napalmd Posted August 10, 2012 Share Posted August 10, 2012 Hello. It's a unRAID Server Plus version: 4.7 I installed on my friend who works in photography. The setup is an old Asus P4p800 with pentium 4 3Ghz and 1 Gb ram. It has 6 Hdd's all Wd green 2Tb each parity drive is connected to onboard sata port Data drives: one on onboard sata, others in pci sata card Sil3114 He complains about slow access when browsing the folders in the unraid, takes a lot of time to open them in adobe bridge cs5 Also copying photos from flash card to server sometimes is really slow, like 1MB/s. I'm thinking about doing an upgrade to corei3, but I'm afraid that it will not improve performance as expected. Syslog is attached PS:They have also 2 other regular NAS machines using raid6 that work with no problems. unraid.zip Link to comment
napalmd Posted August 13, 2012 Author Share Posted August 13, 2012 A little help please? Thanks. Link to comment
Helmonder Posted August 13, 2012 Share Posted August 13, 2012 It kind of depends on what you are viewing as "slow".. read speed is not really improved when you upgrade cpu, write speed does get higher when you use a faster cpu.. http://lime-technology.com/forum/index.php?topic=21989.0 Link to comment
althoralthor Posted August 14, 2012 Share Posted August 14, 2012 Is it possible that the drives having to spin up is making him think things are slow? Link to comment
napalmd Posted August 14, 2012 Author Share Posted August 14, 2012 I don't think that's because of drives have to spin up because it hapens when he is entering and exploring folders and continues slow if he browses to another folder. Could it be because he has a lot of files/folders? He has like a folder for each year, and inside one for each month, inside this for each work that have inside about 1000 or more photos in raw and jpeg, about 10/20 MB each file. I monitored the RAM. free ram is only about 20MB maybe this is the problem? It has 1Gb installed... Link to comment
Helmonder Posted August 14, 2012 Share Posted August 14, 2012 Linux will allways eat all your RAM, free mem does not mean a lot with unraid... If you think it is a folders issue, are you using cache_folders ? This will make sure the folder structure is provided from memory and not from disk.. Actually meant to avoid disks spinning up when folder browsing.. Also check if turning on "spin up groups" helps, it might (but is default on) Link to comment
beckp Posted August 15, 2012 Share Posted August 15, 2012 Looking at your syslog you are having a lot of Unraid shfs: shfs_setxattr: setxattr: ... Operation not supported messages. See similar problem in the following link for a solution. Looks like 5.0b10 fixed it. http://lime-technology.com/forum/index.php?topic=5074.0 Link to comment
napalmd Posted August 17, 2012 Author Share Posted August 17, 2012 I'm a litle afraid about using a beta version on a running server with valuable data.. Today some new issues appear while copying files from a compact flash card to the server. after a while windows says it cannot access network share. if I try again later it continues copying. If I check the properties of the target folder it has the read only attribute semi-checked, I clear it out and then try again the Windows 7 - Error 0x8007003B: An unexpected network error occurred, show's up but then It I click try again and it continues copying fine again. Speed is about 6-9MB/s. if I copy to local drive I get around 20MB/s reading from the Flash card. So this is how slow the Unraid is. Uploaded with ImageShack.us Uploaded with ImageShack.us Link to comment
pantner Posted August 17, 2012 Share Posted August 17, 2012 are you using the onboard network card? or an addon card? i noticed these lines Jul 12 18:21:53 Unraid kernel: e100: Intel® PRO/100 Network Driver, 3.5.24-k2-NAPI Jul 12 18:21:53 Unraid kernel: e100: Copyright© 1999-2006 Intel Corporation Jul 12 18:21:53 Unraid kernel: Broadcom NetXtreme II CNIC Driver cnic v2.0.1 (Oct 01, 2009) the Intel Pro/100 is only a 10/100 network card. I would also check that all cards, cables and switches are gigabit compatible. If your network is only running at 100Mbps you will not get any higher than 10Mb/sec How fast is it to copy a large file off the server onto the computer? Link to comment
