Re: [Xen-users] Slow networking problem in dom0

[Emil Christopher Melar <typeof@xxxxxxxxx>]

Tried swapping card slot. It got a different IRQ, but the problem remained.

It seems like XEN 4.0.1 has a problem with IRQ sharing on my H67 chipset.

I found a workaround: http://lists.xensource.com/archives/html/xen-users/2006-06/msg00176.html

I now have 'noirqdebug' as a kernel param, and it has been fine for 24 hours. This is no ideal solution though.

On Fri, Jun 10, 2011 at 9:00 PM, Emil Christopher Melar <typeof@xxxxxxxxx> wrote:

I am now running the 2.6.32.40 kernel, but I  still use XEN 4.0.1. Still the same problems.
However I discovered something:

dmesg:

[2011-06-10 19:31:48] irq 19: nobody cared (try booting with the "irqpoll" option)

[2011-06-10 19:31:48] Pid: 0, comm: swapper Not tainted 2.6.32.40 #1

[2011-06-10 19:31:48] Call Trace:

[2011-06-10 19:31:48]  <IRQ>  [<ffffffff810c14a2>] __report_bad_irq+0x38/0x87

[2011-06-10 19:31:48]  [<ffffffff810c1603>] note_interrupt+0x112/0x177

[2011-06-10 19:31:48]  [<ffffffff810c1cbb>] handle_fasteoi_irq+0x97/0xbc

[2011-06-10 19:31:48]  [<ffffffff812f8b8d>] __xen_evtchn_do_upcall+0x1bf/0x287

[2011-06-10 19:31:48]  [<ffffffff812f94c2>] xen_evtchn_do_upcall+0x32/0x47

[2011-06-10 19:31:48]  [<ffffffff8103de7e>] xen_do_hypervisor_callback+0x1e/0x30

[2011-06-10 19:31:48]  <EOI>  [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x1000

[2011-06-10 19:31:48]  [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x1000

[2011-06-10 19:31:48]  [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x1000

[2011-06-10 19:31:48]  [<ffffffff81038d7b>] ? xen_safe_halt+0x10/0x1a

[2011-06-10 19:31:48]  [<ffffffff81044127>] ? default_idle+0x3f/0x6a

[2011-06-10 19:31:48]  [<ffffffff8103bde6>] ? cpu_idle+0x66/0xa4

[2011-06-10 19:31:48]  [<ffffffff81602249>] ? rest_init+0x6d/0x6f

[2011-06-10 19:31:48]  [<ffffffff81a18dd5>] ? start_kernel+0x43c/0x447

[2011-06-10 19:31:48]  [<ffffffff81a182c1>] ? x86_64_start_reservations+0xac/0xb0

[2011-06-10 19:31:48]  [<ffffffff81a1caca>] ? xen_start_kernel+0x5b7/0x5be

[2011-06-10 19:31:48] handlers:

[2011-06-10 19:31:48] [<ffffffff813b5ff4>] (e1000_intr+0x0/0xf7)

[2011-06-10 19:31:48] Disabling IRQ #19

After this, it goes slow. I tried disabling everything I could (onboard NIC) but nothing happens. I  also tried irqpoll in kernel params.

e1000 happens to be my intel NIC which I use.

On Sat, May 14, 2011 at 6:54 PM, Joel Witherspoon <joel.witherspoon@xxxxxxxxx> wrote:

So,
You're running IPv6 and IPv4 on the same interface? Can you do without IPv6?

Joel C. Witherspoon

(626) 502 - 7201

joel.witherspoon@xxxxxxxxx

My profiles:

Latest tweet: For the Nintendo and Twitter lovers out there: The Fail Blooper. http://www.cubeecraft.com/blog/fail-blooper/ Follow @joelwitherspoon Reply Retweet   13:37 May-11

  Get this email app!  

Signature powered by WiseStamp 

On Sat, May 14, 2011 at 9:49 AM, Emil Christopher Melar <typeof@xxxxxxxxx> wrote:

On Sat, May 14, 2011 at 6:14 PM, Erwan RENIER <erwan.renier@xxxxxxxxxxx> wrote:
> Did you try tcpdump  allready to examine it with wireshark, to see i there
> are problems with the flows ( like retransmitted paquets ) , i had this
> problem on virtualbox that maid the network very slow.

