June 24, 201214 yr I am not sure what to make of this. When I transfer files from my desktop to my unraid box I can no longer get speeds past 1.1 MB/sec. This just recently happened. I am running a Gigbit switch, and used to get speeds well over 22 MB/sec consistantly, sometimes as fast as 35 MB/sec. I tried a different ethernet cable from the switch to the server, which had no effect. I tried connecting the server to my router instead of the switch, and it would not connect at all. I ran ethtool and it shows I only have a 10Mb/s connection. This really has me baffled. I suspect my NIC may be going bad. Any thoughts? root@unRAID:~# ethtool eth0 Settings for eth0: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised auto-negotiation: Yes Speed: 10Mb/s Duplex: Full Port: MII PHYAD: 0 Transceiver: internal Auto-negotiation: on Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000033 (51) Link detected: yes root@unRAID:~# ifconfig eth0 Link encap:Ethernet HWaddr 90:e6:ba:a2:51:22 inet addr:192.168.1.141 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1 RX packets:110215 errors:0 dropped:0 overruns:0 frame:0 TX packets:62488 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:149577894 (142.6 MiB) TX bytes:6677368 (6.3 MiB) Interrupt:26 Base address:0xe000 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:270 errors:0 dropped:0 overruns:0 frame:0 TX packets:270 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:32121 (31.3 KiB) TX bytes:32121 (31.3 KiB)
June 24, 201214 yr Author The one I was using is cat5e, and the other one I tried is the same. As originally stated though it was working fine with the original cable, and then one day simply slowed down.
June 25, 201214 yr Are you using Win7 to transfer to your server? I had an issue when I updated the NIC driver on my machine. After updating I had to manually tell win7 that the NIC was GB and then reboot. Changed my speed from 7MB/s to 32MB/s.
June 25, 201214 yr Author My PC is dual boot with both Win 7, and Linux Mint 13. I get the same transfer speed to the unraid server no matter which OS I am in. Either one maxes out at 1.1 MB/sec.
June 25, 201214 yr My PC is dual boot with both Win 7, and Linux Mint 13. I get the same transfer speed to the unraid server no matter which OS I am in. Either one maxes out at 1.1 MB/sec. have you rebooted your router?
June 25, 201214 yr Author Joe good point. I had rebooted my Gigabit switch, but not the router. So I just tried it, and unfortunately it didn't change anything.
June 25, 201214 yr I'd try eliminating everything on the network first - i.e. use the switch and just have your unRAID and Win7 box plugged in, with nothing else (you'll probably need static IP's to achieve this if you don't use them already.) If the issue persists, I'd then remove the Win7 PC, and attach some other device (laptop maybe?) and test with that. If it still persists, then it's likely the unRAID end. It *could* also be the switch, but you appear to have eliminated that by trying both that and your router above. I once had a friend with exactly the same issue - it turned out to be that they'd just bought a Siemens VoIP phone, and the base station for that was only capable of a 10baseT/half duplex connection. I'm guessing the issue was actually his switch, as this in turn caused everything to negotiate at the same speed. Removed the phone and everything went back to normal. It was resolved by replacing the switch in the end.
June 25, 201214 yr Author BetaQuasi, per your suggestion I just tried removing other devices from my network. I turned the wifi for my phone off, and I shut off my tablet. Unfortunately nothing changed. I'll try the rest of your ideas over the next night or two. I'm going to check at work to see if I can borrow a NIC card to try in the server as well. I don't have any here at home to try as my HTPC and my regular PC both have built in on the MB. I appreciate the tips. Keep them coming.
June 25, 201214 yr Author Ok I tried this just for the heck of it: ethtool -s eth0 speed 100 duplex full This should have forced the NIC to go to 100 Mb/sec, and it didn't. Results are below. root@unRAID:~# ethtool eth0 Settings for eth0: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised auto-negotiation: Yes Speed: 10Mb/s Duplex: Full Port: MII PHYAD: 0 Transceiver: internal Auto-negotiation: on Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000033 (51) Link detected: yes root@unRAID:~# ethtool -s eth0 speed 100 duplex full root@unRAID:~# root@unRAID:~# ethtool eth0 Settings for eth0: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised auto-negotiation: Yes Speed: 10Mb/s Duplex: Full Port: MII PHYAD: 0 Transceiver: internal Auto-negotiation: on Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000033 (51) Link detected: yes
June 25, 201214 yr Make sure you try different ports on the router and switch too. If you check the speeds on the other machines, you should be able to isolate the issue to just the segment between UnRAID machine and switch. Sounds like you may need either a new cable or a new NIC in your UnRAID machine.
