
000 S 0 199 1 0 60 0 - 299 read_c tty5 00:00:00 getty
000 S 0 200 1 0 60 0 - 299 read_c tty6 00:00:00 getty
000 S 0 790 1 0 60 0 - 299 read_c tty1 00:00:00 getty
140 S 0 5985 185 0 61 0 - 764 select ? 00:00:08 sshd
100 S 1000 5986 5985 0 60 0 - 565 wait4 pts/2 00:00:00 bash
100 S 0 6053 5986 0 65 0 - 579 wait4 pts/2 00:00:02 bash
140 S 0 6629 1 0 60 0 - 319 select ? 00:00:00 atalkd
140 S 0 6633 1 0 60 0 - 520 select ? 00:00:00 afpd
040 S 0 6635 1 0 60 0 - 309 select ? 00:00:00 papd
140 S 0 6662 1 0 60 0 - 9301 select ? 00:00:02 apache
140 S 65534 6664 6662 0 60 0 - 9316 posix_ ? 00:00:00 apache
140 S 65534 6670 6662 0 60 0 - 9307 wait_f ? 00:00:00 apache
140 S 0 6711 185 0 60 0 - 743 select ? 00:00:01 sshd
100 S 1000 6712 6711 0 60 0 - 563 wait4 pts/0 00:00:00 bash
100 S 0 6713 6712 0 60 0 - 573 read_c pts/0 00:00:00 bash
140 S 0 6747 1 0 60 0 - 605 select ? 00:00:00 nmbd
140 S 0 6751 1 0 60 0 - 744 select ? 00:00:00 smbd
140 S 65534 6754 6751 0 64 0 - 1081 select ? 00:00:01 smbd
100 R 0 6759 6053 0 71 0 - 772 - pts/2 00:00:00 ps
12:02pm up 4 days, 18:09, 2 users, load average: 0.00, 0.00, 0.00
31 processes: 30 sleeping, 1 running, 0 zombie, 0 stopped
CPU states: 1.2% user, 1.8% system, 0.0% nice, 97.1% idle
Mem: 21748K av, 21300K used, 448K free, 15464K shrd, 1444K buff
Swap: 127980K av, 2804K used, 125176K free 11728K cached
PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
6760 root 15 0 1368 1368 760 R 0 2.1 6.2 0:00 top
6754 nobody 3 0 2788 2788 2452 S 0 0.5 12.8 0:02 smbd
5985 root 1 0 452 240 140 S 0 0.1 1.1 0:08 sshd
1 root 0 0 112 68 52 S 0 0.0 0.3 0:11 init
2 root 0 0 0 0 0 SW 0 0.0 0.0 0:00 kflushd
3 root 0 0 0 0 0 SW 0 0.0 0.0 0:01 kupdate
4 root 0 0 0 0 0 SW 0 0.0 0.0 0:00 kpiod
5 root 0 0 0 0 0 SW 0 0.0 0.0 0:38 kswapd
137 root 0 0 264 224 172 S 0 0.0 1.0 0:01 syslogd
139 root 0 0 572 164 136 S 0 0.0 0.7 0:01 klogd
171 postgres 0 0 220 0 0 SW 0 0.0 0.0 0:00
postmaster
185 root 0 0 536 444 372 S 0 0.0 2.0 0:43 sshd
196 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
197 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
198 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
199 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
200 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
790 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
5986 andrew 0 0 288 0 0 SW 0 0.0 0.0 0:00 bash
6053 root 3 0 788 656 488 S 0 0.0 3.0 0:02 bash
6629 root 0 0 112 48 40 S 0 0.0 0.2 0:00 atalkd
6633 root 0 0 180 0 0 SW 0 0.0 0.0 0:00 afpd
6635 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 papd
6662 root 0 0 2132 2108 2048 S 0 0.0 9.6 0:02 apache
6664 nobody 0 0 2264 2248 2104 S 0 0.0 10.3 0:00 apache
6670 nobody 0 0 2164 2140 2072 S 0 0.0 9.8 0:00 apache
6711 root 0 0 1448 1420 1184 S 0 0.0 6.5 0:01 sshd
6712 andrew 0 0 1316 1316 1036 S 0 0.0 6.0 0:00 bash
6713 root 0 0 1380 1380 1060 S 0 0.0 6.3 0:00 bash
6747 root 0 0 1280 1280 924 S 0 0.0 5.8 0:00 nmbd
6751 root 0 0 1296 1296 1140 S 0 0.0 5.9 0:00 smbd
AFTER
-----
F S UID PID PPID C PRI NI ADDR SZ WCHAN TTY TIME CMD
100 S 0 1 0 0 60 0 - 303 select ? 00:00:11 init
040 S 0 2 1 0 60 0 - 0 bdflus ? 00:00:00 kflushd
040 S 0 3 1 0 60 0 - 0 kupdat ? 00:00:01 kupdate
040 S 0 4 1 0 60 0 - 0 kpiod ? 00:00:00 kpiod
040 S 0 5 1 0 60 0 - 0 kswapd ? 00:00:38 kswapd
140 S 0 137 1 0 60 0 - 322 select ? 00:00:01 syslogd
140 S 0 139 1 0 60 0 - 403 syslog ? 00:00:01 klogd
000 S 31 171 1 0 60 0 - 1389 select ? 00:00:00
postmaster
140 S 0 185 1 0 60 0 - 653 select ? 00:00:43 sshd
000 S 0 196 1 0 60 0 - 299 read_c tty2 00:00:00 getty
000 S 0 197 1 0 60 0 - 299 read_c tty3 00:00:00 getty
000 S 0 198 1 0 60 0 - 299 read_c tty4 00:00:00 getty