My bridge setup is pretty standard. Speed is OK right after boot. But
over time (hours) it drops to 500-600 k/sec. My setup is stated in the
first post.
tcpdump didn't work:

> Another option is to not bridge but route the network, at least to see if it
> works.
> Also did you check your stats on the interfaces and on the bridge for errors
> ?

tcpdump doesn't say anything unusual when it comes to ICMP at least.
Any suggestions to what to test on?
I did  sudo tcpdump -nnvvi xen-pub  icmp[0] = 8 or icmp[0] = 30.
I am clueless on what to test :-)

I can't change the network configuration as I am only on remote right
now. I just want to know if there's anything inherently wrong with my
kernel and / or bridge configuration.

> can you post the result of : ifconfig -a on the host.
> Is there any relevant logs ( dmesg ).
> Thanks.

eth0      Link encap:Ethernet  HWaddr 00:1b:21:8d:b1:27
         inet6 addr: fe80::21b:21ff:fe8d:b127/64 Scope:Link
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:69748631 errors:0 dropped:22 overruns:0 frame:0
         TX packets:20460331 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:1000
         RX bytes:87875949926 (87.8 GB)  TX bytes:21764017480 (21.7 GB)

eth1      Link encap:Ethernet  HWaddr bc:ae:c5:de:97:0d
         BROADCAST MULTICAST  MTU:1500  Metric:1
         RX packets:0 errors:0 dropped:0 overruns:0 frame:0
         TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:1000
         RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)
         Interrupt:18

lo        Link encap:Local Loopback
         inet addr:127.0.0.1  Mask:255.0.0.0
         inet6 addr: ::1/128 Scope:Host
         UP LOOPBACK RUNNING  MTU:16436  Metric:1
         RX packets:9862 errors:0 dropped:0 overruns:0 frame:0
         TX packets:9862 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:0
         RX bytes:705262 (705.2 KB)  TX bytes:705262 (705.2 KB)

sit0      Link encap:IPv6-in-IPv4
         NOARP  MTU:1480  Metric:1
         RX packets:0 errors:0 dropped:0 overruns:0 frame:0
         TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:0
         RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

vif4.0    Link encap:Ethernet  HWaddr fe:ff:ff:ff:ff:ff
         inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link
         UP BROADCAST RUNNING PROMISC MULTICAST  MTU:1500  Metric:1
         RX packets:59827 errors:0 dropped:0 overruns:0 frame:0
         TX packets:700342 errors:0 dropped:6 overruns:0 carrier:0
         collisions:0 txqueuelen:32
         RX bytes:11462334 (11.4 MB)  TX bytes:618194259 (618.1 MB)

virbr0    Link encap:Ethernet  HWaddr aa:5e:2e:b3:28:32
         inet addr:192.168.122.1  Bcast:192.168.122.255  Mask:255.255.255.0
         inet6 addr: fe80::a85e:2eff:feb3:2832/64 Scope:Link
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:0 errors:0 dropped:0 overruns:0 frame:0
         TX packets:26 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:0
         RX bytes:0 (0.0 B)  TX bytes:1628 (1.6 KB)

xen-dmz   Link encap:Ethernet  HWaddr 5a:d4:56:b2:a8:17
         inet6 addr: fe80::58d4:56ff:feb2:a817/64 Scope:Link
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:0 errors:0 dropped:0 overruns:0 frame:0
         TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:0
         RX bytes:0 (0.0 B)  TX bytes:468 (468.0 B)

xen-intern Link encap:Ethernet  HWaddr 02:99:ff:50:7b:22
         inet6 addr: fe80::99:ffff:fe50:7b22/64 Scope:Link
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:0 errors:0 dropped:0 overruns:0 frame:0
         TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:0
         RX bytes:0 (0.0 B)  TX bytes:468 (468.0 B)

xen-pub   Link encap:Ethernet  HWaddr 00:1b:21:8d:b1:27
         inet addr:192.168.0.3  Bcast:192.168.0.255  Mask:255.255.255.0
         inet6 addr: fe80::21b:21ff:fe8d:b127/64 Scope:Link
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:68302724 errors:0 dropped:0 overruns:0 frame:0
         TX packets:18407729 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:0
         RX bytes:86617595475 (86.6 GB)  TX bytes:22462523015 (22.4 GB)


Thanks!
--

Emil Christopher Solli Melar

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users

--
Emil Christopher Solli Melar

--
Emil Christopher Solli Melar

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users