Bug 150374

Summary: netdump chokes on eth1 device in /etc/sysconfig/netdump
Product: [Fedora] Fedora Reporter: p thompson <pt>
Component: netdumpAssignee: Jeff Moyer <jmoyer>
Status: CLOSED NEXTRELEASE QA Contact:
Severity: high Docs Contact:
Priority: medium    
Version: 3   
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2006-04-18 13:18:52 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description p thompson 2005-03-05 04:13:30 UTC
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.5)
Gecko/20041107 Firefox/1.0

Description of problem:
Ethernet configuration on host, two eepro100 NIC's:

02:0d.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100]
(rev 05)
        Subsystem: Intel Corp. EtherExpress PRO/100+ Management Adapter

02:0d.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100]
(rev 05)
        Subsystem: Intel Corp. EtherExpress PRO/100+ Management Adapter


{public internet}
eth0      Link encap:Ethernet  HWaddr 00:90:27:1C:3A:81
          inet addr:24.106.25.221  Bcast:255.255.255.255  
Mask:255.255.255.0
{private network}
eth1      Link encap:Ethernet  HWaddr 00:90:27:35:D2:B2
          inet addr:192.168.0.1  Bcast:192.168.0.255  Mask:255.255.255.0

/etc/sysconfig/netdump with no DEV= automatically assumes eth0,
but can't communicate with the netdump-server which is on eth1.

see syslog below
<snip>
 LOCALPORT=6666
# DEV=eth1
 NETDUMPADDR=192.168.0.5
 NETDUMPPORT=6666
 NETDUMPMACADDR=00:D0:B7:C0:E9:1E
 IDLETIMEOUT=
 NETDUMPKEYEXCHANGE=none

syslog on 192.168.0.1
Mar  4 21:59:09 monotheletisia kernel: netdump: local port 6666
Mar  4 21:59:09 monotheletisia kernel: netdump: local IP 24.106.25.221
Mar  4 21:59:09 monotheletisia kernel: netdump: interface eth0
Mar  4 21:59:09 monotheletisia kernel: netdump: remote port 6666
Mar  4 21:59:09 monotheletisia kernel: netdump: remote IP 192.168.0.5
Mar  4 21:59:09 monotheletisia kernel: netdump: remote ethernet
address 00:d0:b7
:c0:e9:1e
Mar  4 21:59:09 monotheletisia kernel: netdump: network crash dump enabled
Mar  4 21:59:09 monotheletisia netdump: initializing netdump succeeded
Mar  4 21:59:09 monotheletisia kernel: netconsole: local port 6665
Mar  4 21:59:09 monotheletisia kernel: netconsole: local IP 24.106.25.221
Mar  4 21:59:09 monotheletisia kernel: netconsole: interface eth0
Mar  4 21:59:09 monotheletisia kernel: netconsole: remote port 514
Mar  4 21:59:09 monotheletisia kernel: netconsole: remote IP 192.168.0.5
Mar  4 21:59:09 monotheletisia kernel: netconsole: remote ethernet
address 00:d0
:b7:c0:e9:1e
Mar  4 21:59:09 monotheletisia kernel: netlog: local port 6664
Mar  4 21:59:09 monotheletisia kernel: netlog: local IP 24.106.25.221
Mar  4 21:59:09 monotheletisia kernel: netlog: interface eth0
Mar  4 21:59:09 monotheletisia netdump: initializing netconsole succeeded

When the DEV=eth1 parameter is put back in, netdump won't start
/etc/sysconfig/netdump
{snip}
 DEV=eth1
[root@monotheletisia scripts]# service netdump restart
netdump: eth1 must be an ethernet interface
netdump eth1 configuration                                 [FAILED]



Version-Release number of selected component (if applicable):


How reproducible:
Always

Steps to Reproduce:
1.DEV=eth1 in /etc/sysconfig/netdump
2.  service netdump restart
3.
    

Actual Results:  netdump: eth1 must be an ethernet interface
netdump eth1 configuration                                 [FAILED]



Expected Results:  Dumps pump out eth1

Additional info:

Comment 1 p thompson 2005-03-05 04:14:32 UTC
Linux monotheletisia 2.6.10-1.766_FC3 #1 Wed Feb 9 23:06:42 EST 2005
i686 i686 i386 GNU/Linux
netdump-server-0.7.0-1
kernel-utils-2.4-13.1.49_FC3
kernel-doc-2.6.10-1.766_FC3
netdump-0.7.0-1
kernel-2.6.10-1.766_FC3


Comment 2 Jeff Moyer 2005-03-08 13:50:55 UTC
Yup, I was able to reproduce this in-house.  I'll take a look.

Comment 3 Jeff Moyer 2005-08-01 19:46:35 UTC
Please try the rawhide version of netdump.  It has some changes to the init
script which should address this issue.  You can find a copy of the package here:
  http://people.redhat.com/jmoyer/bz150374/

Let me know how the testing goes.

Thanks.

Comment 4 p thompson 2005-08-01 21:53:59 UTC
service netdump restart
initializing netdump FATAL: Module netdump not found.
                                                           [FAILED]
initializing netconsole                                    [  OK  ]
[root@monotheletisia thompson]# dmesg -c
SMB connection re-established (-5)
netconsole: Unknown parameter `netlog'
netconsole: local port 6665
netconsole: local IP 192.168.0.1
netconsole: interface eth1
netconsole: remote port 514
netconsole: remote IP 192.168.0.5
netconsole: remote ethernet address 00:d0:b7:c0:e9:1e
netconsole: network logging started

kernel-devel-2.6.12-1.1372_FC3
netdump-0.7.10-4

The most promising results so far.  I have not had the problems lately with the
kernel which caused all my trouble that started me down the netdump path but I
will update when appropriate.

Comment 5 Jeff Moyer 2005-08-02 12:01:50 UTC
Hmm, it seems that the netdump module isn't being built in fc3?  I thought that
was resolved.  I'll have a look at that issue.

Thanks for the quick test turn-around!

Comment 6 p thompson 2005-09-20 19:13:18 UTC
Things look promising, is there an easy way to test things without an actual crash?


Comment 7 Jeff Moyer 2005-09-20 19:20:58 UTC
Sure, you could test using netconsole (aka netlog).  there should be a log file
created on the server when the netconsole module is inserted.  Or, you could
trigger one using sysrq-h.


Comment 8 Jeff Moyer 2006-04-17 19:04:53 UTC
It's unclear to me what the state of this bug is.  Is the issue addressed for you?

Comment 9 p thompson 2006-04-18 03:50:12 UTC
I believe this is corrected

Comment 10 Jeff Moyer 2006-04-18 13:18:52 UTC
OK.  I'm guessing this was fixed in fc4, not fc3.  I'm going to resolve it as
NEXTRELEASE.