On 07/05/11 19:10, AG wrote:
I'm going for the "nuclear" option and am downloading the
netinstall.iso and will install from scratch and see if I run into the
same problems again in the future. So far, the only successes have
been when I enable the dhcp option in the /etc/network/interfaces
file, but as I need to have a printer client find the server this is
no good to me. I (strongly?) suspect that the offending culprit is
Network Manager, but cannot demonstrate that, so will endeavour to
avoid or de-install that if possible.
Hi again
I'm now running squeeze from the Feb netinst i386 iso.[1] This is a
fresh install using graphical expert install. I set the parameters of
the network as follows (being a creature of habit):
No auto-configure with DHCP
IP 192.168.1.40
gateway 192.168.1.254
nameserver 87.194.255.154 #modified from previous
hostname valhalla
I haven't altered anything aside from the configuration during install
and updating sources.list to testing rather than squeeze to update
applications. I haven't done any comprehensive dist-upgrade and there
are, according to the Update Mngr some 587 packages awaiting me.
I would like to gradually ease this into a testing, so given my last
experience am wondering what the best way of doing this will be. I want
to give any network related packages a wide berth for now until I can
figure out what went wrong last time.
In pursuit of this I re-ran Camaleón's helpful heuristic for the
previous installation and noted the output. I compared the current
output with that previously posted and can only see differences in three
stanzas, listed below. I can't comment on how significant these
differences might be, so comments welcomed.
# The primary network interface (this is different due to the parameters
manually configured during install, so the difference is to be expected,
especially the "static" clause)
allow-hotplug eth0
iface eth0 inet static
address 192.168.1.40
netmask 255.255.255.0
network 192.168.1.0
broadcast 192.168.1.255
gateway 192.168.1.254
# dns-* options are implemented by the resolvconf package, if installed
dns-nameservers 87.194.255.154 87.194.255.155
dns-search org
# output from /etc/resolv.conf is:
$ cat /etc/resolv.conf
search org
nameserver 87.194.255.154
nameserver 87.194.255.155
# compared with previously (& probably a limited range because the
domain is defined as lan rather than as org):
$ cat /etc/resolv.conf
domain lan
search lan
nameserver 192.168.1.254
# And dmesg | grep -i eth is different too:
# Now:
$ dmesg | grep -i eth
[ 0.231823] ACPI Error (psparse-0537): Method parse/execution failed
[\_SB_.MEM_._CRS] (Node f6c1a0c0), AE_AML_BUFFER_LIMIT
[ 0.231859] ACPI Error (uteval-0250): Method execution failed
[\_SB_.MEM_._CRS] (Node f6c1a0c0), AE_AML_BUFFER_LIMIT
[ 4.312872] 8139cp: 10/100 PCI Ethernet driver v1.3 (Mar 22, 2004)
[ 4.746282] 8139too Fast Ethernet driver 0.9.28
[ 4.747608] eth0: RealTek RTL8139 at 0xec00, 00:90:47:05:a3:07, IRQ 18
[ 9.062298] eth0: link up, 100Mbps, full-duplex, lpa 0x41E1
[ 13.589391] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
[ 19.820518] eth0: no IPv6 routers present
# Previously (don't know if these are renewed signals and addresses or
significant):
$ dmesg | grep -i eth
[ 0.178866] ACPI Error: Method parse/execution failed
[\_SB_.MEM_._CRS] (Node f4c370c0), AE_AML_BUFFER_LIMIT
(20110112/psparse-536)
[ 0.178875] ACPI Error: Method execution failed [\_SB_.MEM_._CRS]
(Node f4c370c0), AE_AML_BUFFER_LIMIT (20110112/uteval-103)
[ 1.014972] 8139too 0000:01:0a.0: eth0: RealTek RTL8139 at 0xec00,
00:90:47:05:a3:07, IRQ 18
[ 10.733628] eth0: link up, 100Mbps, full-duplex, lpa 0x41E1
[ 20.752026] eth0: no IPv6 routers present
I don't know if any of this is of any value, although from my
perspective I am pleased to have a system that works as I want it out of
the box. I will progress through the upgrades slowly, and have
uninstalled Network Manager in case that was the culprit, and in any
event my needs are not that complex that require something like that.
So, in terms of upgrading, would doing an aptitude safe-upgrade be the
wisest approach to updating squeeze to testing/ wheezy?
Thanks
AG
[1] I downloaded and installed using the latest testing/ wheezy netinst
CD iso and when I went to logon after booting up, the mouse was frozen
as was the keyboard (and for several reattempts to logon). Also, during
the install process, it hadn't required me to input anything for the
network section of the installation. Perhaps it is just me, but I am
wondering if that was part of the problem as I had upgraded from squeeze
to testing in the last few days and if that was the straight
installation experience, then perhaps the upgrade was broken as well?
This is, of course, pure speculation on my part, but explains whey I
went back to squeeze as a known release candidate.
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4dc64b11.5060...@gmail.com
