Bug#920618: initscripts: errors with /dev/shm on Hurd
Svante Signell, le sam. 02 févr. 2019 11:25:59 +0100, a ecrit: > On Sat, 2019-02-02 at 06:41 +0100, Samuel Thibault wrote: > > Svante Signell, le sam. 02 févr. 2019 01:07:48 +0100, a ecrit: > > > > > > I'm having a look :) > > > > Thanks! :) > > Seem to work fine, Thanks! Now pending upload. > after removing the broken links /run/shm -> /dev/shm and > /dev/shm -> /run/shm. /run is a tmpfs anyway, so the link would go away anyway. Samuel
Bug#920618: initscripts: errors with /dev/shm on Hurd
[2019-02-02 11:25] Svante Signell > I don't boot with a serial console so I might have missed som output, e.g. the > brightness issue. For what it worth, code in `brightness' script check two possible location is copy-paste fashion. I have only one of those locations, and things work for me. -- Note, that I send and fetch email in batch, once every 24 hours. If matter is urgent, try https://t.me/kaction --
Bug#920618: initscripts: errors with /dev/shm on Hurd
Dmitry Bogatov, le mer. 30 janv. 2019 20:48:56 +, a ecrit: > [2019-01-28 13:40] Samuel Thibault > > > Do you mean, that you plan to resolve current bug via changes in > > > src:hurd? > > > > That's be the idea yes. > > > > > > The same patch creates /run/lock and /run/utmp, I guess initscripts > > > > handles that too? > > > > > > Yes, initscripts handle them too. See /etc/init.d/bootmisc.sh > > > > Ok, can somebody check that things work fine when dropping that part > > too? > > I wouldn't be me, sorry. I wasn't thinking about you :) I was thinking about people on debian-hurd@, who have a running hurd box, and can try to modify these scripts and see if that works fine. People should really not hope that I'll magically have the time to do everything, that will just not work. Samuel
Bug#920618: initscripts: errors with /dev/shm on Hurd
[2019-01-28 13:40] Samuel Thibault > > Do you mean, that you plan to resolve current bug via changes in > > src:hurd? > > That's be the idea yes. > > > > The same patch creates /run/lock and /run/utmp, I guess initscripts > > > handles that too? > > > > Yes, initscripts handle them too. See /etc/init.d/bootmisc.sh > > Ok, can somebody check that things work fine when dropping that part > too? I wouldn't be me, sorry. I start to lose track of recent Hurd issues. As I understand, this bug should be reassigned to src:hurd. If there are more issues in sysvinit=2.93-6, please file them to BTS. -- Note, that I send and fetch email in batch, once every 24 hours. If matter is urgent, try https://t.me/kaction --
Bug#920618: initscripts: errors with /dev/shm on Hurd
[2019-01-27 15:03] Samuel Thibault > Dmitry Bogatov, le dim. 27 janv. 2019 13:52:59 +, a ecrit: > > /run/shm is created by the hurd package, causing a conflict with > > initscripts. > > rgrep /run/shm /etc/hurd > > /etc/hurd/rc:mkdir -p /run/lock /run/shm > > > > dpkg -S /etc/hurd/rc > > hurd: /etc/hurd/rc > > > > - Either the creation of /run/shm in hurd should be removed or creation > > of the > > symbolic link /run/shm in initscripts should be removed for Hurd. > > I guess the shm part of debian/patches/run.patch can now be dropped > indeed. Anybody up to test that it boots fine and the issue goes away? Do you mean, that you plan to resolve current bug via changes in src:hurd? > The same patch creates /run/lock and /run/utmp, I guess initscripts > handles that too? Yes, initscripts handle them too. See /etc/init.d/bootmisc.sh -- Note, that I send and fetch email in batch, once every 24 hours. If matter is urgent, try https://t.me/kaction --
Bug#920618: initscripts: errors with /dev/shm on Hurd
Dmitry Bogatov, le dim. 27 janv. 2019 13:52:59 +, a ecrit: > /run/shm is created by the hurd package, causing a conflict with > initscripts. > rgrep /run/shm /etc/hurd > /etc/hurd/rc:mkdir -p /run/lock /run/shm > > dpkg -S /etc/hurd/rc > hurd: /etc/hurd/rc > > - Either the creation of /run/shm in hurd should be removed or creation > of the > symbolic link /run/shm in initscripts should be removed for Hurd. I guess the shm part of debian/patches/run.patch can now be dropped indeed. Anybody up to test that it boots fine and the issue goes away? The same patch creates /run/lock and /run/utmp, I guess initscripts handles that too? Samuel
Bug#920618: initscripts: errors with /dev/shm on Hurd
Package: initscripts Version: 2.93-5 Severity: wishlist [ Originally reported by Svante Signell ] Hi, we have some problems with the latest fixes of sysvinit for GNU/Hurd. * Make /run/shm symlink to /dev/shm, not other way around (Closes: #851427) Boot messages: mkdir: Cannot create directory /dev/shm: File exists Mount point '/dev/shm' does not exist. Skipping mount. touch: cannot touch '/dev/shm/.tmpfs': Too many levels of symbolic links. Cleaning up temporary files... failed! startpar: service(s) returned failure: checkroot-bootclean.sh mountall- bootclean.sh mountnfs.sh ... failed! /run/shm is created by the hurd package, causing a conflict with initscripts. rgrep /run/shm /etc/hurd /etc/hurd/rc:mkdir -p /run/lock /run/shm dpkg -S /etc/hurd/rc hurd: /etc/hurd/rc - Either the creation of /run/shm in hurd should be removed or creation of the symbolic link /run/shm in initscripts should be removed for Hurd. /dev/shm is already a symbolic link to /run/shm: file /dev/shm /dev/shm: symbolic link to /run/shm ls -l /dev/shm lrwxr-xr-x 1 root root 8 Jan 20 2017 /dev/shm -> /run/shm pgpMQXXRcfh6Z.pgp Description: PGP signature
