January 11, 200917 yr Hello, I'm using 4.4-beta2. I decided to add another hdd and a cache drive to my unraid pro server. When I'm writing directly on cache drive or a user share its too slow. It's between 7-12MB. Reads are like 40MB however. I don't know what the problem is. NIC info (from ethtool) Settings for eth0: Supported ports: [ MII ] 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: MII PHYAD: 1 Transceiver: external Auto-negotiation: on Supports Wake-on: g Wake-on: d Link detected: yes NIC driver info (from ethtool -i) driver: forcedeth version: 0.61 firmware-version: bus-info: 0000:00:08.0 Ethernet config info (from ifconfig) eth0 Link encap:Ethernet HWaddr 00:1A:4D:84:8C:8D inet addr:192.168.1.7 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1 RX packets:7098384 errors:0 dropped:0 overruns:0 frame:0 TX packets:4266173 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1059552922 (1010.4 MiB) TX bytes:3305686272 (3.0 GiB) Interrupt:222 Base address:0x2000 (from /usr/bin/ps -eaf) UID PID PPID C STIME TTY TIME CMD root 1 0 0 05:53 ? 00:00:01 init root 2 0 0 05:53 ? 00:00:00 [kthreadd] root 3 2 0 05:53 ? 00:00:00 [migration/0] root 4 2 0 05:53 ? 00:00:00 [ksoftirqd/0] root 5 2 0 05:53 ? 00:00:00 [migration/1] root 6 2 0 05:53 ? 00:00:00 [ksoftirqd/1] root 7 2 0 05:53 ? 00:00:00 [events/0] root 8 2 0 05:53 ? 00:00:00 [events/1] root 9 2 0 05:53 ? 00:00:00 [khelper] root 54 2 0 05:53 ? 00:00:00 [kblockd/0] root 55 2 0 05:53 ? 00:00:00 [kblockd/1] root 56 2 0 05:53 ? 00:00:00 [kacpid] root 57 2 0 05:53 ? 00:00:00 [kacpi_notify] root 162 2 0 05:53 ? 00:00:00 [ksuspend_usbd] root 168 2 0 05:53 ? 00:00:00 [khubd] root 171 2 0 05:53 ? 00:00:00 [kseriod] root 217 2 0 05:53 ? 00:00:00 [pdflush] root 218 2 0 05:53 ? 00:00:05 [pdflush] root 219 2 0 05:53 ? 00:00:00 [kswapd0] root 263 2 0 05:53 ? 00:00:00 [aio/0] root 264 2 0 05:53 ? 00:00:00 [aio/1] root 273 2 0 05:53 ? 00:00:00 [cifsoplockd] root 274 2 0 05:53 ? 00:00:00 [cifsdnotifyd] root 440 2 0 05:53 ? 00:00:00 [scsi_eh_0] root 441 2 0 05:53 ? 00:00:00 [usb-storage] root 459 2 0 05:53 ? 00:00:00 [rpciod/0] root 460 2 0 05:53 ? 00:00:00 [rpciod/1] root 461 2 0 05:53 ? 00:00:00 [kirqd] root 526 1 0 05:53 ? 00:00:00 /sbin/udevd --daemon root 844 2 0 05:53 ? 00:00:00 [ata/0] root 847 2 0 05:53 ? 00:00:00 [ata/1] root 848 2 0 05:53 ? 00:00:00 [ata_aux] root 1053 2 0 05:53 ? 00:00:00 [scsi_eh_1] root 1054 2 0 05:53 ? 00:00:00 [scsi_eh_2] root 1056 2 0 05:53 ? 00:00:00 [scsi_eh_3] root 1058 2 0 05:53 ? 00:00:00 [scsi_eh_4] root 1060 2 0 05:53 ? 00:00:00 [scsi_eh_5] root 1062 2 0 05:53 ? 00:00:00 [scsi_eh_6] root 1232 1 0 05:53 ? 00:00:00 /usr/sbin/syslogd -m0 root 1237 1 0 05:53 ? 00:00:00 /usr/sbin/klogd -c 3 -x root 1635 1 0 05:53 ? 00:00:00 /usr/sbin/ifplugd -i eth0 -fwI -u0 -d10 root 1664 1 0 05:53 ? 00:00:16 /sbin/dhcpcd -d -t 30 -h Tower eth0 bin 1674 1 0 05:53 ? 00:00:00 /sbin/rpc.portmap nobody 1678 1 0 05:53 ? 00:00:00 /sbin/rpc.statd root 1686 1 0 05:53 ? 00:00:00 /usr/sbin/inetd root 1696 1 0 05:53 ? 00:00:00 /usr/sbin/acpid root 1698 1 0 05:53 ? 00:00:00 /usr/sbin/crond -l10 daemon 1700 1 0 05:53 ? 00:00:00 /usr/sbin/atd -b 15 -l 1 root 1717 1 0 05:53 ? 00:00:00 /usr/local/sbin/emhttp root 1941 2 0 05:53 ? 00:00:00 [mdrecoveryd] root 2080 1 0 05:53 tty1 00:00:00 /sbin/agetty 38400 tty1 linux root 2081 1 0 05:53 tty2 00:00:00 /sbin/agetty 38400 tty2 linux root 2084 1 0 05:53 tty3 00:00:00 /sbin/agetty 38400 tty3 linux root 2085 1 0 05:53 tty4 00:00:00 /sbin/agetty 38400 tty4 linux root 2088 1 0 05:53 tty5 00:00:00 /sbin/agetty 38400 tty5 linux root 2089 1 0 05:53 tty6 00:00:00 /sbin/agetty 38400 tty6 linux root 2353 1 0 06:30 ? 00:00:00 /bin/bash ./uu root 2354 2353 0 06:30 ? 00:00:00 awk -W re-interval -f ./unmenu.awk root 3681 1 0 07:01 ? 00:00:00 /usr/sbin/nmbd -D root 3683 1 0 07:01 ? 