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

Bug#278042: installation report (Fujitsu Lifebook P2040)



Package: installation-reports

Debian-installer-version: http://cdimage.debian.org/pub/cdimage-testing/daily/i386/current/sarge-i386-netinst.iso (Oct-22-2004)
uname -a: Linux arianus 2.4.27-1-386 #1 Fri Sep 3 06:24:46 UTC 2004 i686 GNU/Linux
Date: Oct 23rd, 2004
Method: Netinst ISO, packages installed from debian mirror over NAT'ed 802.11b.

Machine: Fujitsu Lifebook P2040
Processor: Transmeta(tm) Crusoe(tm) Processor TM5800
Memory: 240MB
Root Device: TOSHIBA MK2018GAP, ATA DISK drive
Root Size/partition table:

Disk /dev/hda: 20.0 GB, 20003880960 bytes
16 heads, 63 sectors/track, 38760 cylinders
Units = cylinders of 1008 * 512 = 516096 bytes

   Device Boot      Start         End      Blocks   Id  System
/dev/hda1   *           1       37328    18813280+  83  Linux
/dev/hda2           37329       38760      721728    f  W95 Ext'd (LBA)
/dev/hda5           37329       38760      721696+  82  Linux swap

Output of lspci and lspci -n:

0000:00:00.0 Host bridge: Transmeta Corporation LongRun Northbridge (rev 01)
0000:00:00.1 RAM memory: Transmeta Corporation SDRAM controller
0000:00:00.2 RAM memory: Transmeta Corporation BIOS scratchpad
0000:00:02.0 USB Controller: ALi Corporation USB 1.1 Controller (rev 03)
0000:00:04.0 Multimedia audio controller: ALi Corporation M5451 PCI AC-Link Controller Audio Device (rev 01)
0000:00:06.0 Bridge: ALi Corporation M7101 Power Management Controller [PMU]
0000:00:07.0 ISA bridge: ALi Corporation M1533 PCI to ISA Bridge [Aladdin IV]
0000:00:0c.0 CardBus bridge: Texas Instruments PCI1410 PC card Cardbus Controller (rev 01)
0000:00:0f.0 IDE interface: ALi Corporation M5229 IDE (rev c3)
0000:00:12.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10)
0000:00:13.0 FireWire (IEEE 1394): Texas Instruments TSB43AB21 IEEE-1394a-2000 Controller (PHY/Link)
0000:00:14.0 VGA compatible controller: ATI Technologies Inc Rage Mobility P/M (rev 64)

0000:00:00.0 0600: 1279:0395 (rev 01)
0000:00:00.1 0500: 1279:0396
0000:00:00.2 0500: 1279:0397
0000:00:02.0 0c03: 10b9:5237 (rev 03)
0000:00:04.0 0401: 10b9:5451 (rev 01)
0000:00:06.0 0680: 10b9:7101
0000:00:07.0 0601: 10b9:1533
0000:00:0c.0 0607: 104c:ac50 (rev 01)
0000:00:0f.0 0101: 10b9:5229 (rev c3)
0000:00:12.0 0200: 10ec:8139 (rev 10)
0000:00:13.0 0c00: 104c:8026
0000:00:14.0 0300: 1002:4c52 (rev 64)

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Initial boot worked:    [O]
Configure network HW:   [O]
Config network:         [E]
Detect CD:              [O]
Load installer modules: [O]
Detect hard drives:     [O]
Partition hard drives:  [O]
Create file systems:    [O]
Mount partitions:       [O]
Install base system:    [O]
Install boot loader:    [O]
Reboot:                 [O]

Comments/Problems:

(Issue #1) Wireless config wizard doesn't allow for ad-hoc networks.

I had to drop to shell and "iwconfig eth1 mode ad-hoc" to enable 
support for my wireless network. I think the best solution would 
be to put the card into monitor mode and list all networks 
(both ad-hoc and managed) and allow the user to select one. 
Failing that (for cloaked ESSIDs), the user should be able to 
select additional parameters in "expert" mode.

(Wish #2) Single-partition installs should be streamlined.

I had to click through the defaults on several pages of the 
installer. I think the simple installation method should be
streamlined to be "one-click" or at least as few as possible.

(Wish #3) Installer should remember the paritioning settings.

Installer should have remembered my "use the entire drive" 
choice from the partitioner and not bothered me with 
"we might have failed to detect other OSs on the system". 
Occurs during: Install the GRUB boot loader on a hard disk.

(Wish #4) GMT should be selected automatically if single-OS

If Debian is the only OS on the system (based on a single-OS)
choice in partitioner, it should just select "system time is
in GMT" without prompting.

If no other OSs are detected, it should default to that.

If other OSs are detected, it should default to local time.

(Wish #5) APT spew during package installation is ugly.

The console spew from APT that we show during package 
installation looks somewhat alarming. It ought to be hidden
behind a nice ncurses frontend that shows installation progress.
Something like the following would be nice:

   Package              Version         Size
libglib2.0-0         2.4.7-1        [1502kB]
[another package
[another package

[divider]

[Current package title]
[progress bar]
[package description (good for learning about what is on the system)]
[etc]

(Issue #6) Place to choose a proxy during installation?

Maybe I missed it. Where do I configure my http proxy for APT?
I have a caching proxy to reduce load on the mirrors when I
install onto more than one machine. Obviously, I could configure
this manually in /etc/apt/apt.config.

(Issue #7) X Configuration 

This is really package config for X -- but is also an essential
part of installation for a desktop.

* Could autodetect video card (XFree86 -configure), but does not.
* Could autodetect "is this an LCD", but does not.
* Shouldn't ask to autodetect mouse (just try it).
* Should autodetect resolution (esp. with LCD devices)
* Resolution of my LCD is not autodetected (1280x768)

Manual fix:

* ln -s /dev/psaux /dev/mouse
* Run XFree86 -configure
* Remove all display depths except 24
* Add DefaultDepth 24
* Clobber old config with new config

(Issue #8) Services accepting connections from network by default

Desktop machines should not listen on the network, by default.
This exposes surface area to attack, in the event of a security 
problem in one of the services. If the services were bound to 
127.0.0.1, they would not be vulnerable, even with unpatched 
security holes.

arianus:/home/lydickaw# netstat -atunp
Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name
tcp        0      0 0.0.0.0:37              0.0.0.0:*               LISTEN     837/inetd
tcp        0      0 0.0.0.0:681             0.0.0.0:*               LISTEN     9827/rpc.statd
tcp        0      0 0.0.0.0:9               0.0.0.0:*               LISTEN     837/inetd
tcp        0      0 0.0.0.0:13              0.0.0.0:*               LISTEN     837/inetd
tcp        0      0 0.0.0.0:111             0.0.0.0:*               LISTEN     9794/portmap
tcp        0      0 0.0.0.0:113             0.0.0.0:*               LISTEN     837/inetd
tcp        0      0 0.0.0.0:22              0.0.0.0:*               LISTEN     9954/sshd
udp        0      0 0.0.0.0:9               0.0.0.0:*                          837/inetd
udp        0      0 0.0.0.0:675             0.0.0.0:*                          9827/rpc.statd
udp        0      0 0.0.0.0:678             0.0.0.0:*                          9827/rpc.statd
udp        0      0 0.0.0.0:111             0.0.0.0:*                          9794/portmap

Particularly:

* The RPC services have a long history of security problems.
* They don't need to run unless NFS is enabled (and it isn't by default)
* portmap is only used for famd and the NFS-supporting services, and 
should also be bound only to 127.0.0.1
* inetd doesn't provide any useful services and should be disabled
* SSHD does not need to be on by default for a workstation.




Reply to: