Hello,
i am using ClearOS Community release 6.6.0 (Final)
Kernel Version 2.6.32-504.8.1.v6.i686
System Time Thu Jun 4 13:23:06 EEST 2015
CPU Model Genuine Intel(R) CPU T2050 @ 1.60GHz
Memory Size 1.84 GB
Uptime 1 Days 4.3 Hours
Load 0.00 0.00 0.00
I have great issues with SAMBA running very slow, writitng of a file from W7 to samba is about 11.9 MB and reading is quite the same (tried with a 1.1 GB iso file)
the output hdparm is:
hdparm -Tt /dev/sda
/dev/sda:
Timing cached reads: 2176 MB in 2.00 seconds = 1087.83 MB/sec
Timing buffered disk reads: 336 MB in 3.01 seconds = 111.69 MB/sec
For now i have tried all methods of speeding up SAMBA from different forums and i upgraded the NIC to a gigabit one but the result is the same.
What else could i try or is this how it should go down?
The share is holding our accounting program wich has aprox. 150 mb large dbf database (consisting of aprox 100 files).
For sure the program was running faster from a WXP machine so there is for sure something i am missing in this.
i am using ClearOS Community release 6.6.0 (Final)
Kernel Version 2.6.32-504.8.1.v6.i686
System Time Thu Jun 4 13:23:06 EEST 2015
CPU Model Genuine Intel(R) CPU T2050 @ 1.60GHz
Memory Size 1.84 GB
Uptime 1 Days 4.3 Hours
Load 0.00 0.00 0.00
I have great issues with SAMBA running very slow, writitng of a file from W7 to samba is about 11.9 MB and reading is quite the same (tried with a 1.1 GB iso file)
the output hdparm is:
hdparm -Tt /dev/sda
/dev/sda:
Timing cached reads: 2176 MB in 2.00 seconds = 1087.83 MB/sec
Timing buffered disk reads: 336 MB in 3.01 seconds = 111.69 MB/sec
For now i have tried all methods of speeding up SAMBA from different forums and i upgraded the NIC to a gigabit one but the result is the same.
What else could i try or is this how it should go down?
The share is holding our accounting program wich has aprox. 150 mb large dbf database (consisting of aprox 100 files).
For sure the program was running faster from a WXP machine so there is for sure something i am missing in this.
Share this post:
Responses (12)
-
Accepted Answer
-
Accepted Answer
Using SMB2 may well help as it reduces the multiple command overhead you can get with lots of files being accessed concurrently. By default I think Samba 3.6 uses NT1 (CIFS) and not SMB2. If another device tries to access Samba with NT1 and max protocol = SMB2 it should not matter as the two systems will negotiate NT1 between them.
Reading the man pages, the "max protocol" bit should be put in the [global] section of smb.conf. -
Accepted Answer
on the wxp machine it was a shared folder, te dbf of the program is about 150 mb big with many small files , there are one or two files wich are something like 15 20 mb. the 11mb transfer speed was when uploading a iso file to the samba drive from a w7 machine.
as far as i remember there were no tx/rx errors. ah, the xwxp machine was upgraded with a ssd drive and booting time changed from 3 mins to 10 seconds while the program worked at best 5% faster so the results were not as expected.
ill test the new drivers tomorrow and ill also try to move the program to a synology NAS for some testing and see what comes out of this. -
Accepted Answer
I am surprised it was faster on a WXP machine with a 100Mbps card as 11.9MB/s is just a bit faster than a 100Mbps connection can handle.
Have you tried the alternative driver, although I don't think it will necessarily improve anything?
How were you using the WXP machine? Did you just share a folder?