00:00:00 /usr/sbin/smbd -D root 3685 2 0 07:01 ? 00:00:00 [unraidd] root 3687 3683 0 07:01 ? 00:00:00 /usr/sbin/smbd -D root 3717 2 0 07:01 ? 00:00:00 [reiserfs/0] root 3718 2 0 07:01 ? 00:00:00 [reiserfs/1] root 3728 1 1 07:01 ? 00:00:09 /usr/local/sbin/shfs /mnt/user 1 2000000 root 3735 1 0 07:01 ? 00:00:00 /usr/local/sbin/shfs /mnt/user0 0 root 3827 3683 1 07:02 ? 00:00:07 /usr/sbin/smbd -D root 3985 2354 0 07:12 ? 00:00:00 gawk -v ConfigFile=unmenu.conf -v MyHost=Tower -v ScriptDirectory=. -v LocalConfigFile=unmenu_local.conf -v MyPort=8080 -W re-interval -f ./29-unmenu-sysinfo.awk STARTED GET /sys_info?option=Ps+info |Main|array_management|Array Mgmt|disk_management|Disk Mgmt|system_log|Syslog|myMain|myMain|links|Useful Links|usage|Disk Usage|dupe_files|Dupe Files|sys_info|System Info|file_browser|File Browser|user_scripts|User Scripts|package_manager|Package Manager|unraid_main|unRAID Main|about|About|help|Help| root 3987 3985 0 07:12 ? 00:00:00 sh -c /usr/bin/ps -eaf 2>&1 root 3988 3987 0 07:12 ? 00:00:00 /usr/bin/ps -eaf (from /sbin/lsmod) Module Size Used by md_mod 52428 3 forcedeth 45072 0 sata_nv 19976 5 libata 126816 1 sata_nv amd74xx 8584 0 [permanent] ide_core 85108 1 amd74xx
January 11, 200917 yr Thanks for all the info. Unfortunately, I could not see a single thing wrong! Not even anything that could be improved! Write speeds that bounce around between 7-12MB are fairly normal, when writing TO the User Share, without a Cache drive. But writing to the Cache drive directly, or to a User Share with a Cache drive installed, should be almost as fast as reading from your unRAID drives. There is nothing in your syslog or other info to explain why you are getting slow writes to the Cache drive. Did this syslog cover a period of slow writes to the Cache drive? You are sure the User Share is configured to use the Cache drive?
January 11, 200917 yr Author Thank you Rob. I restarted the system. Did a parity check. And now I have new data about the problem. When I start a file transfer, the first 30 seconds or so, the data rate is between 3-13MB but after 30 seconds to 1 minute it goes up to 40MB then falls down to 30 then goes up again then falls down to 25 etc. So its between 25MB-40MB range after 30 seconds. I don't remember unraid formatting my cache drive to reiserfs even though it says it is reiserfs. So my guess is that it is formatting it as I write data on the disc thats why it's slow. But I'm probably wrong Other than that all of my user shares are configured to use cache drive
January 11, 200917 yr Your syslog still looks very good, and very good performance on that parity check. The mover script kicked in earlier, only took 2 minutes to search for files and move one video, which seems pretty good, although I don't know how big the video was. The numbers you mention seem very good, with typical inconsistency at the beginning. I'm not sure why you think there is a problem. Writing is often somewhat 'spiky' as the data filters down through the various buffers and caches. It's better (and less confusing) to look at the overall average speed, by timing a transfer and dividing the size of the file by the seconds it took to transfer. As to the formatting, ReiserFS formats very fast, less than a minute per disk. Whereas FAT and NTFS file systems include a search of every sector for bad sectors when they format a drive, the Reiser file system skips that check, and does it on the fly. And on the Cache drive, there is no need for the disk to be cleared, so the prepping of a Cache drive finishes very quickly, and you could easily have missed it.
January 11, 200917 yr Author Thanks again Rob If these figures are the normal, than I think, I don't have to worry anymore. cheers
Archived
This topic is now archived and is closed to further replies.