topic title: antiX-14R-a4-a5 patch
Posts: 2,238
dolphin_oracle
Joined: 16 Dec 2007
#76
Note
In all cases of failure ceni can be used to manually set up networking.
you should be able to set up the networking in wicd as well. you may need to change the interface names in preferences. in samk's report above, eth0 is already set as default wired device in wicd, but his wireless (looks like eth1 in the reports) would have to be changed in the wicd preferences because wicd is set to wlan0 by default.

this is also an issue on live w/ persistence media, since every system you boot from will have new interface names different from the systems you've booted on before. (wlan0, wlan1, wlan2, so on).
Posts: 1,308
BitJam
Joined: 31 Aug 2009
#77
dolphin_oracle wrote:this is also an issue on live w/ persistence media, since every system you boot from will have new interface names different from the systems you've booted on before. (wlan0, wlan1, wlan2, so on).
I thought we fixed this. If we didn't fix it then it needs to be fixed. Do you remember which file this information is stored in?
Posts: 2,238
dolphin_oracle
Joined: 16 Dec 2007
#78
the information is stored in /etc/udev/rules.d/70-persistent-net.rules. The interface names are stored here by mac address. the information is generated by /lib/udev/rules.d/75-persistent-net-generator.rules.
Posts: 1,308
BitJam
Joined: 31 Aug 2009
#79
In antiX-14-a5 there is the following entry in /usr/local/lib/antiX/antiX-excludes.sh:

Code: Select all

    etc/udev/rules.d/*-persistent-net.rules
I think this prevents the problem you mentioned. PLMK if it does not.
Posts: 2,238
dolphin_oracle
Joined: 16 Dec 2007
#80
I don't think so. Here is my current /etc/udev/rules.d/70-persistent-net.rules

Code: Select all

# This file was automatically generated by the /lib/udev/write_net_rules
# program, run by the persistent-net-generator.rules rules file.
#
# You can modify it, as long as you keep each rule on a single
# line, and change only the value of the NAME= key.

# PCI device 0x10ec:0x8168 (r8169)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="64:31:50:6d:81:b4", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"

# PCI device 0x8086:0x0089 (iwlwifi)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:23:15:7b:a8:14", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="wlan0"

# PCI device 0x168c:0x001c (ath5k)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:22:43:56:88:b5", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="wlan1"

# PCI device 0x1969:0x1026 (ATL1E)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:23:54:81:78:aa", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth1"

# PCI device 0x8086:0x10de (e1000e)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="b8:ac:6f:4c:a5:66", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth2"

# PCI device 0x14e4:0x4328 (b43)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="5c:ac:4c:87:a2:7f", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="wlan2"
My question is, when does that excludes list take effect? because 70-pesistent-net.rules doesn't exist until 75-persistent-net-generator.rules makes it. Maybe 75-persistent-net-generator.rules is the rule that needs excluded.

obligatory inxi -F for this live system

Code: Select all

demo@antix1:~
$ inxi -F
System:    Host: antix1 Kernel: 3.18.1-antix.1-486-smp i686 (32 bit)
           Desktop: IceWM 1.3.8
           Distro: antiX-14-a5-RV_386-full Killah_P 14 January 2015
Machine:   System: Dell product: OptiPlex 780
           Mobo: Dell model: 0C27VV v: A01 Bios: Dell v: A04 date: 04/30/2010
CPU:       Dual core Intel Core2 Duo E7500 (-MCP-) cache: 3072 KB 
           clock speeds: max: 2933 MHz 1: 2400 MHz 2: 2400 MHz
Graphics:  Card: Intel 4 Series Integrated Graphics Controller
           Display Server: X.Org 1.16.2.901 drivers: intel (unloaded: fbdev,vesa)
           Resolution: 1280x1024@60.02hz
           GLX Renderer: Mesa DRI Intel Q45/Q43 x86/MMX/SSE2
           GLX Version: 2.1 Mesa 10.3.2
Audio:     Card Intel 82801JD/DO (ICH10 Family) HD Audio Controller
           driver: snd_hda_intel
           Sound: ALSA v: k3.18.1-antix.1-486-smp
Network:   Card-1: Intel 82567LM-3 Gigabit Network Connection driver: e1000e
           IF: eth2 state: up speed: 1000 Mbps duplex: full
           mac: b8:ac:6f:4c:a5:66
           Card-2: Broadcom BCM4321 802.11a/b/g/n driver: b43-pci-bridge
           IF: wlan2 state: down mac: 5c:ac:4c:87:a2:7f
Drives:    HDD Total Size: 266.0GB (0.5% used)
           ID-1: /dev/sda model: WDC_WD2500AAJS size: 250.0GB
           ID-2: USB /dev/sdb model: Cruzer_Glide size: 16.0GB
Sensors:   System Temperatures: cpu: 33.0C mobo: N/A
           Fan Speeds (in rpm): cpu: N/A
Info:      Processes: 115 Uptime: 11 min Memory: 238.2/2996.3MB
           Client: Shell (bash) inxi: 2.2.16 
demo@antix1:~
$
Posts: 1,308
BitJam
Joined: 31 Aug 2009
#81
dolphin_oracle wrote:My question is, when does that excludes list take effect? because 70-pesistent-net.rules doesn't exist until 75-persistent-net-generator.rules makes it. Maybe 75-persistent-net-generator.rules is the rule that needs excluded.
I tested it here on antiX-14.-a5b (soon to be available) and it worked as intended. I mounted the rootfs file and there was no / etc/udev/rules.d/70-persistent-net.rules file in it even though I had just created that file. Maybe my memory is clouded by wishful thinking but I thought you had already tested this change and reported that it worked.

