Bug 600607

Summary: kdump via bond device doesn't work for non-basic config.
Product: Red Hat Enterprise Linux 6 Reporter: Qian Cai <caiqian>
Component: kexec-toolsAssignee: Cong Wang <amwang>
Status: CLOSED CURRENTRELEASE QA Contact: Han Pingtian <phan>
Severity: high Docs Contact:
Priority: medium    
Version: 6.0CC: antillon.maurizio, caiqian, enakai0, mgahagan, nhorman, phan, rkhan, syeghiay
Target Milestone: betaKeywords: Regression
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: kexec-tools-2_0_0-93_el6 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: 459879 Environment:
Last Closed: 2010-11-11 14:45:40 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---
Bug Depends On:    
Bug Blocks: 600568    
Attachments:
Description Flags
proposed patch
none
updated version
none
delta patch
none
a correct patch none

Comment 4 Han Pingtian 2010-06-24 03:48:04 UTC
None-basic bonding config still doesn't work well with -84.el6. Kdump drop into shell with this error messages:
...
Creating block device sda
 sda: sda1 sda2
Creating block device sr0
Making device-mapper control node
mdadm: No arrays found in config file or automatically
Scanning logical volumes
  Reading all physical volumes.  This may take a while...
  Found volume group "vg_ibmx365503" using metadata type lvm2
Activating logical volumes
  3 logical volume(s) in volume group "vg_ibmx365503" now active
mapping eth0 to eth4		<---------------?
ifup: duplicate interface "eth4"
ifconfig: bad address 'eth4'
eth4
eth4 failed to come up
Dropping to shell. exit to reboot
/ # 
...

/etc/sysconfig/network-scripts/ifcfg-eth0:
EVICE=eth0
BOOTPROTO=dhcp
HWADDR=00:14:5E:5B:8D:43
ONBOOT=yes

/etc/sysconfig/network-scripts/ifcfg-bond1:
DEVICE=bond1
ONBOOT=yes
BONDING_OPTS="mode=1 miimon=100"

/etc/sysconfig/network-scripts/ifcfg-eth2:
DEVICE=eth2
ONBOOT=yes
SLAVE=yes
MASTER=bond1

/etc/sysconfig/network-scripts/ifcfg-eth3:
DEVICE=eth3
ONBOOT=yes
SLAVE=yes
MASTER=bond1

/etc/modprobe.conf:
alias eth0 bnx2
alias eth1 bnx2
alias eth2 e1000e
alias eth3 e1000e
alias eth4 e1000e
alias eth5 e1000e
alias eth6 ixgbe
alias bond1 bonding

Comment 5 Han Pingtian 2010-06-24 07:59:27 UTC
Made a mistake. Added two 'net nest.test.redhat.com:/mnt/qa' in /etc/kdump.conf. I am sorry.

Comment 6 Han Pingtian 2010-06-24 08:05:13 UTC
The verification of this bug is also blocked by 605011.

Comment 7 Cong Wang 2010-06-24 08:16:41 UTC
Created attachment 426482 [details]
proposed patch

Although this is a mistake in kdump.conf, but I think this patch is still needed.

Comment 9 Cong Wang 2010-06-28 10:24:31 UTC
Created attachment 427375 [details]
updated version

Comment 10 Cong Wang 2010-06-29 06:54:04 UTC
Neil,

With this patch applied, if there are two 'net' dump targets in kdump.conf, the last one will overwrite the previous one, without this patch, we will get the above errors in the second kernel. I know, we only support one dump target. :)

What do you think about this change?

Comment 11 Neil Horman 2010-06-29 13:18:36 UTC
yeah, I'm ok with it.  multiple dump targets aren't supported, but behavior isn't defined in terms of what happens if you do that, so I'm ok with the overwrite aspect.  Thanks!