napalmd Posted August 17, 2012 Author Share Posted August 17, 2012 I'm using onboard gigabit marvel 88E8001 chip. Jul 12 18:21:53 Unraid kernel: skge 0000:02:05.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22 Jul 12 18:21:53 Unraid kernel: skge 1.13 addr 0xfbffc000 irq 22 chip Yukon rev 1 Jul 12 18:21:53 Unraid kernel: skge eth0: addr 00:0e:a6:20:a8:21 don't know why those e100 lines appear. I get around 30MB/s reading and 10MB/s writing a 1Gb file Link to comment
dgaschk Posted August 18, 2012 Share Posted August 18, 2012 Enter the following: ifconfig ethtool eth0 hdparm -tT /dev/[hs]d? Post the output. Link to comment
napalmd Posted August 20, 2012 Author Share Posted August 20, 2012 Hi the hdparm command can't do right now, takes a long time to show anything must be because I'm copying 30gb of photos from a laptop to the server and the rate is 7MB/s. also is this normal? uptime 12:57:42 up 2 days, 21:00, 2 users, load average: 18.09, 17.14, 11.81 root@Unraid:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:0e:a6:20:a8:21 inet addr:192.168.10.40 Bcast:192.168.10.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:101332595 errors:0 dropped:0 overruns:0 frame:0 TX packets:135087177 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:3243497874 (3.0 GiB) TX bytes:1137344453 (1.0 GiB) Interrupt:22 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:487 errors:0 dropped:0 overruns:0 frame:0 TX packets:487 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:39448 (38.5 KiB) TX bytes:39448 (38.5 KiB) root@Unraid:~# ethtool eth0 Settings for eth0: Supported ports: [ TP ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Full Advertised auto-negotiation: Yes Speed: 1000Mb/s Duplex: Full Port: Twisted Pair PHYAD: 0 Transceiver: internal Auto-negotiation: on Supports Wake-on: pg Wake-on: g Current message level: 0x00000037 (55) Link detected: yes Link to comment
dgaschk Posted August 20, 2012 Share Posted August 20, 2012 Hi the hdparm command can't do right now, takes a long time to show anything must be because I'm copying 30gb of photos from a laptop to the server and the rate is 7MB/s. also is this normal? uptime This rate may be normal for the laptop. Link to comment
napalmd Posted August 20, 2012 Author Share Posted August 20, 2012 that might be true, because that laptop only has a 100mbit lan, but copying to other servers/computers from this lapotop the rate is 11MB/s... here is the hdparm: root@Unraid:~# hdparm -tT /dev/[hs]d? /dev/sda: Timing cached reads: 1768 MB in 2.00 seconds = 883.85 MB/sec Timing buffered disk reads: 262 MB in 3.02 seconds = 86.72 MB/sec /dev/sdb: Timing cached reads: 1776 MB in 2.00 seconds = 888.07 MB/sec Timing buffered disk reads: 258 MB in 3.00 seconds = 85.99 MB/sec /dev/sdc: Timing cached reads: 1784 MB in 2.00 seconds = 892.18 MB/sec Timing buffered disk reads: 254 MB in 3.01 seconds = 84.47 MB/sec /dev/sdd: Timing cached reads: 1796 MB in 2.00 seconds = 897.79 MB/sec Timing buffered disk reads: 264 MB in 3.00 seconds = 87.98 MB/sec /dev/sde: Timing cached reads: 1762 MB in 2.00 seconds = 880.80 MB/sec Timing buffered disk reads: 324 MB in 3.01 seconds = 107.63 MB/sec /dev/sdf: Timing cached reads: 1672 MB in 2.00 seconds = 836.09 MB/sec Timing buffered disk reads: 344 MB in 3.02 seconds = 114.06 MB/sec /dev/sdg: Timing cached reads: 1702 MB in 2.00 seconds = 851.42 MB/sec Timing buffered disk reads: 48 MB in 3.02 seconds = 15.88 MB/sec Link to comment
dgaschk Posted August 20, 2012 Share Posted August 20, 2012 Make sure that the parity drive is on the MB. The PCI bus will only support a single drive with good speed. Parity and data drives are written concurrently. If they are both on the PCI bus it will slow things down. Link to comment
napalmd Posted August 20, 2012 Author Share Posted August 20, 2012 Yes it is. the MB only has 2 sata ports and one is parity drive. I've read the faq about this speed issue and I don't know what else to do but upgrade MB/CPU... Link to comment
JonathanM Posted August 20, 2012 Share Posted August 20, 2012 I don't think the CPU power is the issue, I think it's the shared PCI bus. Since that motherboard has no way to use a faster disk interface, a motherboard upgrade would seem to be the only solution. If you could get away with 3 or 4 total drives, the old motherboard would probably run ok. Link to comment
Helmonder Posted August 21, 2012 Share Posted August 21, 2012 That is testable though... Create a share that only uses one drive, spin the rest down and measure ... If you are right then speed should be increased substantially. Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.