Welcome, Guest. Please login or register.
Did you miss your activation email?

Author Topic: [EN] inxi bugs or failures - please report  (Read 35940 times)

Offline dsat

  • User
  • Posts: 130
[EN] Re: inxi bugs or failures - please report
« Reply #45 on: 2014/04/28, 20:38:26 »
Hi,

referring to the used percentage of my disc I suspect there might be still a bug:
Code: [Select]
System:    Host: x230t Kernel: 3.14-2.towo-siduction-amd64 x86_64 (64 bit gcc: 4.8.2)
           Desktop: KDE 4.12.4 (Qt 4.8.6) Distro: siduction 12.1 Desperado - kde - (201205212202)
Machine:   System: LENOVO product: 34382BG v: ThinkPad X230 Tablet serial: R9VREMH
           Mobo: LENOVO model: 34382BG serial: 1ZPA02C11M9
           Bios: LENOVO v: GCET63WW (2.03 ) date: 11/14/2012
CPU:       Dual core Intel Core i5-3320M CPU (-HT-MCP-) cache: 3072 KB
           flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 10379
           Clock Speeds: 1: 1204 MHz 2: 1204 MHz 3: 1204 MHz 4: 1204 MHz
Graphics:  Card: Intel 3rd Gen Core processor Graphics Controller bus-ID: 00:02.0
           Display Server: X.org 1.15.1 drivers: intel (unloaded: fbdev,vesa)
           tty size: 104x25 Advanced Data: N/A for root
Network:   Card-1: Intel 82579LM Gigabit Network Connection
           driver: e1000e v: 2.3.2-k port: 5080 bus-ID: 00:19.0
           IF: eth0 state: down mac: 3c:97:0e:5a:b2:69
           Card-2: Intel Centrino Advanced-N 6205 [Taylor Peak]
           driver: iwlwifi v: in-tree: bus-ID: 03:00.0
           IF: wlan0 state: up mac: 84:3a:4b:06:78:74
Drives:    HDD Total Size: 120.0GB (3446.5% used) ID-1: model: INTEL_SSDSC2CW12
Info:      Processes: 213 Uptime: 30 min Memory: 868.7/15755.2MB
           Init: systemd runlevel: 5 Gcc sys: 4.8.2 Client: Shell (bash 4.3.111) inxi: 2.1.23

Regards dsat

Offline dsat

  • User
  • Posts: 130
Re: inxi bugs or failures - please report
« Reply #46 on: 2014/04/28, 20:53:28 »
Hi,

sorry, it's not a bug, it's a feature: the space of my nfs-mount is included. When umounted the space used looks correct:
Code: [Select]
Drives:    HDD Total Size: 120.0GB (49.2% used) ID-1: model: INTEL_SSDSC2CW12

Regard dsat

Offline h2

  • User
  • Posts: 64
    • smxi.org
Re: inxi bugs or failures - please report
« Reply #47 on: 2014/04/28, 22:55:52 »
Of more general interest, inxi 2.1.24 fixes _some_ false reading issues for cpu temp.

This fix may break a few working setups, but sensors are very difficult, it's all guesswork in most cases.

If you think inxi is wrong where it was right, please post the output of: sensors
and the output of inxi 2.1.24: inxi -s

I believe _more_ wrong cases will be corrected than previously right cases but now wrong cases will be created.

There's a limit however to what I can get right and wrong, and of course, inxi has to be 'right' for my test systems first of all, ;-)

However, this handles a major bug with temp in msi/asus mobos, and also handles some bad temp data in other systems I have had the sensors output for.

I have added in temp3 readouts as well, though they are only used to get best cpu temp where temp2 temp2 and temp3 exist, they cannot be used to get mobo temp because temp3 is not reliable as a low value.

This fix handles the bad CPUTIN: 127C by using PECI 0 data in that case if there is no core0 data.

It's all hacks and guess work though in the end.

Offline h2

  • User
  • Posts: 64
    • smxi.org
Re: inxi bugs or failures - please report
« Reply #48 on: 2014/04/28, 23:04:51 »
No, that's a bug, re including remote mounts in the disk percent used.

Please generate a: inxi -xx@ 14
so I can look at it, or just paste in:

df -P -T --total

So I can see what was not excluded. make sure that thing is mounted, the one that caused the huge percent used result.

Again, that is a bug, probably caused by my using the --total feature of df.

I need to see the remote fs type.

Post as quickly as possible so I cna squeeze this into the 2.1.24 release.

