Fast news: @MX_Linux

The Forum is now on mxlinux.org and uses an encrypted (https) connection

Please read this Important information: Linux kernel exploit

  • antiX-16, release info here -- watch the video. Torrents here
  • MX-15 (antiX MX), release info here: here. Monthly update/upgrade available here.
  • Older MX Linux releases available on the download page

New users
  • Please read this first, and don't forget to add system and hardware information to posts!

new inxi features of possible interest

Forum for discussion and implementation of features relating directly to development of MEPIS releases.
User avatar
Forum Regular
Forum Regular
Posts: 273
Joined: Sat Nov 15, 2008 3:16 pm

Re: new inxi features of possible interest

#41 Postby h2-1 » Sun May 01, 2011 6:52 pm


Code: Select all

state: unknown

Is the actual content of the state file, I get that from lenny 2.6.26 kernels too. If you see anything other than N/A, the file actually had contents, and what you see is the contents, if you see N/A inxi had no value for that, and simply shows N/A to indicate no content. In other parts, I simply do not show anything when the value is null, but in this case, I definitely want to see when it's null so I can see bugs and get them fixed

So state: unknown is in fact the value given, which is 'unknown'. But it's odd that you have speed and duplex with value, and the state contains 'unknown', I've only seen that be unknown when speed and duplex were null. But there will be huge variance.

But I appreciate the various samples of what the output can be, when working off something as difficult as this, where I cannot just inject user data files to get values, the only way to see what different systems return is to see what different systems return. But yours is working perfectly, all data files contained values.

User avatar
Forum Guide
Forum Guide
Posts: 2485
Joined: Wed Jul 12, 2006 4:43 am

Re: new inxi features of possible interest

#42 Postby ko » Mon May 02, 2011 5:59 am

h2-1 wrote:ko, no, you have something else going on, I'll need to see:

First, download this file: http://smxi.org/opt/tree-traverse.py
save it to some directory. That's a script that will show us the contents of your /sys directory, so we can find where the actual locations of the usb wifi data is.

To run it, do: chmod +x tree-traverse.py
then do: ./tree-traverse.py >> sys-tree.txt

from inside the directory, this will create the text file we need.

After you have created it, upload it somewhere, like pastebin, and then send me a link to the file.

note, zip or gzip the file before uploading it somewhere, can be here.

Any time you see both IF: and mac: blank, it means the logic failed. speed/duplex might be blank, ie, N/A, but not IF or mac.

The cases of failure are what I need, so if I can get that data file I'll see what's going wrong there. We may integrate this into inxi, not sure yet, ie, run a command, like: inxi -@ 12, and it will collect all the data, zip it into a file and then you can upload it somewhere. But one thing at a time.

Also let me know what kernel you are using, this is highly kernel dependent data, so it's good to get a sense of which kernels give what data, and which are missing what.

Sorry for the long quote, but I get a syntax error when running the tree-traverse.py script:

Code: Select all

[ko@laptop Downloads]$ chmod +x tree-traverse.py
[ko@laptop Downloads]$  ./tree-traverse.py >> sys-tree.txt
./tree-traverse.py:10: Warning: 'with' will become a reserved keyword in Python 2.6
  File "./tree-traverse.py", line 10
    with open(fullPathFile, "r") as someFile:
SyntaxError: invalid syntax
[ko@laptop Downloads]$

I'm running Mepis 8.5 32bit with the 2.6.32-1-mepis-smp kernel. I compiled the .deb for my DWA140 (and blacklisted the opensource rt2x00 and rt2800 driver modules and 'decommisioned' the rt2870sta staging driver). I cannot use the Mepis 2.6.36 kernel with the DWA140 wireless device as the latest source code that I used from Ralinktech still does not support the 2.6.36 kernel.

The sys-tree.txt file created has size 0 bytes.
Ko Bros Linux User #312596.  MX14.4, MX14.3, Mepis 12, Debian 7, antiX 15, Peppermint 5...and sometimes Win7

User avatar
Forum Regular
Forum Regular
Posts: 273
Joined: Sat Nov 15, 2008 3:16 pm

Re: new inxi features of possible interest

#43 Postby h2-1 » Mon May 02, 2011 5:03 pm

Card-2 Intel PRO/Wireless 3945ABG [Golan] Network Connection bus-ID: 02:00.0
IF: N/A state: N/A speed: N/A duplex: N/A mac: N/A

No matter the status of the DWA chip, Intel should in no case be blank, so that's definitely a real bug.

I will ask trash80 who wrote the python script what's wrong there. He has a newer version up but it may have the same issue.

Forum Novice
Forum  Novice
Posts: 1
Joined: Mon May 02, 2011 8:04 pm

Re: new inxi features of possible interest

#44 Postby trash80 » Mon May 02, 2011 8:15 pm

tree-traverse.py and xiin.py(newer version) both require python 2.6+ and will work with the python 3.x branch, also.

I'll request that xiin be used instead of tree-traverse (tt). tt was a quick and dirty script for me, after some feature requests by h2, I made it little more user friendly and renamed it xiin. (It's a play on ixni, invert the 'i's and reverse the 'x' and 'n', so you have inxi inside out and backwards). (pronounced 'zen')

feel free to submit bugs and/or feature requests at http://www.inxi.org or to me at myname@myname.org (that's right, replace myname with my name ;) )

Although I won't support older versions of python, I will accept patches from others that would like to.
Last edited by trash80 on Mon May 02, 2011 8:16 pm, edited 1 time in total.

Return to “Dev Team”

Who is online

Users browsing this forum: No registered users and 2 guests