000 S 0 199 1 0 60 0 - 299 read_c tty5 00:00:00 getty
000 S 0 200 1 0 60 0 - 299 read_c tty6 00:00:00 getty
000 S 0 790 1 0 60 0 - 299 read_c tty1 00:00:00 getty
140 S 0 5985 185 0 61 0 - 764 select ? 00:00:09 sshd
100 S 1000 5986 5985 0 60 0 - 565 wait4 pts/2 00:00:00 bash
100 S 0 6053 5986 0 67 0 - 579 wait4 pts/2 00:00:02 bash
140 S 0 6629 1 0 60 0 - 319 select ? 00:00:00 atalkd
140 S 0 6633 1 0 60 0 - 520 select ? 00:00:00 afpd
040 S 0 6635 1 0 60 0 - 309 select ? 00:00:00 papd
140 S 0 6662 1 0 60 0 - 9301 select ? 00:00:02 apache
140 S 65534 6664 6662 0 60 0 - 9316 posix_ ? 00:00:00 apache
140 S 65534 6670 6662 0 60 0 - 9307 wait_f ? 00:00:00 apache
140 S 0 6711 185 0 60 0 - 743 select ? 00:00:01 sshd
100 S 1000 6712 6711 0 60 0 - 563 wait4 pts/0 00:00:00 bash
100 S 0 6713 6712 0 60 0 - 573 read_c pts/0 00:00:00 bash
140 S 0 6747 1 0 60 0 - 605 select ? 00:00:00 nmbd
140 S 0 6751 1 0 60 0 - 744 select ? 00:00:00 smbd
140 S 65534 6754 6751 0 60 0 - 1081 select ? 00:00:03 smbd
100 R 0 6770 6053 0 79 0 - 772 - pts/2 00:00:00 ps
12:10pm up 4 days, 18:17, 2 users, load average: 0.07, 0.02, 0.00
31 processes: 30 sleeping, 1 running, 0 zombie, 0 stopped
CPU states: 1.6% user, 0.8% system, 0.0% nice, 97.7% idle
Mem: 21748K av, 21324K used, 424K free, 15464K shrd, 488K buff
Swap: 127980K av, 2804K used, 125176K free 12708K cached
PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
6771 root 18 0 1368 1368 760 R 0 1.5 6.2 0:00 top
5985 root 4 0 452 240 140 S 0 0.7 1.1 0:10 sshd
1 root 0 0 112 68 52 S 0 0.0 0.3 0:11 init
2 root 0 0 0 0 0 SW 0 0.0 0.0 0:00 kflushd
3 root 0 0 0 0 0 SW 0 0.0 0.0 0:01 kupdate
4 root 0 0 0 0 0 SW 0 0.0 0.0 0:00 kpiod
5 root 0 0 0 0 0 SW 0 0.0 0.0 0:38 kswapd
137 root 0 0 264 224 172 S 0 0.0 1.0 0:01 syslogd
139 root 0 0 572 164 136 S 0 0.0 0.7 0:01 klogd
171 postgres 0 0 220 0 0 SW 0 0.0 0.0 0:00
postmaster
185 root 0 0 536 444 372 S 0 0.0 2.0 0:43 sshd
196 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
197 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
198 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
199 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
200 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
790 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 getty
5986 andrew 0 0 288 0 0 SW 0 0.0 0.0 0:00 bash
6053 root 1 0 788 656 488 S 0 0.0 3.0 0:02 bash
6629 root 0 0 112 48 40 S 0 0.0 0.2 0:00 atalkd
6633 root 0 0 180 0 0 SW 0 0.0 0.0 0:00 afpd
6635 root 0 0 64 0 0 SW 0 0.0 0.0 0:00 papd
6662 root 0 0 2132 2108 2048 S 0 0.0 9.6 0:02 apache
6664 nobody 0 0 2264 2248 2104 S 0 0.0 10.3 0:00 apache
6670 nobody 0 0 2164 2140 2072 S 0 0.0 9.8 0:00 apache
6711 root 0 0 1448 1420 1184 S 0 0.0 6.5 0:01 sshd
6712 andrew 0 0 1316 1316 1036 S 0 0.0 6.0 0:00 bash
6713 root 0 0 1380 1380 1060 S 0 0.0 6.3 0:00 bash
6747 root 0 0 1280 1280 924 S 0 0.0 5.8 0:00 nmbd
6751 root 0 0 1296 1296 1140 S 0 0.0 5.9 0:00 smbd
6754 nobody 0 0 2788 2788 2452 S 0 0.0 12.8 0:03 smbd
Normally there is a lot more memory free ...
--------------------------------------------------------------
Andrew Halliday | Phone : +61 8 8217 0500
Recall Design Pty Ltd | Fax : +61 8 8217 0555
53 Gilbert Street | Mobile : 0407 779980
Adelaide SA 5000 | E-mail : andrew@recalldesign.com
AUSTRALIA | Website : http://www.recalldesign.com
--------------------------------------------------------------
--
LinuxSA WWW: http://www.linuxsa.org.au/ IRC: #linuxsa on irc.linux.org.au
To unsubscribe from the LinuxSA list:
mail linuxsa-request@linuxsa.org.au with "unsubscribe" as the subject
Index:
[thread]
[date]
[subject]
[author]
[stats]
Return to the LinuxSA Mailing List Information Page