Offline melmarker

  • User
  • Posts: 2.799
    • g-com.eu
Re: inxi bugs or failures - please report
« Reply #49 on: 2014/04/28, 23:07:16 »
inxi_2.1.24.r2265.1_all.deb is in the repo ...
Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. (Benjamin Franklin, November 11, 1755)
Never attribute to malice that which can be adequately explained by stupidity. (Hanlons razor)

Offline h2

  • User
  • Posts: 64
    • smxi.org
Re: inxi bugs or failures - please report
« Reply #50 on: 2014/04/28, 23:17:28 »
Heh, you were too fast, I just fixed the bug with including nfs/nfs4/smbfs sizes in the -D disk used output.

I didn't boost the version number since 2.1.24 was just released.

But these are exactly the types of bug reports I was hoping to get.

I am now wondering about using --total at all, because every non anticipated file system can break it if it's not explicitly excluded.

that's what caused this error, the non --total method checks to see that it's a dev mounted file system before adding it.

For now the issue is corrected, but only for nfs/smbfs/nfs4 but it's a poorly future proofed method in my opinion, maybe I should drop the --total method.

Offline melmarker

  • User
  • Posts: 2.799
    • g-com.eu
Re: inxi bugs or failures - please report
« Reply #51 on: 2014/04/28, 23:19:50 »
h2 - sorry, read your second post after uploading  - but a inxi upload is fast and cheap so this should be no problem :)

EDIT: can i upload again or is this still work in progress?
Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. (Benjamin Franklin, November 11, 1755)
Never attribute to malice that which can be adequately explained by stupidity. (Hanlons razor)

Offline h2

  • User
  • Posts: 64
    • smxi.org
Re: inxi bugs or failures - please report
« Reply #52 on: 2014/04/28, 23:30:11 »
I decided that this false used bug report was too serious and exposed a serious lack of future proofing, so I removed using --total from inxi, now it only uses partitions that are found in /dev somewhere.

To make sure this mistake dies, I did 2.1.25.

I am excluding those fs types now as well, but that just changes how many loops gawk has to do, nothing else.

thanks for catching that one, I knew --total was too good to be true, and it is.

2.1.25 is stable since it just removes the method. That's two days in row I did two numbered releases, now it's time for work though.

<update> after checking some data, I decided to not use temp3, that item is very unreliable.

There may be other sensors issues, but do keep in mind, some sensors output combinations simply cannot be guessed at without breaking others instantly, so inxi needs to try to be right more than it is wrong in such fringe cases, that's the best it can do.

But always post sensors output and inxi -s output if you think inxi is wrong, sometimes I can find a pattern that is safe to handle without breaking other things, that's how I got PECI, temp3, and Core0 cpu temp overrides working more or less now for example
« Last Edit: 2014/04/28, 23:49:59 by h2 »

Offline melmarker

  • User
  • Posts: 2.799
    • g-com.eu
Re: inxi bugs or failures - please report
« Reply #53 on: 2014/04/29, 01:51:58 »
inxi_2.1.25.r2269.1_all.deb  is in the repo
Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. (Benjamin Franklin, November 11, 1755)
Never attribute to malice that which can be adequately explained by stupidity. (Hanlons razor)

Offline musca

  • User
  • Posts: 725
  • sid, fly high!
Re: virtualization report - add to wish list,pls
« Reply #54 on: 2014/04/29, 12:02:26 »
Hello h2,

we enjoy your nice work almost every single day, thank you so much!

Some time ago we talked about Virtualization support and at that time it was difficult to do without root.
Ofcourse there are some indirect hints like type of Graphics or Motherboard.

But there is the package imvirt in sid which now seems to do it without root, e.g.

Code: [Select]
$ imvirt
VirtualBox

$imvirt
KVM

$ imvirt
Physical

and i would like to see such a hint even in the short output of `inxi` without options.

Please add this to the wish list for inxi 3.0   ;)

greetings
musca
« Last Edit: 2014/04/29, 12:16:33 by musca »
„Es irrt der Mensch, solang er strebt.“  (Goethe, Faust)

Offline dsat

  • User
  • Posts: 130
Re: inxi bugs or failures - please report
« Reply #55 on: 2014/04/29, 17:48:57 »
Hi,

here is the output:
Code: [Select]
root@x230t:# df -P -T --total
Dateisystem            Typ      1024-Blöcke    Benutzt  Verfügbar Kapazität Eingehängt auf
/dev/sda1              ext4        15995848    8740336    6419928       58% /
udev                   devtmpfs       10240          0      10240        0% /dev
tmpfs                  tmpfs        3226676       3312    3223364        1% /run
tmpfs                  tmpfs        8066688          0    8066688        0% /dev/shm
tmpfs                  tmpfs        8066688          0    8066688        0% /sys/fs/cgroup
192.168.x.x:/media/md0 nfs       5813963776 3976600576 1544334336       73% /media/<directory>
tmpfs                  tmpfs         102400          4     102396        1% /run/user
tmpfs                  tmpfs           5120          0       5120        0% /run/lock
/dev/sda2              ext4          991512     293244     630684       32% /media/System
/dev/sda3              xfs         75739000   48597612   27141388       65% /media/Storage
total                  -         5926167948 4034235084 1598000832       72% -

inxi is up to date and gives:
Code: [Select]
Drives:    HDD Total Size: 120.0GB (49.2% used) ID-1: model: INTEL_SSDSC2CW12
Info:      Processes: 196 Uptime: 8 min Memory: 822.7/15755.2MB
           Init: systemd runlevel: 5 Gcc sys: 4.8.2 Client: Shell (bash 4.3.111) inxi: 2.1.25