June 25, 201214 yr Author Rob I have tried plugging the unraid box into different ports on the switch. I've also tried plugging it directly into the router, but for some reason it won't connect when I do that. When I was swapping stuff around I moved the main PC from the switch to the router. I ran ethtool on the main PC, and it comes back as 100 Mb/sec, which is correct for my router.
June 26, 201214 yr Author What about directly from unRAID to the NIC on your PC, just out of curiosity? I just looked up the procedure for doing this, and it fairly involved. Considering my NIC refuses to connect to my router, and at this point will only connect with my switch I'm feeling pretty positive that the NIC has an issue, so before going through an involved process only to find it likely won't work I think I'm going to just bite the bullet, and buy a new NIC card (didn't have one at the office to try out). Once I've done that, and tested it I'll post back. Thank you everyone for your assistance. I'll let you know how I make out.
June 26, 201214 yr I used to have a TrendNet 8 port gigabit switch, and at times, whenever I rebooted my server, it would got to 100mb/s only. I had a mixture between 100 mb/s and gigabit devices. I ended up putting the 100 mb/s devices from port 1 to 3, and the gigabit device from 4 to 8. This seems to fix the issue or so I thought. Whenever I rebooted my server for some reason or another it would go back to 100 mb/s. I ended up buying a Inter Pro gibabit NIC card, and the issue was still present whenever I booted my Server. I had a NetGear gigabit 8 port switch, which replaced my TrendNet and problem was solved. The trendnet 8 port gigabit switch was brand new so I don't understand why it had that issue. I had a 5 port gigabit trendnet switch, which I still use without any issues. Oh, I still use the intel NIC by the way. I also ended up replacing all of my Cat5e cables with Cat6, with the exception of one since its almost 100 feet long.
June 26, 201214 yr Considering my NIC refuses to connect to my router, and at this point will only connect with my switch I'm feeling pretty positive that the NIC has an issue ... Why do you believe that it is the NIC which has an issue ... why not the router?
June 28, 201214 yr Author Peter I suspect the NIC because all of my other devices are working correctly, but sure I suppose it could be the router. I just did a 30/30/30 reset on the router, and still my unraid box refuses to connect to it when plugged in directly. I plugged the unraid box back into the switch, and I am now working again, but still at the crippled speed. I have not ordered a new NIC yet as I have not completely given up, but the only thing left to try is directly connecting the unraid box to my PC, and for that it seems I have to edit files on the unraid thumb drive. I have been pushing off trying that, and am not sure yet if I will do that, or try a new NIC first. I will keep you posted.
July 1, 201214 yr Author Today's update. As you know I tried this ethtool command before, but it failed to effect NIC's speed. I figured trying to force things up one level seemed logical, but absolutely nothing happened. ethtool -s eth0 speed 100 duplex full So I tried another tool called net-tools, and ran this command. mii-tool -F 100baseTx-HD This time the server was booted off the network. I rebooted, and it connected back at 10Mb/s. I know that this point I can't be sure that the NIC is bad, but with multiple other devices running fine on the network it seems logical. I've decided I'm going to purchase a new NIC, and see what happens. I will post back with the results good or bad. Edit: I ordered this card from Newegg - (Intel EXPI9301CTBLK 10/ 100/ 1000Mbps PCI-Express Network Adapter Item #: N82E16833106033). I was going to go with this card - (Intel PWLA8391GT PRO/1000 GT Desktop Adapter 10/ 100/ 1000Mbps PCI 1 x RJ45 - OEM Item#: N82E16833106123) since the Pro/1000 is on the known working card list, and since it wouldn't take up a PCI-e slot, but I changed my mind right before placing the order. I figure my server is currently only using 4 drives, and I found out that my motherboard (Asus M4A785-M) has compatibility issues with the Supermicro AOC-SASLP-MV8 card. So realistically my current setup isn't going over 6 drives anyway, and I figure a newer model PCIe card can be re-used in another system down the road. It's shelf life ought to be a bit better. The Pro/1000 is a pretty old card at this point.
July 5, 201214 yr Do you have any progress on this? I have the same thing happening to me. It says its going to take weeks to transfer my media from my Lacie Network Space 2 NAS 2TB to my unraid server. I was thinking it because its still somehow being slowed down by win7 ultimate. I drag the files from the network share from the NAS to Uraid and on my laptop I can see the transfer speed/rate. I though I could do it over the NAS's web interface but I cant, I cant only copy to other folders on the NAS. Could this be a reason? I also install my parity drive BEFORE transferring my media across, would that slow it down so all that much? I have read the thread from the start and googled for the answer but im stumped (would it be worth going from 4.7 to unRAID Server Release 5.0-rc5-r8168 as I am using a Realtek NIC)
July 5, 201214 yr I hate to say this but I too am getting this issue. Just started today while trying to copy some files from it. I'm running 5.0rc3 and this just started happening, I was getting 20-30mb copying to and from the server last month. I'm also using a realtek nic. I'm afraid to upgrade to the 5.0-rc5-r8168 release without a valid recent backup and at this speed unraid 7.0 will be released by the time it finishes. Also just to add, only one drive was spun up when doing this recent copy when the issue came up. Last monthly parity check with all drives was on the 1st and it took 8 hours to complete. I don't think it's the server. I can copy to other windows shares super fast from the same machine so I know it's not the machines issue.
