Subject: Re: [Fwd: NSLU2 builds and boots, but seems to disappear when running
To: Donald T Hayford <don@donhayford.com>
From: Donald T Hayford <don@donhayford.com>
List: port-arm
Date: 01/20/2008 08:22:13
This is a multi-part message in MIME format.
--------------090405030201060504090701
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Problems described below are definitely NFS problems. I checked
/var/log/messages on the TFTP/NFS server (Fedora 7) and saw the following:
Jan 20 09:53:03 mythtv xinetd[2207]: START: tftp pid=26560 from=192.168.0.1
Jan 20 09:53:38 mythtv dhcpd: DHCPDISCOVER from 00:18:39:a2:26:7c via eth0
Jan 20 09:53:38 mythtv dhcpd: DHCPOFFER on 192.168.1.240 to
00:18:39:a2:26:7c via eth0
Jan 20 09:53:38 mythtv dhcpd: DHCPREQUEST for 192.168.1.240
(192.168.1.102) from 00:18:39:a2:26:7c via eth0
Jan 20 09:53:38 mythtv dhcpd: DHCPACK on 192.168.1.240 to
00:18:39:a2:26:7c via eth0
Jan 20 09:53:41 mythtv mountd[2265]: authenticated mount request from
slug:1023 for /video/client/root (/video)
Jan 20 09:53:43 mythtv setroubleshoot: SELinux is preventing
/usr/sbin/rpc.mountd (nfsd_t) "ioctl" to /dev/mapper/control
(lvm_control_t). For complete SELinux messages. run sealert -l
3f5ecf3b-4146-484b-92e3-c9819a90ceea
Jan 20 09:53:44 mythtv setroubleshoot: SELinux is preventing the
nfs daemon from allow remote clients to read local files. For
complete SELinux messages. run sealert -l
054092b2-6c43-489e-9fe5-81b7507ba7f7
Jan 20 09:53:44 mythtv setroubleshoot: SELinux is preventing
/usr/sbin/rpc.mountd (nfsd_t) "ioctl" to /dev/sda2
(fixed_disk_device_t). For complete SELinux messages. run sealert
-l 996e2623-ced8-4008-8fd8-7998ffd8bb2b
Jan 20 10:08:03 mythtv xinetd[2207]: EXIT: tftp status=0 pid=26560
duration=900(sec)
Jan 20 10:23:57 mythtv setroubleshoot: SELinux is preventing
/usr/sbin/rpc.mountd (nfsd_t) "ioctl" to /dev/sda2
(fixed_disk_device_t). For complete SELinux messages. run sealert
-l 996e2623-ced8-4008-8fd8-7998ffd8bb2b
So, even though SELinux on that machine is set to "permissive", it is
still preventing the slug from using the NFS files as root (maybe
because it doesn't have the right password? I can connect to them from
another machine without a problem.)
At any rate, I'm going to try recompiling the kernel with ext2 support
and see if I can create a root USB drive from a linux machine (I don't
have a machine running NetBSD). Trying to figure out SELinux seems like
a no-win situation, though I suppose I could always disable it.
If somebody has a better idea, I'd appreciate hearing about it.
Thanks, Don
Donald T Hayford wrote:
> Excuse me for probably sending the same message twice, but I was
> having network problems last night (not related to NSLU2) and never
> got a confirmation that the message below ever got to port-arm.
>
> -------- Original Message --------
> Subject: NSLU2 builds and boots, but seems to disappear when running
> /sbin/init
> Date: Sat, 19 Jan 2008 18:26:19 -0500
> From: Donald T Hayford <don@donhayford.com>
> To: port-arm@netbsd.org
>
>
>
> I am able to build the NSLU2 arm port, and constructed a root file
> system using NFS. The first time I booted the system, I got a warning
> that /dev/console was missing, and then the NSLU2 stopped with the
> following message:
> /
> [...snip...]
> root device: npe0
> dump device:
> file system (default generic): nfs
> root on npe0
> nfs_boot: trying DHCP/BOOTP
> nfs_boot: DHCP next-server: 192.168.1.102
> nfs_boot: my_addr=192.168.1.240
> nfs_boot: my_mask=255.255.255.0
> nfs_boot: gateway=192.168.1.1
> root on 192.168.1.102:/video/client/root
> root time: 0x4792690f
> xrtc0: xrtc_clock_read: failed to read rtc at 0x0
> xrtc0: xrtc_clock_read: failed to read rtc at 0x0
> WARNING: preposterous TOD clock time
> WARNING: using filesystem time
> WARNING: CHECK AND RESET THE DATE!
