Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • 14:38

    oxr463 on master

    Add link to usage survey [skip… (compare)

  • Jan 17 14:24
    KimonHoffmann commented #342
  • Jan 17 10:16
    KimonHoffmann commented #342
  • Jan 17 09:07
    avian2 commented #319
  • Jan 16 16:51
    oxr463 milestoned #342
  • Jan 16 16:51
    oxr463 milestoned #342
  • Jan 16 16:50
    oxr463 assigned #342
  • Jan 16 16:50
    oxr463 commented #342
  • Jan 16 15:17
    marzban2030 commented #94
  • Jan 16 15:15
    marzban2030 commented #94
  • Dec 31 2022 18:07
    oxr463 closed #341
  • Dec 31 2022 00:27
    FredericGuilbault opened #341
  • Dec 26 2022 18:15
    m13253 commented #334
  • Oct 29 2022 19:32
    oxr463 commented #339
  • Oct 29 2022 18:06
    oxr463 labeled #338
  • Oct 29 2022 18:06

    oxr463 on master

    Add: faccess2(2) syscall Remove extra tab for PR_faccess… (compare)

  • Oct 29 2022 18:06
    oxr463 closed #338
  • Oct 29 2022 17:34
    oxr463 commented #340
  • Oct 29 2022 17:32
    oxr463 milestoned #338
  • Oct 29 2022 17:32
    oxr463 milestoned #338
oxr463
@oxr463:matrix.org
[m]
But if a program doesn't allow ptrace then it won't work
So you can use gdb or strace on PRoot itself... but depends on what runs inside the PRoot if you can ptrace it
pie_
@jcie74:matrix.org
[m]
uhuh
pie_
@jcie74:matrix.org
[m]
oxr463: proot doesnt translate (bind) mount does it?
1 reply
pie_
@jcie74:matrix.org
[m]
well half my problem was I kept messing up a file name, so I did end up solving it with the fake mounts, but I was more wondering if (I really should just check with -v) internal mount --bind could /is be caught and emulated (if that makes sense)
1 reply
pie_
@jcie74:matrix.org
[m]
oxr463: any idea what will break when trying to use proot on a 2.6.32 kernel? (2013 build)
1 reply
oxr463: another suggestion: allow specifying a file to output -v output to
pie_
@jcie74:matrix.org
[m]
proot info: vpid 47: translate("/" + "/nix/store/ga9q4ikprn36wm32gqm4k521iv9kic5p-libsodium-1.0.18/lib/libsodium.so.23")
proot info: vpid 47:          -> "/Lustre01/home/qosnebc/pls/store/ga9q4ikprn36wm32gqm4k521iv9kic5p-libsodium-1.0.18/lib/libsodium.so.23.3.0"
proot info: vpid 47: sysenter start: read(0x3, 0x7fff0e468db8, 0x340, 0x0, 0x7fff0e468d9f, 0x0) = 0xffffffffffffffda [0x7fff0e468ad8, 0]
proot info: vpid 47: sysenter start: newfstatat(0x3, 0x6f0000028954, 0x7fff0e468c40, 0x1000, 0x7f911fecdf30, 0x6f0000033220) = 0xffffffffffffffda [0x7fff0e468bb8, 0]
proot info: vpid 47: sysenter start: close(0x3, 0x6f0000028954, 0x7fff0e468c40, 0x1000, 0x7f911fecdf30, 0x6f0000033220) = 0xffffffffffffffda [0x7fff0e468bb8, 0]
proot info: vpid 47: sysenter start: writev(0x2, 0x7fff0e468900, 0xa, 0x20, 0x7fff0e468d48, 0x7f911fecdf30) = 0xffffffffffffffda [0x7fff0e4688c0, 0]
/nix/store/xdlpraypxdimjyfrr4k06narrv8nmfgh-nix-2.11.1/bin/nix-store: error while loading shared libraries: libsodium.so.23: cannot stat shared object: Invalid argument
proot info: vpid 47: sysenter start: exit_group(0x7f, 0x3c, 0x7f, 0x20, 0xe7, 0x7f911fecdf30) = 0xffffffffffffffda [0x7fff0e468d38, 0]
proot info: vpid 47: exited with status 127
Does that make any sense to you?
Well, I guess it's not really a proot issue;
$ LD_LIBRARY_PATH=pls/store/ga9q4ikprn36wm32gqm4k521iv9kic5p-libsodium-1.0.18/lib/ pls/store/scd5n7xsn0hh0lvhhnycr9gx0h8xfzsl-glibc-2.34-210/lib64/ld-linux-x86-64.so.2 pls/store/xdlpraypxdimjyfrr4k06narrv8nmfgh-nix-2.11.1/bin/nix-store
