Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

I am trying to isolate the root cause of slow transfer speeds from my host OS to a KVM client. Both are Ubuntu 12.04 LTS Linux.

Rsync on the host 192.168.1.72

rsync -auv --progress rut3.img  /tmp/   [54.09MB/s]

Rsync to the client:

rsync -auv --progress rut3.img  192.168.1.80:/tmp/ [25.52MB/s]

ifconfig:

br0       Link encap:Ethernet  HWaddr 00:24:21:e2:1d:36  
          inet addr:192.168.1.73  Bcast:192.168.1.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:11117216 errors:0 dropped:0 overruns:0 frame:0
          TX packets:22809214 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:9188723833 (9.1 GB)  TX bytes:25471033638 (25.4 GB)

eth0      Link encap:Ethernet  HWaddr 00:24:21:e2:1d:36  
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:6382610 errors:0 dropped:0 overruns:0 frame:0
          TX packets:6113095 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:9027466854 (9.0 GB)  TX bytes:529074209 (529.0 MB)
          Interrupt:42 Base address:0xa000 

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:3870749 errors:0 dropped:0 overruns:0 frame:0
          TX packets:3870749 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:99586308664 (99.5 GB)  TX bytes:99586308664 (99.5 GB)

rut3      Link encap:Ethernet  HWaddr 32:04:a8:24:23:eb  
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:2588596 errors:0 dropped:0 overruns:0 frame:0
          TX packets:11362124 errors:0 dropped:0 overruns:19613 carrier:0
          collisions:0 txqueuelen:500 
          RX bytes:174124246 (174.1 MB)  TX bytes:16828306882 (16.8 GB)

I realize that there will be some TCP overhead on the transfer but over 50%? Can someone enlighten me on what could be slowing down the transfers so much?

share|improve this question
    
please provide more information - what configs are used, OS types and versions etc. Also, comparing rsync of a file on the local disk and over network is not very correct –  dyasny Dec 20 '12 at 9:37
    
I have added more information. I am open to any metric that can get me a resolution but rsync is something I use a lot and easily displays bit rate. –  indian maiden Dec 20 '12 at 15:49
    
try to count physical to physical host transfer speeds, and compare it to a physical to virtual. Besides, you are on ubuntu, it is known to have some issues around the virt stack, if you can try on fedora, that would eliminate that part of the problem –  dyasny Dec 20 '12 at 17:05
    
Host to host is actually slower 17.31MB/s. I will do a fedora build over the weekend. If there are any custom recommendations please advise if not I will just do a generic setup. –  indian maiden Dec 21 '12 at 16:54
    
if host to host is slow, this has nothing to do with KVM –  dyasny Dec 22 '12 at 16:39

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.