> warning: no /dev/console
> init path (default /sbin/init):
> init: copying out path `/sbin/init' 11
> [a few seconds go by]
> panic: init died (signal 0, exit 12)
> Stopped in pid 1.1 (init) at netbsd:cpu_Debugger+0x4:
> bx r14
> db>
> [end of text - NSLU2 is running the debugger]
> /
> Using the nbmknod command, I created dev/console using "sudo nbmknod
> root/dev/console c 0 0" and rebooted the system. The NSLU2 appears to
> boot ok (though with the same message about failing to read the rtc),
> and continues through the "init path" message. The system no longer
> panics and reboots, but neither does it allow me to log in. If I ping
> the NSLU2 address, it responds, but I can neither ssh or telnet into
> the NSLU2. If I press the power button on the front, the system
> powers down with the following message stream:
>
> /[...snip...]
> init path (default /sbin/init): /sbin/init
> init: copying out path `/sbin/init' 11
>
> [NSLU2 appears to hang at this point, but responds to a ping]
> [when the power button is pressed...]
>
> slugbutt0: power button pressed, shutting down!
> boot: howto=00000808 curproc=0xc048e6e4
> syncing disks... done
> unmounting file systems...
> unmounting / (192.168.1.102:/video/client/root)... done
> powering down...
>
> [NSLU2 powers off]./
>
> I've checked /var/log for anything that the NSLU2 might have written
> there that would indicate the problem, but don't see anything. I've
> used FreeBSD before, but I'm new to NetBSD, so I'm not sure where to
> go next. I'm thinking that most likely I have a problem with how I
> set up the NFS root, and I'm going to try to set up a usb drive to use
> as root. Definitely sailing some uncharted waters here (for me, anyway).
>
> As always, any help is greatly appreciated.
>
> Thanks, Don
>
--------------090405030201060504090701
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Problems described below are definitely NFS problems. I checked
/var/log/messages on the TFTP/NFS server (Fedora 7) and saw the
following:<br>
<br>
Jan 20 09:53:03 mythtv xinetd[2207]: START: tftp pid=26560
from=192.168.0.1<br>
Jan 20 09:53:38 mythtv dhcpd: DHCPDISCOVER from 00:18:39:a2:26:7c via
eth0<br>
Jan 20 09:53:38 mythtv dhcpd: DHCPOFFER on 192.168.1.240 to
00:18:39:a2:26:7c via eth0<br>
Jan 20 09:53:38 mythtv dhcpd: DHCPREQUEST for 192.168.1.240
(192.168.1.102) from 00:18:39:a2:26:7c via eth0<br>
Jan 20 09:53:38 mythtv dhcpd: DHCPACK on 192.168.1.240 to
00:18:39:a2:26:7c via eth0<br>
Jan 20 09:53:41 mythtv mountd[2265]: authenticated mount request from
slug:1023 for /video/client/root (/video)<br>
Jan 20 09:53:43 mythtv setroubleshoot: SELinux is preventing
/usr/sbin/rpc.mountd (nfsd_t) "ioctl" to /dev/mapper/control
(lvm_control_t). For complete SELinux messages. run sealert -l
3f5ecf3b-4146-484b-92e3-c9819a90ceea<br>
Jan 20 09:53:44 mythtv setroubleshoot: SELinux is preventing the
nfs daemon from allow remote clients to read local files. For
complete SELinux messages. run sealert -l
054092b2-6c43-489e-9fe5-81b7507ba7f7<br>
Jan 20 09:53:44 mythtv setroubleshoot: SELinux is preventing
/usr/sbin/rpc.mountd (nfsd_t) "ioctl" to /dev/sda2
(fixed_disk_device_t). For complete SELinux messages. run sealert
-l 996e2623-ced8-4008-8fd8-7998ffd8bb2b<br>
Jan 20 10:08:03 mythtv xinetd[2207]: EXIT: tftp status=0 pid=26560
duration=900(sec)<br>
Jan 20 10:23:57 mythtv setroubleshoot: SELinux is preventing
/usr/sbin/rpc.mountd (nfsd_t) "ioctl" to /dev/sda2
(fixed_disk_device_t). For complete SELinux messages. run sealert
-l 996e2623-ced8-4008-8fd8-7998ffd8bb2b<br>
<br>
So, even though SELinux on that machine is set to "permissive", it is
still preventing the slug from using the NFS files as root (maybe
because it doesn't have the right password? I can connect to them from
another machine without a problem.)<br>
<br>
At any rate, I'm going to try recompiling the kernel with ext2 support
and see if I can create a root USB drive from a linux machine (I don't
have a machine running NetBSD). Trying to figure out SELinux seems
like a no-win situation, though I suppose I could always disable it.<br>
<br>
If somebody has a better idea, I'd appreciate hearing about it.<br>
<br>
Thanks, Don<br>
<br>
Donald T Hayford wrote:
<blockquote cite="mid479304D8.8060909@donhayford.com" type="cite">
