March 8, 201511 yr Hi, first of all I-m on 4.7 still (yeah I know), but ther- no forum for it and it-s kind of a generic question so here goes... I realized I screwed up my split level for TV Shows (or rather my folder creations..( since it's set to 2 but I had been creating season folders directly (not series --> season). So now I want to fix it, and thought I would just create the series folder and move content into it and that should work. So now I have: TV Shows | ---> Show X - Season 1 ---> Show X - Season 2 etc. I wil do: TV Shows | Show X | -----> Show X - Season 1 -----> Show X - Season 2 etc. So the questions are: 1) I tried creating the base folder in windows, and then MOVING the season folder into it. While Technically the base folder will get created on any disk and then the season folder would be created/mapped to a single disk now, after doing that I didn't see any hdd activity -at all-... Now I know that normally when you move files Windows will just move the pointers for it, but here the files are physically on different disks, so what gives..? Due to the above, my next one changed the process to create the base folder, COPY the season folder, and then delete the original season folder. But doing this.... 2) I am getting pretty crappy speeds, even though I am on wired and Gb. This is topic for a different post, but what I want to know is: I understand that when you copy on a windos machine the files would come from the source disk, pass through my PC go back to the dest disk to be written. Now, is there any tool, menu , or whatever to avoid this unnecessary traffic being I-m not actually wanting to send new files t the server but rather move them between it-s own disks? Im-m taling about something that while I would have a nice gui, I-m actually operating within the server itself (like a terminal server / rdp session would do for example). Thanks!
March 8, 201511 yr You can use midnight commander (although honestly I'm not sure if it comes with 4.7 -> before my time) login at the console or telnet and type mc http://lime-technology.com/forum/index.php?topic=38603.msg359068#msg359068 Just make sure that you're not moving from disk shares to user shares (there's a bug when you do this).
March 8, 201511 yr Author You can use midnight commander (although honestly I'm not sure if it comes with 4.7 -> before my time) login at the console or telnet and type mc http://lime-technology.com/forum/index.php?topic=38603.msg359068#msg359068 Just make sure that you're not moving from disk shares to user shares (there's a bug when you do this). Thanks, I'll give it a try and make sure I'm moving from share to share. So this would not replicate windows behaviour? (When I do it in windows it does't really copy the files to the different disk though it seems). Now if I could only figure out why my (in theory) Gb setup is writing to the array at an average of <10 MB/s, that would really be nice...
March 9, 201511 yr Now if I could only figure out why my (in theory) Gb setup is writing to the array at an average of <10 MB/s, that would really be nice... Because you're reading from drive x, transferring it to windows, transferring it back to unraid, writing it to the same drive and updating parity at the same time Alot of overhead involved. Your mileage will vary depending upon all of the hardware involved however. If I move stuff around on my cache drive via Windows (no parity involved), I'm getting ~20-30MB/s. Just a straight copy to cache and I can hit ~70-80MB/s Before you start figuring out why these particular transfers are so slow, it would be good to see just how fast you can copy from unRaid to Windows, and then from Windows to unRaid in two separate steps.
March 9, 201511 yr Author Now if I could only figure out why my (in theory) Gb setup is writing to the array at an average of <10 MB/s, that would really be nice... Because you're reading from drive x, transferring it to windows, transferring it back to unraid, writing it to the same drive and updating parity at the same time Alot of overhead involved. Your mileage will vary depending upon all of the hardware involved however. If I move stuff around on my cache drive via Windows (no parity involved), I'm getting ~20-30MB/s. Just a straight copy to cache and I can hit ~70-80MB/s Before you start figuring out why these particular transfers are so slow, it would be good to see just how fast you can copy from unRaid to Windows, and then from Windows to unRaid in two separate steps. Uhm.. No.. That's not the case. Sorry I should have clarified. While what ypu say is true, this last post was just referencing slow speed in general, straight from PC to unraid, not the other thing. I am getting an average of roughly 10MB/s, with occasional rises to 15 tops. Reading is ok, and numbers are around the 70MB/s region...
March 9, 201511 yr ... I'll give it a try and make sure I'm moving from share to share ... NO -- You do NOT want to do a share to share (assuming you're referring to YOUR shares) ... you want to reference the disks (these are also "shares" ... but they're the "disk shares" -- not "user shares"). i.e. use references like \\Tower\disk2\Movies ... NOT \\Tower\Movies
March 9, 201511 yr I am getting an average of roughly 10MB/s, with occasional rises to 15 tops. Reading is ok, and numbers are around the 70MB/s region... You may have a problem with one of the wires in your Ethernet cable or one of the connectors. Try a different Ethernet cable (preferably a new Cat-6 cable).