pls/store/xdlpraypxdimjyfrr4k06narrv8nmfgh-nix-2.11.1/bin/nix-store: error while loading shared libraries: libsodium.so.23: cannot stat shared object: Invalid argument
oxr463
@oxr463:matrix.org
[m]
Like a verbose log file?
pie_
@jcie74:matrix.org
[m]
also reasonable
re log: yeah
IIRC strace has something like it for example
pie_
@jcie74:matrix.org
[m]
two libuv tests failed under proot but im not sure if its actually proot's fault
specifically fs_file_enametoolong and signal_multiple_loops
just mentioning it
pie_
@jcie74:matrix.org
[m]
Is this a proot issue?
ECC:
 Ed25519        |  nanosecs/iter   cycles/iter
           mult | Fatal: getentropy is not supported: Function not implemented
Hm ok seems like a kernel syscall interface issue I guess
hm
(old kernel)
Though I think this should be going through musl so it should work...hm...
pie_
@jcie74:matrix.org
[m]
oxr463: do you know if that is a proot issue? ^
oxr463
@oxr463:matrix.org
[m]
I don't think so?
pie_
@jcie74:matrix.org
[m]
hm
pie_
@jcie74:matrix.org
[m]
yeah so it was a kernel syscall / libc thing
I used a patch to musl to get a version of it
I'm running into some other issues now
angyongen
@angyongen
hi trying to compile proot for the first time. using the command "make -C src loader.elf loader-m32.elf build.h". anyone know why does it say "make: * No rule to make target 'loader-m32.elf'. Stop."?????
oxr463
@oxr463:matrix.org
[m]
On the master branch?
angyongen
@angyongen
Yes. Tried the latest release as well (5.3.1). Took me a while to figure out how to install the dependencies (configure; make; make install; then later i found out that apt-get works too by looking at the github ci), and it seems to compile with a few warnings? But in the last step many tests failed. Is this normal?
The make rule message was for both the master branch and latest release
Can i assume the "make: * No rule to make target 'loader-m32.elf" is normal too?
oxr463
@oxr463:matrix.org
[m]
Yeah, lots of tests need to be fixed
I've seen that loader-m32.elf before but I can't remember the resolution
angyongen
@angyongen
Oh no.... then how will i know whether my build is broken or not? Or if i made some changes that break something? Or if the loader-m32 is causing some failed tests?
1 reply
Btw i think the apt get command should be added to the section about dependencies so others can know how to install them easily
1 reply
(Can anyone tell me which tests are supposed to pass right now?)
angyongen
@angyongen
Also i couldn’t find the descriptions about each test, right now each test has some hex code after it, and i do see issues that refer to these hex codes, but the tests dont seem to refer back to the issues. Not sure if there is a list or table somewhere to keep track of what each test does? Then maybe at least know which features are not working?
oxr463
@oxr463:matrix.org
[m]
Here is a list of the tests, proot-me/proot#164
The "hex" is actually a git commit ID
angyongen
@angyongen
Oh, thanks! However some files have hex codes that are not valid git commit IDs? Do i need to search through the github commits manually to figure out what they do?
Or maybe looking through the code is faster...?
angyongen
@angyongen
(For the valid git commit ids, I found this command to print a description. Perhaps it can be automated with a script to print all the descriptions. "git show --no-notes --oneline --no-patch <commit>")
oxr463
@oxr463:matrix.org
[m]
I think those are just fake lol
It would be nice to rename all the tests, but that's a lot of effort
Frederic Guilbault
@FredericGuilbault

Hi im having a file permission problem ( show d????????? instead of rwxr-xr-x ) inside proot since I changed the host operating system.

I was on Mint 20 ext4 everything was working fine.
I copied my code base(iso and scritps that use proot ) on an external drive (btrfs) and imported it on a fresh install of mint21.1( ubuntu 22.04) with kernel 5.15.0-56 ext4

My script mount the iso file add folders and file to the iso then use proot to run the files. All files living inside the ISO file have correct permissions set, but others files that have been copied from the host don't have correct permissions


total 76K
drwxr-xr-x  20 root root 4.0K Dec 25 19:02 .
drwxr-xr-x  20 root root 4.0K Dec 25 19:02 ..
d?????????   ? ?    ?       ?            ? apt
lrwxrwxrwx   1 root root    7 Dec 25 15:10 bin -> usr/bin
drwxr-xr-x   5 root root 4.0K Dec 25 20:22 boot
drwxr-xr-x  21 root root 4.8K Dec 25 21:38 dev
drwxr-xr-x 150 root root  12K Dec 25 17:06 etc
drwxr-xr-x   3 root root 4.0K Dec 25 15:14 home
d?????????   ? ?    ?       ?            ? host-rootfs
d?????????   ? ?    ?       ?            ? install-scripts
lrwxrwxrwx   1 root root    7 Dec 25 15:10 lib -> usr/lib
drwx------   2 root root  16K Dec 25 15:09 lost+found
drwxr-xr-x   3 root root 4.0K Dec 25 15:19 media
drwxr-xr-x   5 root root 4.0K Dec 25 16:36 mnt
drwxr-xr-x   4 root root 4.0K Dec 25 16:44 opt
dr-xr-xr-x 431 root root    0 Dec 25 21:27 proc
drwx------   8 root root 4.0K Dec 25 19:02 root
drwxr-xr-x  37 root root 1.1K Dec 25 22:00 run
lrwxrwxrwx   1 root root    8 Dec 25 15:10 sbin -> usr/sbin
drwxr-xr-x   2 root root 4.0K Dec 17 15:08 srv
dr-xr-xr-x  13 root root    0 Dec 25 21:35 sys
drwxrwxrwt  20 root root 4.0K Dec 25 22:50 tmp
drwxr-xr-x  14 root root 4.0K Dec 17 15:08 usr
drwxr-xr-x  11 root root 4.0K Dec 17 15:43 var

My proot command looks like this :
cp -r "../install-scripts" "$rootfs/"
cp -r "/etc/apt" "$rootfs/"

  if [[ $cpuArch == arm64 ]]; then
      qemuArch=" -q qemu-aarch64"
  elif [[ $cpuArch == armhf ]]; then
      qemuArch=" -q qemu-arm"
  fi
  proot $qemuArch \
    --root-id \
    --rootfs=$workDir/rootfs \
    --cwd=/install-scripts \
    --mount=/etc/resolv.conf:/etc/resolv.conf \
    --mount=/dev:/dev \
    --mount=/sys:/sys \
    --mount=/proc:/proc \
    $buildCmd

  return=$?
  [ $return -ne 0 ] && exit 1

```

Did I find a bug or im doing something wrong ?

Frederic Guilbault
@FredericGuilbault
oh and overlayfs is involved and both old and new install run proot version 5.1.0 :)
oxr463
@oxr463:matrix.org
[m]
Huh, that's odd
Probably a bug