I have been working on my new unraid server for some time now and finally have it sorta coming together.

It uses an sil3124 SATA card because I needed the port multiplier part.

 

I have 3 disks, 2tb parity, 2tb data and 1tb data.

 

I haven't run thru the parity sync yet, I thought I would be able to copy my data to it and run a parity sync after it was complete, but initial transfer speeds (over gigabit network) can start off at 30-35 megs a second, then can drop to just a few hundred kilobytes a sec and stay like that for a while, then shot back up to 35MB a sec, and do it all over again, this was over SMB or AFP

 

Would not having done the parity sync yet cause this?

It will do a parity sync all day at 40MB a sec and not miss a beat, and it will also read really fast, but writing is all over.  Ethtool shows a 1GB link and ifconfig shows no errors.

 

 

 

 

I just installed an Adaptec 1430SA PCI-E 4x card and am having a similar issue. I'm moving data from one of my old shares where the data is on the motherboard controller onto the PCI-E card and at times the copy drops right down to <1MB/Sec, then will speed right back up to ~35-50MB/Sec writing directly to the array. I figured it was just because I'm copying a folder full of small files (148000 files ~145gb) because most of the time I copy larger files over 1gb.

 

Edit: We recently ran an old CAT5e cable downstairs and moved the unRAID box down there, both ends are a gigabit switch but the cable is pretty old.. I checked ifconfig and noticed this:

 

root@Tower:~# ifconfig
eth0      Link encap:Ethernet  HWaddr 1c:6f:65:80:5f:a4
          inet addr:192.168.1.5  Bcast:192.168.1.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:112167945 errors:0 dropped:0 overruns:0 frame:0
          TX packets:116821682 errors:0 dropped:7453 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:4290389147 (3.9 GiB)  TX bytes:326380770 (311.2 MiB)
          Interrupt:32 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:1425 errors:0 dropped:0 overruns:0 frame:0
          TX packets:1425 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:107132 (104.6 KiB)  TX bytes:107132 (104.6 KiB)

root@Tower:~#

 

I'm guessing the dropped packets are because of a bad network connection? ie. Cable. Might have to pick up a long CAT6 from somewhere to replace it.

Not sure what your issue may be either.

 

After parity sync completed, copying files to the array was now fast and consistent.

 

 

chickensoup,

 

I have the same issue as well.  It was worse when I was using the built-in mobo lan.  I ended up buying a PCI intel NIC, and the speed has been staying above 20 MB/S now.  Some times, it will start at 55 MB/s, and slow down to 26 MB/s, which is a lot better than what I was getting before 11.5 MB/s.

 

When I finished transferring a file I get drop pockets as well, but not all the time.  It is sporadic.  I terminated my own cable, with the exception of the cable for my unraid server.  I bought a cable testing too, and all of my cables were fine.  I think I might end up buying a couple of Cat6 cables.  One for my W7 PC, and the other for my server and see if that changes anything.

 

The other thing I know might be acting up is my router's current FW.  I updated the FW a few weeks ago, and it started dropping the wireless connection all of the sudden.  I've flash my router with an older FW, and it seems it is working better.

 

I am thinking of trying out a different router and see if that helps...anyone can suggest a good router that can replace my WNDR3700 for a max price of $100 bucks.

I'm swapping out my shitty Netcomm router tomorrow for a new TP-Link one I've had sitting around for a bit although I don't think that will fix the issues since the two machines are copying between gigabit switches independant of the router. I'm pretty certain my problem is bad wiring or a bad NIC/switch but when I figure it out I'll let you know.

Yeah, I was messing around with my computer settings, and I found out that my W7 PC mobo nic is the realtek 8111C, which is problematic with UnRaid.  So, I am wondering if this is causing the issue for dropped pockets when I do transfers only?

I think it is only 8111E that had issues (with some of the v5 betas).  I have 8111D for example and no issues at all.  As far as I can see 8111C is also OK.

I think it is only 8111E that had issues (with some of the v5 betas).  I have 8111D for example and no issues at all.  As far as I can see 8111C is also OK.

 

Cool.  The jumbo frame were disabled in my W7 PC, which I just enabled, and I also did the MTU=9000 to my unraid, and see how that works out.

I enable jumbo frames for my W7 PC and server, and I noticed it was a little bit faster.  It constantly stayed at 35 MB/s.  I'm going to try out like this for a while, and see how it works out.  After I finish the transfer I am going to see if I've got any error / dropped pockets in eth0.

NIC info (from ethtool)

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: 1000Mb/s

Duplex: Full

Port: Twisted Pair

PHYAD: 0

Transceiver: internal

Auto-negotiation: on

MDI-X: Unknown

Supports Wake-on: umbg

Wake-on: g

Current message level: 0x00000007 (7)

Link detected: yes

 

NIC driver info (from ethtool -i)

driver: e1000

version: 7.3.21-k8-NAPI

firmware-version:

bus-info: 0000:02:06.0

 

Ethernet config info (from ifconfig)

eth0      Link encap:Ethernet  HWaddr 90:e2:ba:11:a3:ac 

          inet addr:192.168.1.2  Bcast:192.168.1.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:9000  Metric:1

          RX packets:2204338 errors:0 dropped:0 overruns:0 frame:0

          TX packets:929826 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:1000

          RX bytes:2479680770 (2.3 GiB)  TX bytes:97901211 (93.3 MiB)

 

 

Nice....no more errors, dropped, or frame errors at all.  Seems that it is working now.  Thanks.