Regards dsat

Offline h2

  • User
  • Posts: 64
    • smxi.org
Re: inxi bugs or failures - please report
« Reply #56 on: 2014/04/29, 21:06:45 »
musca, thanks, however, in inxi, I try to keep to an absolute minimum recommended packages, always preferring to query the system directly to get the data, it's more reliable and means it will work for most users right away.

If simple raw virtualization detection is required, all that is needed is a set of checks, for example, vbox has a distinct id in the name of its devices, disks, etc, that is easy to spot.

vmware I haven't seen for a while, but doing inxi -Fxx on any system should show the distinct identifiers used in some cases.

However, you'd have to get in line, 2.2 is reserved for the long waited for -m option, memory, I was hoping that data would appear in /sys but I guess it isn't, so that will require root and dmidecode, sadly. And 3.0 is far far away, that's 9 new major features, I don't know if a bash+gawk script can get that big to be honest.

I won't have time for any other things in the coming months though I think, I have never found a major feature that requires less than 2 weeks of datacollection at least, plus coding, plus fixing exceptions, so I always have to give careful thought to any new feature.

However, people can help this process by collecting data on the feature type, all possible outputs, data sources, etc.

imvirt is getting its data from somewhere, so all you have to do is figure out some method to do that test for each vm type. My guess is a short function would do it. kvm might be harder because it has more options, like using host kernel, etc.
« Last Edit: 2014/04/29, 21:16:59 by h2 »

Offline melmarker

  • User
  • Posts: 2.799
    • g-com.eu
Re: inxi bugs or failures - please report
« Reply #57 on: 2014/05/04, 03:56:32 »
hi h2 - a special wish: is it possible, that you let inxi detect LXQt? That would be nice for our planned Release at LinuxTag Berlin
Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. (Benjamin Franklin, November 11, 1755)
Never attribute to malice that which can be adequately explained by stupidity. (Hanlons razor)

Offline h2

  • User
  • Posts: 64
    • smxi.org
Re: inxi bugs or failures - please report
« Reply #58 on: 2014/05/04, 19:33:09 »
to add a desktop identification, I need a few things:

latest: inxi -xx@ 14
from the system running that desktop.

That gives me the data required to see if the standard methods detect it.

Then, I need the actual process name..

let's say it's lxqt

then I need how to obtain its version data: lxqt --version
for example, the actual command, and then I  need the actual literal output of that command

Then, because I've found some fringe cases, I further need this, assuming it's called lxqt:

lxqt --version && echo goes to stdout || echo goes to stderr

this is because in a few cases desktops decide to putput their version data to stderr. weird, silly, but it happens.

Once I have this data I can add support.

I also need to know if the desktop is a derived type, like some are, where the actual id is openbox, for example, inxi in those cases finds the process then queries the primary thing, like openbox, for version data.


Offline h2

  • User
  • Posts: 64
    • smxi.org
Re: inxi bugs or failures - please report
« Reply #59 on: 2014/05/05, 19:30:53 »
If you guys want this, you'll have to do the research and provide all the data required.

Here's what some siduction person's -xx@14 showed me

inxi shows LXDE (Openbox v...)

xprop -root shows the primary container is Openbox

$XDG_CURRENT_DESKTOP is 'Razor'

in ps aux, no razor-desktop appears, but lxqt-session does.

startlxde-qt is present which is why it shows as lxde.

so to get anywhere, I need this:

razor-desktop --version
lxqt-session --version

or any other command that gives version info, like -v.

If not null, post the literal output. if null, post that fact too.

Then someone who understands what the difference is here between razor and lxqt is to post the technical explanations.

Failing this, I'll just leave this alone until it's clarified.

inxi as is will show this as lxde because of the startlxde-qt in the ps aux, but I don't remember what lxde itself looks like in ps aux, though I believe

$XDG_CURRENT_DESKTOP is 'LXDE' but I also believe I could not rely on XDG beiing set, so I added in a fallback ps aux test.

So you guys will have to figure it out then post the literal ways to differentiate the three, lxde, qt-razor, and lx-qt.

with all verison tests etc.

Once i have that, I'll look at the problem again.