July 6, 201214 yr I hate to say this but I too am getting this issue. Just started today while trying to copy some files from it. I'm running 5.0rc3 and this just started happening, I was getting 20-30mb copying to and from the server last month. I'm also using a realtek nic. I'm afraid to upgrade to the 5.0-rc5-r8168 release without a valid recent backup and at this speed unraid 7.0 will be released by the time it finishes. Also just to add, only one drive was spun up when doing this recent copy when the issue came up. Last monthly parity check with all drives was on the 1st and it took 8 hours to complete. I don't think it's the server. I can copy to other windows shares super fast from the same machine so I know it's not the machines issue. Can you post a syslog please?
July 6, 201214 yr I've noticed it will take 1-2 minutes for the web interface to respond now, though keeping a constant ping on it, it doesn't lose any replies. I don't mean to take over the thread, I'm just stating I'm seeing this issue like the OP and it's out of the blue. syslog-2012-07-06.txt
July 6, 201214 yr I've noticed it will take 1-2 minutes for the web interface to respond now, though keeping a constant ping on it, it doesn't lose any replies. I don't mean to take over the thread, I'm just stating I'm seeing this issue like the OP and it's out of the blue. Your log has a bunch of Jun 30 03:26:01 NAS2 crond[1238]: failed parsing crontab for user root: cron="" messages which is a known bug with the addon SimpleFeatures, see the following thread if you want that to go away (workaround required each boot until newer version is released): http://lime-technology.com/forum/index.php?topic=20689.msg183461#msg183461 Can you please post the output of: ethtool eth0 and ifconfig The syslog says you're running an Intel PCI-E 4x PRO/1000 NIC not a Realtek, is that correct?
July 6, 201214 yr Interesting...on the 27th I shut the server down and installed an Intel nic to switch over to "eventually", but it's not connected with a cable. I'm still using the on-board realtek. I'm at work so I can't run the commands until I get home. Is it using an Intel driver for that realtek nic, could that the issue? Guess I should remove it and see if my issues go away. I didn't think it would have a problem with the other nic installed just sitting there but it could explain the issue I guess. The crontab issue isn't bothering me, I knew where it was coming from but it doesn't seem to be affecting anything so I ignore the issue.
July 6, 201214 yr Interesting...on the 27th I shut the server down and installed an Intel nic to switch over to "eventually", but it's not connected with a cable. I'm still using the on-board realtek. I'm at work so I can't run the commands until I get home. Is it using an Intel driver for that realtek nic, could that the issue? Guess I should remove it and see if my issues go away. I didn't think it would have a problem with the other nic installed just sitting there but it could explain the issue I guess. The crontab issue isn't bothering me, I knew where it was coming from but it doesn't seem to be affecting anything so I ignore the issue. I'm not sure sorry, I tried searching your log for realtek, rtl, 8111, 8168, 8169 and got nothing, so i searched for eth and found: ... Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.0: eth0: (PCI Express:2.5GT/s:Width x4) 00:1f:29:57:f8:5c Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.0: eth0: Intel(R) PRO/1000 Network Connection Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.0: eth0: MAC: 0, PHY: 4, PBA No: D51930-004 Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.1: Disabling ASPM L1 Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.1: PCI INT B -> GSI 17 (level, low) -> IRQ 17 Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.1: setting latency timer to 64 Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.1: irq 48 for MSI/MSI-X Jun 28 12:28:55 NAS2 kernel: scsi10 : usb-storage 1-5:1.0 Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.1: eth1: (PCI Express:2.5GT/s:Width x4) 00:1f:29:57:f8:5d Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.1: eth1: Intel(R) PRO/1000 Network Connection Jun 28 12:28:55 NAS2 kernel: e1000e 0000:02:00.1: eth1: MAC: 0, PHY: 4, PBA No: D51930-004 ... Jun 28 12:28:57 NAS2 kernel: e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx ... I might need someone with a bit more expertise to decipher what is going on but to me it looks like eth0 & eth1 are both the same Intel NIC? The "active" NIC appears to be eth0, which, as unRAID states is an Intel. Edit: I would recommend disabling the onboard LAN in the BIOS and moving your network cable to the Intel NIC, then see how it goes from there.
Archived
This topic is now archived and is closed to further replies.