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
Excuse me for probably sending the same message twice, but I was having
network problems last night (not related to NSLU2) and never got a
confirmation that the message below ever got to port-arm.<br>
<br>
-------- Original Message --------
<table class="moz-email-headers-table" border="0" cellpadding="0"
cellspacing="0">
<tbody>
<tr>
<th align="right" nowrap="nowrap" valign="baseline">Subject: </th>
<td>NSLU2 builds and boots, but seems to disappear when running
/sbin/init</td>
</tr>
<tr>
<th align="right" nowrap="nowrap" valign="baseline">Date: </th>
<td>Sat, 19 Jan 2008 18:26:19 -0500</td>
</tr>
<tr>
<th align="right" nowrap="nowrap" valign="baseline">From: </th>
<td>Donald T Hayford <a class="moz-txt-link-rfc2396E"
href="mailto:don@donhayford.com"><don@donhayford.com></a></td>
</tr>
<tr>
<th align="right" nowrap="nowrap" valign="baseline">To: </th>
<td><a class="moz-txt-link-abbreviated"
href="mailto:port-arm@netbsd.org">port-arm@netbsd.org</a></td>
</tr>
</tbody>
</table>
<br>
<br>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
I am able to build the NSLU2 arm port, and constructed a root file
system using NFS. The first time I booted the system, I got a warning
that /dev/console was missing, and then the NSLU2 stopped with the
following message:<br>
<i><br>
[...snip...]<br>
root device: npe0<br>
dump device:<br>
file system (default generic): nfs<br>
root on npe0<br>
nfs_boot: trying DHCP/BOOTP<br>
nfs_boot: DHCP next-server: 192.168.1.102<br>
nfs_boot: my_addr=192.168.1.240<br>
nfs_boot: my_mask=255.255.255.0<br>
nfs_boot: gateway=192.168.1.1<br>
root on 192.168.1.102:/video/client/root<br>
root time: 0x4792690f<br>
xrtc0: xrtc_clock_read: failed to read rtc at 0x0<br>
xrtc0: xrtc_clock_read: failed to read rtc at 0x0<br>
WARNING: preposterous TOD clock time<br>
WARNING: using filesystem time<br>
WARNING: CHECK AND RESET THE DATE!<br>
warning: no /dev/console<br>
init path (default /sbin/init):<br>
init: copying out path `/sbin/init' 11<br>
[a few seconds go by]<br>
panic: init died (signal 0, exit 12)<br>
Stopped in pid 1.1 (init) at netbsd:cpu_Debugger+0x4: bx
r14<br>
db><br>
[end of text - NSLU2 is running the debugger]<br>
</i><br>
Using the nbmknod command, I created dev/console using "sudo nbmknod
root/dev/console c 0 0" and rebooted the system. The NSLU2 appears to
boot ok (though with the same message about failing to read the rtc),
and continues through the "init path" message. The system no longer
panics and reboots, but neither does it allow me to log in. If I ping
the NSLU2 address, it responds, but I can neither ssh or telnet into
the NSLU2. If I press the power button on the front, the system powers
down with the following message stream:<br>
<meta http-equiv="CONTENT-TYPE" content="text/html; ">
<title></title>
<meta name="GENERATOR" content="OpenOffice.org 2.0 (Linux)">
<meta name="AUTHOR" content="Donald Hayford">
<meta name="CREATED" content="20080111;5130400">
<meta name="CHANGEDBY" content="Donald Hayford">
<meta name="CHANGED" content="20080119;13413200">
<style type="text/css">
<!--
@page { size: 8.5in 11in; margin: 0.79in }
P { margin-bottom: 0.08in }
--> </style><br>
<i>[...snip...]<br>
init path (default /sbin/init): /sbin/init<br>
init: copying out path `/sbin/init' 11<br>
<br>
[NSLU2 appears to hang at this point, but responds to a ping]<br>
[when the power button is pressed...]<br>
<br>
slugbutt0: power button pressed, shutting down!<br>
boot: howto=00000808 curproc=0xc048e6e4<br>
syncing disks... done<br>
unmounting file systems...<br>
unmounting / (192.168.1.102:/video/client/root)... done<br>
powering down...<br>
<br>
[NSLU2 powers off].</i><br>
<br>
I've checked /var/log for anything that the NSLU2 might have written
there that would indicate the problem, but don't see anything. I've
used FreeBSD before, but I'm new to NetBSD, so I'm not sure where to go
next. I'm thinking that most likely I have a problem with how I set up
the NFS root, and I'm going to try to set up a usb drive to use as
root. Definitely sailing some uncharted waters here (for me, anyway).<br>
<br>
As always, any help is greatly appreciated.<br>
<br>
Thanks, Don<br>
<br>
</blockquote>
</body>
</html>
--------------090405030201060504090701--