December 6, 201411 yr Hi, new to the forum, and new to unraid overall (only running it about a week with some much needed help from other forum people) Over the course of the week, I've added 2 drives, 1 being parity. About 600gb was transferred over, no issues, at approximately 35m/sec. Great, super happy. 2 days ago, I purchased a drive to make as the cache (Seagate Barracuda ST1000DM003 1TB 64MB Cache SATA 6.0Gb/s), installed it, up and running, files are moved with mover, so on and so on, no problems... except for one. The write speed is 11m/sec! Not only that but it's 11m read AND write. All the cache drive posts I see mention 3-5X FASTER speeds than simply writing to the array without a cache drive, so what could be causing this issue? I've seen several other posts mention problems (some related to the mobo, some related to memory issues) so I'm not sure where I stand with this, but I'd greatly appreciate any help in resolving it. I should mention that there is a half a second spike up to 50m/sec when first transferring over a file, but it quickly declines down to the 11m-12m/sec range. I only noticed this spike because I was uploading through ftp, rather than windows explorer/teracopy. I realize this isn't the only post related to this issue. Below I've also posted disk speed test results taken from the option in unmenu/user scripts. Current build: Motherboard: GIGABYTE GA-H97N-WIFI LGA 1150 Intel H97 HDMI SATA 6Gb/s USB 3.0 Mini ITX Intel PSU: CORSAIR CX series CX430 430W ATX12V v2.3 80 PLUS BRONZE Certified Active PFC Power Supply PCU: Intel Core i3-4130 Haswell Dual-Core 3.4GHz LGA 1150 54W Desktop Processor Intel HD Graphics 4400 BX80646I34130 Flash for unRAID: CORSAIR Flash Voyager 8GB USB 2.0 Flash Drive Model CMFUSB2.0-8GB Memory: Kingston 4GB 240-Pin DDR3 SDRAM DDR3 1600 Desktop Memory Model KVR16N11S8/4 HD Parity: Western Digital Red NAS Hard Drive WD30EFRX 3TB IntelliPower 64MB Cache SATA 6.0Gb/s 3.5" NAS Hard Drive /dev/sdd: Timing cached reads: 14952 MB in 2.00 seconds = 7489.41 MB/sec Timing buffered disk reads: 436 MB in 3.01 seconds = 144.79 MB/sec HD Share: Western Digital Red NAS Hard Drive WD30EFRX 3TB IntelliPower 64MB Cache SATA 6.0Gb/s 3.5" NAS Hard Drive /dev/sdb: Timing cached reads: 15930 MB in 2.00 seconds = 7979.89 MB/sec Timing buffered disk reads: 428 MB in 3.00 seconds = 142.59 MB/sec HD Cache: Seagate Barracuda ST1000DM003 1TB 64MB Cache SATA 6.0Gb/s /dev/sdc: Timing cached reads: 15914 MB in 2.00 seconds = 7972.11 MB/sec Timing buffered disk reads: 574 MB in 3.01 seconds = 190.75 MB/sec Files being transferred to it are always from my windows 7 64bit home pc, using Samba because there's almost no support for NFS. UPDATE: After some more tinkering, I decided to see what could be causing this. Tried disabling cache_dirs, same speeds occur. Tried stopping the array and turning off the cache drive, leaving only the share drive and the parity, STILL just 11m read/write. This really makes no sense. Thousands of people have these 3tb red drives, they are definitely not 11m read/write drives.
December 6, 201411 yr The write speed is 11m/sec! Not only that but it's 11m read AND write.Confirm a good Cat5E or better gigabit connection for all points, server, switch, and client. It sounds like you got bumped back to 100Mbps at some point. If you have a laptop with a gigabit ethernet port, you should be able to put a cable directly from the server to the laptop and test speeds that way for a good baseline.
December 6, 201411 yr Author Yeah I'm slowly little by little realizing this is definitely a network issue. Work in progress.
December 6, 201411 yr I had a similar issue. I ended up replacing the NIC on the server and my speeds were back up to normal. Might be some thing to investigate if trying new network cables isn't having a positive effect.
December 7, 201411 yr Author As of right now, from various sources, if I "mc" into my server and try to transfer files from /mnt/cache into /mnt/user, or vice versa, I'm getting 30-40m/sec transfers. If I try and do transfers from my windows pc to one of the shares, I'm still bottlenecking at 11m/sec... so this would lead me to believe it's a network issue and not a drive issue. I'll have to tinker with new ethernet cables, and hopefully that'll fix the issue. Otherwise, I'm not sure where to try next (router settings?)
December 7, 201411 yr As of right now, from various sources, if I "mc" into my server and try to transfer files from /mnt/cache into /mnt/user, or vice versa, I'm getting 30-40m/sec transfers. If I try and do transfers from my windows pc to one of the shares, I'm still bottlenecking at 11m/sec... so this would lead me to believe it's a network issue and not a drive issue. I'll have to tinker with new ethernet cables, and hopefully that'll fix the issue. Otherwise, I'm not sure where to try next (router settings?) This is definitely a network issue and based on the consistent speed probably not a cable. When you say router, is it a gigabit router? Most routers in-built switch is 100MB/Sec. Check the specifications online if you aren't sure. To check your link speed, assuming eth0 is your network port, type: root@Tower:~# ethtool eth0
December 7, 201411 yr Author Looks as though it is definitely the router limiting me. Don't know how I've never noticed that before. Router is the Netgear WNR3500v2, though it claims it's a gigabit router. Knocked me down to 100? How would I reset that 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 pause frame use: No Advertised auto-negotiation: Yes Speed: 100Mb/s Duplex: Full Port: Twisted Pair PHYAD: 2 Transceiver: internal Auto-negotiation: on MDI-X: on Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000007 (7) Link detected: yes
December 7, 201411 yr Looks as though it is definitely the router limiting me. Don't know how I've never noticed that before. Router is the Netgear WNR3500v2, though it claims it's a gigabit router. Knocked me down to 100? How would I reset that 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 pause frame use: No Advertised auto-negotiation: Yes Speed: 100Mb/s Duplex: Full Port: Twisted Pair PHYAD: 2 Transceiver: internal Auto-negotiation: on MDI-X: on Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000007 (7) Link detected: yes Normally speed is auto-sensed. It could be just a port on the router going bad - you could try other ports. Also, sometimes it is the cable to the router rather than the router itself. Gigabit Ethernet has more stringent requirements on the cable than 100Mb Ethernet so sometimes swapping cables can help.
December 7, 201411 yr Author I'm leaning towards the cord or port, since I had original 30-40m/sec speeds. I'll swap and see. Thanks for the help all!
December 8, 201411 yr Author All fixed on the issue. Now running 1000mbps speeds - 120m read/write! Turns out, it was part bad cord/NIC, and part me plugging the ethernet into the APC UPS and then out to the NAS. Router really didn't like that, but seeing as I'm not powering any devices via ethernet, I probably shouldn't have even plugged it into the UPS to begin with. Thanks again for all the help.
Archived
This topic is now archived and is closed to further replies.