Re: some information for shadow 4.20.
| Newsgroups | gmane.linux.lfs.beyond.devel,gmane.linux.lfs.devel |
|---|---|
| Message-ID | <[email protected]> |
On 8/13/26 5:00 PM, Joe Locash ([email protected] via blfs-dev Mailing List) wrote: > On 8/13/26 4:04 PM, Bruce Dubbs ([email protected] via blfs-dev Mailing List) wrote: >> On 8/13/26 2:43 PM, Pierre Labastie ([email protected] via blfs-dev Mailing >> List) wrote: -- >>>> Yes, 'sudo su - <userid>' does show the same problem. >>>> >>> Have you tried to set /proc/sys/dev/tty/legacy_tiocsti to 0, as said in >>> one of the last comments of >>> https://github.com/shadow-maint/shadow/issues/1704 >>> ? >>> >>> We could do that when mounting /proc >> That also changes the value for the host. I don't know if that is an issue or not. >> >> I'm currently doing a full build with all tests and the latest shadow version. >> I notice on my current kernel I have CONFIG_LEGACY_TIOCSTI=y so /proc/sys/dev/tty/ >> legacy_tiocsti is set to 1. I didn't do that explicitly. It probably depends on >> the host distro. >> >> I'll see what the results are when it's done. I'm building at -j4 for timing so >> that will take a while. > I ran into this a few weeks ago. It's because CONFIG_LEGACY_TIOCSTI is enabled > in the kernel. It should be safe to disable in the kernel. For running kernels > set /proc/sys/dev/tty/legacy_tiocsti to 0. > > For sysv you can add "dev.tty.legacy_tiocsti = 0" to /etc/sysctl.conf I don't know why I am not seeing the problem. I did a full jhalfs build with all tests and everything appears to work OK with /proc/sys/dev/tty/legacy_tiocsti set to 1. For instance gawk is installed right after shadow in Chapter 8 and the gawk tests do su tester -c "PATH=$PATH make check" I checked logs/829-gawk-5.4.1 and test-logs/829-gawk-5.4.1. The only indication of something wrong is the line: 'egrep: warning: egrep is obsolescent; using grep -E' which we know about and is only because gawk's tests have not been updated. What I do see is "ALL TESTS PASSED". Note: This really should be a part of lfs-dev, not blfs-dev. Please answer at lfs-dev. -- Bruce -- http://lists.linuxfromscratch.org/sympa/info/blfs-dev Unsubscribe: See the above information page