The plan was to leave the rule generator in place so networks are persistent while you are running but then we remove the 70-persistent-net.rules file during a remaster or a persist-save so there is no memory between boots. The exclude happens when you run persist-save or remaster-live. If you are using static root persistence then it won't take effect. If the problem is static root persistence then we could run code at shutdown that erases certain files.

The excludes should show up in /live/config/rsync.exclude after you run a persist-save manually.

You can also look for the file under /live/linux/ etc/udev/rules.d/. If the file is there then it made it into a remaster but even then, the deletion of the file should be remembered in the rootfs (using tricky hidden files).

You should be able to mount the rootfs file to look inside it with:

Code: Select all

sudo mount /live/boot-dev/antiX/rootfs /live/root-persist
Alternatively you can use the"Explore" option in persist-makefs.
Posts: 2,238
dolphin_oracle
Joined: 16 Dec 2007
#82
yep static root persistence. that's the culprit. when we tested it earlier I was probably using the standard persistence.
Posts 1,308
BitJam
Joined: 31 Aug 2009
#83
Thanks. Good to know. It will be easy to erase some files at shutdown if static root persistence is enabled. The bigger question is: which files to erase? I suppose that question is already answered with the exclude list for dynamic root persistence.
Posts: 604
thriftee
Joined: 27 Feb 2009
#84
Tried it on my Dell Optiplex gx270 and I get

Code: Select all

graphics initialization failed
error setting up gfxboot 

it then gives a little text menu
Welcome
antiX
blah blah
memtest86

boot:
but if i type antiX as shown it says it can't find the OS

Same problem with 13.2 or 14.5

The system comes up fine 1024x768 using driver radeon under knoppix 7.05

The original built in intel graphics don't work at all and so it has a radeon 7000 pci board added and bios set for auto instead of onboard.
Posts: 1,308
BitJam
Joined: 31 Aug 2009
#85
@thriftee, I am delighted you have stumbled across this bug. It gives us a chance to test code that is in place to deal with this situation. It should actually be working in antiX-14-a5 but the menu you are given is wrong. The actual menu is:

Code: Select all

live                antiX-14-a5-RV_386-full (14 January 2015)
cli                 Command Line Install
frugal              Frugal Install
root-persist        Root Persistence
root-persist-static Static Root Persistence
home-persist        Home Persistence
harddisk            Boot_from_Hard_Disk
memtest             Memory_test_(memtest86+)
1) Please let me know if you can boot using"live".

2) obviously, the menu needs to get fixed. The question is: what should we use for labels and short descriptions? The main menu will be changed in the next release. The current plan is for it to be:

Code: Select all

live       %FULL DISTRO NAME% (%RELEASE DATE%)
safe       Safe Video Mode
failsafe   Failsafe Boot
cli        Command Line Boot
harddisk   Boot from Hard Disk
memtest    Memory test (memtest86+)
where I've listed the shortcut and then the label from the main menu. Does this look okay? Should this be what is in the menu that gets displayed when the graphics mode fails?
Posts: 604
thriftee
Joined: 27 Feb 2009
#86
Bitjam,

Some people in this world are lucky, and THEN there's people like you that are REAL LUCKY.