Comment 13 Han Pingtian 2010-06-30 05:28:14 UTC
(In reply to comment #9)
> Created an attachment (id=427375) [details]
> updated version    

Looks like this patch still doesn't work:

...  
Reading all physical volumes.  This may take a while...
  Found volume group "vg_ibmx365503" using metadata type lvm2
Activating logical volumes
  3 logical volume(s) in volume group "vg_ibmx365503" now active
mapping bond1 to bond0
mapping eth0 to eth0
mapping eth1 to eth1
ifup: ignoring unknown interface bond0
bond0 failed to come up
Dropping to shell. exit to reboot
/ # ifconfig
/ # 
/ # cat /etc/iface_to_activate 
bond0
/ #

Comment 14 Cong Wang 2010-06-30 06:02:51 UTC
Created attachment 427867 [details]
delta patch

Pingtian, try if this patch can fix the problem.

Comment 15 Cong Wang 2010-06-30 06:23:46 UTC
Created attachment 427875 [details]
a correct patch

Comment 17 Han Pingtian 2010-07-01 06:53:56 UTC
Reproduced with -72.el6:
... ...
mdadm: No arrays found in config file or automatically
Scanning logical volumes
  Reading all physical volumes.  This may take a while...
  Found volume group "vg_ibmx365503" using metadata type lvm2
Activating logical volumes
  3 logical volume(s) in volume group "vg_ibmx365503" now active
mapping bond1 to bond0
mapping eth0 to eth0
mapping eth1 to eth1
mapping eth2 to eth2
mapping eth3 to eth3
ADDRCONF(NETDEV_UP): bond0: link is not ready
bond0 Link Up.  Waiting 60 Seconds
Continuing
searching for slaves
msh: /etc/ifcfg-eth4: No such file or directory
msh: /etc/ifcfg-eth5: No such file or directory
msh: /etc/ifcfg-eth6: No such file or directory
msh: /etc/ifcfg-lo: No such file or directory
msh: /etc/ifcfg-sit0: No such file or directory
udhcpc (v1.15.1) started
Sending discover...
Sending discover...
Sending discover...
No lease, failing
bond0 failed to come up
Dropping to shell. exit to reboot
/ # 

Verified with -95.el6:
... ...
mdadm: No arrays found in config file or automatically
Scanning logical volumes
  Reading all physical volumes.  This may take a while...
  Found volume group "vg_ibmx365503" using metadata type lvm2
Activating logical volumes
  3 logical volume(s) in volume group "vg_ibmx365503" now active
mapping bond1 to bond0
mapping eth0 to eth0
mapping eth1 to eth1
ADDRCONF(NETDEV_UP): bond0: link is not ready
bond0 Link Up.  Waiting 60 Seconds
Continuing
searching for slaves
enslaving eth0 tbonding: bond0: Adding slave eth0.
o bond0
bnx2: eth0: using MSI
bonding: bond0: Warning: failed to get speed and duplex from eth0, assumed to be 100Mb/sec and Full.
bonding: bond0: enslaving eth0 as an active interface with an up link.
ADDRCONF(NETDEV_CHANGE): bond0: link becomes ready
enslaving eth1 to bond0
bonding: bond0: Adding slave eth1.
bnx2: eth1: using MSI
bonding: bond0: Warning: failed to get speed and duplex from eth1, assumed to be 100Mb/sec and Full.
bonding: bond0: enslaving eth1 as an active interface with an up link.
udhcpc (v1.15.1) started
Sending discover...
Sending discover...
bnx2: eth0 NIC Copper Link is Up, 1000 Mbps full duplex
Sending discover...
Sending select for 10.16.68.96...
Lease of 10.16.68.96 obtained, lease time 86400
deleting routers

Comment 18 releng-rhel@redhat.com 2010-11-11 14:45:40 UTC
Red Hat Enterprise Linux 6.0 is now available and should resolve
the problem described in this bug report. This report is therefore being closed
with a resolution of CURRENTRELEASE. You may reopen this bug report if the
solution does not work for you.