March 9, 201511 yr Author I am getting an average of roughly 10MB/s, with occasional rises to 15 tops. Reading is ok, and numbers are around the 70MB/s region... You may have a problem with one of the wires in your Ethernet cable or one of the connectors. Try a different Ethernet cable (preferably a new Cat-6 cable). I can definitely try that (and will), but my logic is that if read speed is ok, then the unraid cable's Tx pair and my PC's Rx pair, have to be OK. That being the case, if I were to swap cables at the Computer end, that would efectively confirm or rule out a problem with the opposite pairs. I need to try more things out, but my gut tells me it may just be a crappy controller on the unRAID box... (Just checked, it was Realtek...)
March 9, 201511 yr I have a mobo and a switch that occasionally fall back to 100Mbit. Then I have to reset the switch and they will negotiate for Gbit again. In other terms it's working pretty well, just this one thing will cost it's life one day.
March 9, 201511 yr Any continuity issue in a wire or a connector can cause the speed to revert to 100Mb. I've seen bad switch ports; bad connectors; and bad cables all have this effect. The vast majority of the time it's either the cable or the connector (sometimes it's got a bit of pressure on it that, when relieved, will result in a return to Gb speed.
June 20, 201511 yr Author Hi Folks, Resuming this issue, I really appreciate everyone's input and apologize for the delay in replying, but sometimes other things take priority and well, you know... Even though I know there is nothing wrong with the switch or it's ports, specially because if in one direction it works, in the other it should as well with such a simple test as inverting the cables/ports, but still: I bought a brand new Cat6 cable, and I have just tested connecting the 2 computers *directly*: i.e. Computer "A" is directly connected to the unRAID server. Once again speed is the same (And at moments worse) and averages at about 9-10MB/sec... So at this point I am putting the focus on one of these two: 1) There is something wrong with my server (be it hardware or configuration) 2) The NIC on the server motherboard is just not doing it's job... What throws me off the most is that in the past long time ago I *did* copy files to my server at Gb speed (averaging between 40-70MB/sec) so I really don't know what could be happening Can you suggest things to check to try to pinpoint the culprit of the slow speeds?
June 20, 201511 yr Author Show the outputs of "ifconfig" and "ethtool eth0" Ok, this is it: eth0 Link encap:Ethernet HWaddr 6c:f0:49:6c:db:65 inet addr:192.168.0.7 Bcast:192.168.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:64823920 errors:0 dropped:0 overruns:0 frame:0 TX packets:7624635 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:2688330116 (2.5 GiB) TX bytes:546470681 (521.1 MiB) Interrupt:26 Base address:0xc000 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:80 errors:0 dropped:0 overruns:0 frame:0 TX packets:80 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:6896 (6.7 KiB) TX bytes:6896 (6.7 KiB) 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: 1000Mb/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
November 10, 201510 yr Since the topic has changed over the course of the thread, what exactly are you talking about now? Post a syslog that includes the timeframe of the behavior you are concerned with.
November 10, 201510 yr Author Just the speed issue, last is what I was requested to post. Write speed is constant at about 10Mb/s still
November 10, 201510 yr Just the speed issue, last is what I was requested to post. Write speed is constant at about 10Mb/s still Do you mean you have already posted the syslog I requested? I don't see it anywhere in this thread. Now that I have looked at your post history I see that you have also been posting in v6 support. Do you have multiple servers on different versions? Are you aware that the forums are mostly just users helping other users for free?
November 10, 201510 yr Author Just the speed issue, last is what I was requested to post. Write speed is constant at about 10Mb/s still Do you mean you have already posted the syslog I requested? I don't see it anywhere in this thread. Now that I have looked at your post history I see that you have also been posting in v6 support. Do you have multiple servers on different versions? Are you aware that the forums are mostly just users helping other users for free? Yes, well aware of it although sometime some of the nice folks of lime-tech also chime in. Anyway I was refering to the post immediately before today's where I had posted the last request for something specific. I'll get the syslog as soon as I get home, do you think that would show something relative to speed? Posted in v6 because I am going to upgrade but still on 4.7 Thanks!!
Archived
This topic is now archived and is closed to further replies.