1. I had a 14.4 DVD in my"furture tree ornaments" pile
2. The old thing has a DVD drive not a CD drive because that's what I replaced the floppy with 8 yrs ago
3. I typed"live" at the boot prompt and it went off booting

Ok, now to get 4 for 4, I have a black screen after seeing a login prompt go by, so I think it died trying to get X running. My guess is that if I could get X to use a Radeon driver in 1024x768 it would all work. Is there a way to tell it to try that combination?

Ok, I did a Ctrl-Alt-F1, and logged into console, and I think if I could blacklist i915 the radeon would take over the things it didn't. Does that sound right?

I got it to work temporarily by ctrl-alt-bksp from X

Code: Select all

modprobe -r i915
modprobe radeon
exit
login as demo/demo

Code: Select all

startx
So all I need now is to know how to do that from the boot command line. It seems to me I saw that somewhere...
Posts: 1,308
BitJam
Joined: 31 Aug 2009
#87
thriftee wrote:So all I need now is to know how to do that from the boot command line. It seems to me I saw that somewhere...
In the next release there will be a"blacklist=xxx" boot parameter.

Have you tried using"F5 Video Mode" -->"safe" or"F5 Video Mode" -->"vesa"? You can also try typing in"xdrvr=radeon" as a boot parameter. You may need to add"nomodeset" which is what F5 -->"safe" gives you. Use F12 to see the current boot parameters.

I suggest combining F5 -->"safe" with typing in"xdrvr=radeon".
Posts: 604
thriftee
Joined: 27 Feb 2009
#88
hmmmm, I dont get the screen where you would do the F5 etc

yes,

i915.blacklist=yes
blacklist=i915

neither worked.

I will try xdrvr=radeon

Nope, caused panic. No worries. I don't plan to use this machine, but if you want me to test again with it just let me know. I just need to switch 4 cables from another box and put the cd in
Posts: 1,028
SamK
Joined: 21 Aug 2011
#89
antiX-14-a4-R_386-full.iso patched with a4-a5.delta

Tests conducted booting live USB with root persistence, using shipped defaults

Recording symptoms only (not investigated)


Prior to testing, system booted using persistent root & home
Wired networking set up via ceni
Changes saved at shutdown (automatic)


Tests show wired networking inconsistencies of subsequent re-boots.


Test 1
Boot param appended: None
Using Desktop=Default in boot screen (i.e. making no selection)
Network cable connected to system (i.e. wired and wireless)

Automatic Configuration of Network Successful
Automatic Configuration of Network Successful

Network automatic configuration successful


Test 2
Boot param appended: None
Using Desktop=Default in boot screen (i.e. making no selection)
Network cable connected to system (i.e. wired and wireless)

Automatic Configuration of Network Failure
Automatic Configuration of Network Failure

Network automatic configuration failed



The condition is unpredictable and cannot be recreated on demand. When the failure occurs (Test 2), networking can be successfully established via ceni.

I intended to attach the results of a third test but was unable to capture the state in the time available. In this state networking is automatically configured and works OK, however conky does not display the network monitors.


Reference Point
This system has run 13.2-Full-Stable (and continues to) since the day of its release. The above behaviour has never occurred with this antiX version. It has been entirely predictable and dependable throughout the period.


Addendum
Managed to get missing capture
Test 3
Boot param appended: None
Using Desktop=Default in boot screen (i.e. making no selection)
Network cable connected to system (i.e. wired and wireless)

Mixed State Networking & Conky
Mixed State Networking & Conky

Network automatic configuration successful
Conky display monitors failure
Last edited by SamK on 18 Feb 2015, 11:06, edited 1 time in total.
Posts: 1,308
BitJam
Joined: 31 Aug 2009
#90
thriftee wrote:I will try xdrvr=radeon

Nope, caused panic. No worries. I don't plan to use this machine, but if you want me to test again with it just let me know. I just need to switch 4 cables from another box and put the cd in
I suggest trying:

Code: Select all

nomodeset xdrvr=vesa
and then trying:

Code: Select all

nomodeset xdrvr=radeon