I know very little about samba tuning and many references on the internet are obsolete. One thing you can do is try turning on smb2. It may help for small files. Add the line:
to /etc/samba/smb.conf. I've added it under a section called "# Other handy directives" but I may have added that as well! You may be able to reload samba with a "service smb reload". If not then you'll need to restart it.max protocol = SMB2
Another thing worth checking is for network errors:
Are there many TX or RX errors?ifconfig eth1
-
Accepted Answer
hi, the two speeds are from because i forgot to connect the cable to the gb port of the switch, second test is with gigabit. the strange part is that the tests look ok, hdd and network speed, just samba is painfully slow. when the program was hosted on a xp machine it was considerably faster even on 100 mbit network. coukd the source of the slow speed be the many small dbf files? -
Accepted Answer
Disk speed is fine and way faster than you are seeing in Samba. Network speeds look odd with one way slower than the other. I am no sure what is happening there. and you may need to investigate. Could it be a dodgy cable on either the transmit or receive pairings?
Looking at your original post, you are running the 32-bit version of ClearOS and I can't compile drivers for that. However Tim has the sk98lin driver compiled here. You could download and install it then reboot the server. Check that ClearOS is now using the new driver with:
If it is still using the sky2 driver, please post back as it may need blacklisting. I don't have a huge amount of hope for this but it is worth a try.lspci -k | grep Eth -A 3
-
Accepted Answer
So:
We have :
[me@clearos ~]# lspci -vn | grep 0200
Memory at d0200000 (32-bit, non-prefetchable) [size=512K]
01:00.0 0200: 11ab:4362 (rev 15)
Memory at d0020000 (64-bit, non-prefetchable) [size=16K]
02:08.0 0200: 8086:27dc (rev 02)
Then:
[root@clearos ~]# hdparm -tT /dev/sda
/dev/sda:
Timing cached reads: 2204 MB in 2.00 seconds = 1102.05 MB/sec
Timing buffered disk reads: 304 MB in 3.01 seconds = 100.88 MB/sec
and:
[me@clearos ~]# hdparm -I /dev/sda
/dev/sda:
ATA device, with non-removable media
Model Number: WDC WD10EFRX-68PJCN0
Serial Number: WD-WCC4JPWUU7P0
Firmware Revision: 01.01A01
Transport: Serial, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6, SATA Rev 3.0
Standards:
Supported: 9 8 7 6 5
Likely used: 9
Configuration:
Logical max current
cylinders 16383 16383
heads 16 16
sectors/track 63 63
--
CHS current addressable sectors: 16514064
LBA user addressable sectors: 268435455
LBA48 user addressable sectors: 1953525168
Logical Sector size: 512 bytes
Physical Sector size: 4096 bytes
Logical Sector-0 offset: 0 bytes
device size with M = 1024*1024: 953869 MBytes
device size with M = 1000*1000: 1000204 MBytes (1000 GB)
cache/buffer size = unknown
Nominal Media Rotation Rate: 5400
Capabilities:
LBA, IORDY(can be disabled)
Queue depth: 32
Standby timer values: spec'd by Standard, with device specific minimum
R/W multiple sector transfer: Max = 16 Current = 16
DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6
Cycle time: min=120ns recommended=120ns
PIO: pio0 pio1 pio2 pio3 pio4
Cycle time: no flow control=120ns IORDY flow control=120ns
Commands/features:
Enabled Supported:
* SMART feature set
Security Mode feature set
* Power Management feature set
* Write cache
* Look-ahead
* Host Protected Area feature set
* WRITE_BUFFER command
* READ_BUFFER command
* NOP cmd
* DOWNLOAD_MICROCODE
Power-Up In Standby feature set
* SET_FEATURES required to spinup after power up
SET_MAX security extension
* 48-bit Address feature set
* Device Configuration Overlay feature set
* Mandatory FLUSH_CACHE
* FLUSH_CACHE_EXT
* SMART error logging
* SMART self-test
Media Card Pass-Through
* General Purpose Logging feature set
* 64-bit World wide name
* URG for READ_STREAM[_DMA]_EXT
* URG for WRITE_STREAM[_DMA]_EXT
* IDLE_IMMEDIATE with UNLOAD
* WRITE_UNCORRECTABLE_EXT command
* {READ,WRITE}_DMA_EXT_GPL commands
* Segmented DOWNLOAD_MICROCODE
* Gen1 signaling speed (1.5Gb/s)
* Gen2 signaling speed (3.0Gb/s)
* Gen3 signaling speed (6.0Gb/s)
* Native Command Queueing (NCQ)
* Host-initiated interface power management
* Phy event counters
* Idle-Unload when NCQ is active
* NCQ priority information
* unknown 76[15]
DMA Setup Auto-Activate optimization
Device-initiated interface power management
* Software settings preservation
* SMART Command Transport (SCT) feature set
* SCT Write Same (AC2)
* SCT Error Recovery Control (AC3)
* SCT Features Control (AC4)
* SCT Data Tables (AC5)
unknown 206[12] (vendor specific)
unknown 206[13] (vendor specific)
Security:
Master password revision code = 65534
supported
not enabled
not locked
not frozen
not expired: security count
supported: enhanced erase
144min for SECURITY ERASE UNIT. 144min for ENHANCED SECURITY ERASE UNIT.
