[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: RE: [Xen-users] 3.0.0 and gplpv



> I'm still keep to identify the problem. Have you done a TCP dump looking
> for bad packets or anything?

> tcpdump -s0 -v -n -i vifX.0

> where vifX.0 is the interface for your DomU

Well, this has been a fun week. A few days ago, I got a major kmail update. 
It's now fully converted over to using akonadi (ugh!). I thought I had 
everything converted over after a couple of days, but I just looked at my 
converted saved mails, and they are nothing but headers! Fortunately, kmail 
still keeps track of the originals.

Then yesterday, fedora rawhide put out kernel-3.1.0-0.rc0, and I got all 
excited. Hmm - it doesn't boot under the hypervisor. And when I tried to boot 
bare metal, it couldn't find modules.dep, or any kernel modules. Turns out 
dmesg and uname are advertising the kernel as 3.0.1-0.rc0. And yet, the new 
kernel config options are there:

4394a4423,4424
> CONFIG_XEN_SELFBALLOONING=y
> CONFIG_XEN_BALLOON_MEMORY_HOTPLUG=y
4405a4436,4437
> CONFIG_XEN_TMEM=y
> CONFIG_XEN_PCIDEV_BACKEND=m

that you would expect from kernel 3.1. A simple soft link in /lib/modules 
from:

3.0.1-0.rc0.git9.1.fc17.x86_64 -> 3.1.0-0.rc0.git9.1.fc17.x86_64

solved the bare metal boot problem. I just got another kernel-3.1.0-0.rc0 
update (from git9.1 to git11.2). Hopefully, fedora has sorted out it's kernel 
identity crisis!

Anyway, as promised, I did the tcpdump under dom0 3.0.0-1 on my winxp domu. 
The gplpv network interface died while I was asleep. What's interesting about 
it is it appears only tcp died. You'll see udp packets on 6646 (mcafee) and 
smb, but no responses to arp. The ifconfig output appears uninteresting. dom0 
syslog had two things of interest. My dom0 is setup as a local browser in 
smb.conf, but my winxp domu (inspxpvm) kept announcing itself as the local 
browser, as if it couldn't hear anyone else on the network. Finally, earlier 
(shortly after booting the domu), syslog reported a kmalloc poisoning.

I'll continue to try to get a dump closer to when xennet actually froze, and 
post that. Hope this was of interest.

tcpdump -s0 -v -n -i vif1.0 :

    192.168.1.102.6646 > 192.168.1.255.6646: UDP, length 1147
15:50:02.842410 IP (tos 0x0, ttl 128, id 401, offset 0, flags [none], proto 
UDP (17), length 255)
    192.168.1.102.netbios-dgm > 192.168.1.255.netbios-dgm: NBT UDP PACKET(138)
15:50:26.036034 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 
192.168.1.1 tell 192.168.1.102, length 28
15:50:30.029020 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 
192.168.1.1 tell 192.168.1.102, length 28
15:50:34.028941 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 
192.168.1.1 tell 192.168.1.102, length 28
15:50:41.040211 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 
192.168.1.1 tell 192.168.1.102, length 28
15:50:44.091453 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 
192.168.1.1 tell 192.168.1.102, length 28
15:50:50.106342 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 
192.168.1.1 tell 192.168.1.102, length 28
15:50:53.687827 IP (tos 0x0, ttl 128, id 414, offset 0, flags [none], proto 
UDP (17), length 198)
    192.168.1.102.6646 > 192.168.1.255.6646: UDP, length 170
15:50:55.522504 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 
192.168.1.1 tell 192.168.1.102, length 28
^C
196319 packets captured
375563 packets received by filter
179244 packets dropped by kernel
root@insp6400 07/30/11  3:51PM:~
[507] > (ifconfig vif1.0)
vif1.0    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:9000  Metric:1
          RX packets:184519 errors:0 dropped:0 overruns:0 frame:0
          TX packets:191061 errors:0 dropped:3 overruns:0 carrier:0
          collisions:0 txqueuelen:32
          RX bytes:150101405 (143.1 MiB)  TX bytes:211552022 (201.7 MiB)

two snippets from syslog:

Jul 30 16:26:05 Insp6400 nmbd[3035]: [2011/07/30 16:26:05.918094,  0] 
nmbd/nmbd_incomingdgrams.c:308(process_local_master_announce)
Jul 30 16:26:05 Insp6400 nmbd[3035]:   process_local_master_announce: Server 
INSPXPVM at IP 192.168.1.102 is announcing itself as a local master browser 
for workgroup LINKSYS and we think we are master. Forcing election.
Jul 30 16:26:05 Insp6400 nmbd[3035]: [2011/07/30 16:26:05.994876,  0] 
nmbd/nmbd_become_lmb.c:148(unbecome_local_master_success)
Jul 30 16:26:05 Insp6400 nmbd[3035]:   *****
Jul 30 16:26:05 Insp6400 nmbd[3035]:
Jul 30 16:26:05 Insp6400 nmbd[3035]:   Samba name server INSP6400 has stopped 
being a local master browser for workgroup LINKSYS on subnet 192.168.1.100
Jul 30 16:26:05 Insp6400 nmbd[3035]:
Jul 30 16:26:05 Insp6400 nmbd[3035]:   *****
Jul 30 16:26:23 Insp6400 nmbd[3035]: [2011/07/30 16:26:23.219147,  0] 
nmbd/nmbd_become_lmb.c:395(become_local_master_stage2)
Jul 30 16:26:23 Insp6400 nmbd[3035]:   *****
Jul 30 16:26:23 Insp6400 nmbd[3035]:
Jul 30 16:26:23 Insp6400 nmbd[3035]:   Samba name server INSP6400 is now a 
local master browser for workgroup LINKSYS on subnet 192.168.1.100
Jul 30 16:26:23 Insp6400 nmbd[3035]:
Jul 30 16:26:23 Insp6400 nmbd[3035]:   *****


Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] FIX kmalloc-2048: Restoring 
0xffff88005e584ac8-0xffff88005e5850cb=0x6b
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064]
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] FIX kmalloc-2048: Marking all 
objects used
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] 
=============================================================================
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] BUG kmalloc-2048 (Not 
tainted): Poison overwritten
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] 
-----------------------------------------------------------------------------
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064]
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] INFO: 
0xffff88005a37a160-0xffff88005a37a763. First byte 0xe6 instead of 0x6b
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] INFO: Allocated in 
__netdev_alloc_skb+0x1f/0x3c age=418 cpu=0 pid=5224
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] INFO: Freed in 
skb_release_data+0xa5/0xaa age=225 cpu=0 pid=5224
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] INFO: Slab 0xffffea00013bc240 
objects=15 used=14 fp=0xffff88005a37a120 flags=0x200000000040c1
Jul 29 23:15:43 Insp6400 kernel: [ 2254.454064] INFO: Object 
0xffff88005a37a120 @offset=8480 fp=0x          (null)


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


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.