Logical Unit WWN Device Identifier: 50014ee25fbb5127
NAA : 5
IEEE OUI : 0014ee
Unique ID : 25fbb5127
Checksum: correct
And :
[me@clearos ~]# iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[ 4] local 10.0.0.138 port 5001 connected with 10.0.0.106 port 50210
[ ID] Interval Transfer Bandwidth
[ 4] 0.0-10.0 sec 110 MBytes 92.2 Mbits/sec
[ 5] local 10.0.0.138 port 5001 connected with 10.0.0.106 port 50226
[ ID] Interval Transfer Bandwidth
[ 5] 0.0-10.0 sec 890 MBytes 746 Mbits/sec -
Accepted Answer
Can you give the device ID's:
There may be a better/different driver that you can use (sk98lin) on your gigabit card. You did not give the output of "uname -r", but if you are running the 64bit version of ClearOS then I have compiled the driver here, but I'd like to check the device ID.lspci -vn | grep 0200
Have you been able to do a network speed test with something like iperf? If it shows no issues, then, when doing a file transfer, have a look at "top" for resource usage.
Do you know what mode your disks are running in? (ATA/IDE or AHCI)
What is the output of:
Where X is your drive letter (try a if you have a single disk) Also what is the output of:hdparm -tT /dev/sdX
Change X as necessary.hdparm -I /dev/sdX
-
Accepted Answer
01:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8053 PCI-E Gigabit Ethernet Controller (rev 15)
Subsystem: SysKonnect Device 4340
Kernel driver in use: sky2
Kernel modules: sky2
--
02:08.0 Ethernet controller: Intel Corporation NM10/ICH7 Family LAN Controller (rev 02)
Subsystem: Intel Corporation Device 0000
Kernel driver in use: e100
Kernel modules: e100
[me@clearos ~]# ethtool eth1
Settings for eth1:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Supported pause frame use: No
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Advertised pause frame use: No
Advertised auto-negotiation: Yes
Speed: 1000Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 0
Transceiver: internal
Auto-negotiation: on
MDI-X: Unknown
Supports Wake-on: pg
Wake-on: d
Current message level: 0x000000ff (255)
drv probe link timer ifdown ifup rx_err tx_err
Link detected: yes -
Accepted Answer
-
Accepted Answer
-
Accepted Answer
11.9MB/s os more or less 100mb/s, so the speed of fast ethernet. You say you changed the server to a gigabit NIC. Are all switches between the server and workstation gigabit rated? Also does the workstation have a gigabit NIC? Everything in the chain will need to be gigabit capable.
Can you check your link speed with "ethtool"? You'll have to google for the switches.
You can check your network speed running iperf on the server and workstation (there are WinX and Linux versions)
Again on the server, what is the output of:lspci -k | grep Eth -A 3
uname -r
Please login to post a reply
You will need to be logged in to be able to post a reply. Login using the form on the right or register an account if you are new here.
Register Here »