You are not logged in.
Since the upgrade to 210-3 I'm getting more verbose output on the console after an error has occurred with systemd-remount-fs. This seems to occur sometimes, but not all the time. Looks like a race condition.
When failed, /run is mounted like this:
tmpfs on /run type tmpfs (rw,nosuid,nodev,mode=755)
when succeeded, /run is mounted like this:
tmpfs on /run type tmpfs (rw,nosuid,nodev,noexec,noatime,size=5120k,mode=755)
Running /usr/lib/systemd/systemd-remount-fs afterwards alleviates the problem. I did a strace, the mount() call from the binary looks okay:
[mount][pid 2411] mount("run", "/run", 0x986d978, MS_MGC_VAL|MS_NOSUID|MS_NODEV|MS_NOEXEC|MS_REMOUNT|MS_NOATIME, "mode=0755,size=5m") = 0
Below is a failure:
-- Reboot --
mrt 10 15:29:19 delta systemd[1]: Starting Remount Root and Kernel File Systems...
-- Subject: Unit systemd-remount-fs.service has begun with start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/li … temd-devel
--
-- Unit systemd-remount-fs.service has begun starting up.
mrt 10 15:29:19 delta systemd-remount-fs[422]: /bin/mount for /run exited with exit status
mrt 10 15:29:19 delta systemd-remount-fs[422]: mount: /run not mounted or bad option
mrt 10 15:29:19 delta systemd-remount-fs[422]: In some cases useful info is found in syslog
mrt 10 15:29:19 delta systemd-remount-fs[422]: dmesg | tail or so.
mrt 10 15:29:19 delta systemd[1]: systemd-remount-fs.service: main process exited, code=exi
mrt 10 15:29:19 delta systemd[1]: Failed to start Remount Root and Kernel File Systems.
-- Subject: Unit systemd-remount-fs.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/li … temd-devel
--
-- Unit systemd-remount-fs.service has failed.
--
-- The result is failed.
mrt 10 15:29:19 delta systemd[1]: Unit systemd-remount-fs.service entered failed state.
mrt 11 06:50:50 delta systemd-remount-fs[428]: /bin/mount for /run exited with exit status
mrt 11 06:50:50 delta systemd-remount-fs[428]: mount: /run not mounted or bad option
mrt 11 06:50:50 delta systemd-remount-fs[428]: In some cases useful info is found in syslog
mrt 11 06:50:50 delta systemd-remount-fs[428]: dmesg | tail or so.
Here is a succes:
-- Reboot --
mrt 10 15:35:54 delta systemd[1]: Starting Remount Root and Kernel File Systems...
-- Subject: Unit systemd-remount-fs.service has begun with start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/li … temd-devel
--
-- Unit systemd-remount-fs.service has begun starting up.
mrt 10 15:35:54 delta systemd[1]: Started Remount Root and Kernel File Systems.
-- Subject: Unit systemd-remount-fs.service has finished start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/li … temd-devel
--
-- Unit systemd-remount-fs.service has finished starting up.
--
-- The start-up result is done.
Is anyone else suffering from this?
Last edited by Rexilion (2014-03-11 06:11:03)
fs/super.c : "Self-destruct in 5 seconds. Have a nice day...\n",
Offline
So, the system affected by this is a 32bit UP system. It's like 10,5 years old. It has a custom kernel, but so does the other laptop that is not affected by this issue.
Two other laptop's both running x64 which are slightly more recent (7 and maybe 4 years). These are not suffering from this issue.
EDIT: And yes, they use the exact same fstab for mounting run.
Last edited by Rexilion (2014-03-16 13:37:30)
fs/super.c : "Self-destruct in 5 seconds. Have a nice day...\n",
Offline
Ow wait, it happens on my laptop as well. Both the laptop and the machine this was first observed on have a custom stripped down kernel. The second laptop where this issue is not observed (I just double checked) has a stock kernel and no issue's.
So it's the custom kernel I guess. How odd...
Last edited by Rexilion (2014-03-16 13:55:01)
fs/super.c : "Self-destruct in 5 seconds. Have a nice day...\n",
Offline
Replaced a binary with a bascript that called strace.
[pid 459] mount("run", "/run", 0x1a0b240, MS_MGC_VAL|MS_NOSUID|MS_NODEV|MS_NOEXEC|MS_REMOUNT|MS_NOATIME, "mode=0755,size=5m") = -1 EINVAL (Invalid argument) <0.000039>
[mount][pid 2411] mount("run", "/run", 0x986d978, MS_MGC_VAL|MS_NOSUID|MS_NODEV|MS_NOEXEC|MS_REMOUNT|MS_NOATIME, "mode=0755,size=5m") = 0
Eh?
EINVAL
source had an invalid superblock. Or, a remount (MS_REMOUNT) was attempted, but source was not already mounted on target. Or, a move (MS_MOVE) was attempted, but source was not a mount point, or was '/'.
int mount(const char *source, const char *target,
const char *filesystemtype, unsigned long mountflags,
const void *data);
I guess /run is not attached in time.
EDIT: Just prefixed a mounpoint in my systemd-remount-fs replacement script. It says that /run is a mountpoint. This just got weird. I'll guess I update my kernel someday and try again.
Last edited by Rexilion (2014-03-16 14:37:29)
fs/super.c : "Self-destruct in 5 seconds. Have a nice day...\n",
Offline
I opened a bug here.
I'm starting to think this is somehow more kernel related. Something that the systemd upgrade exposed.
fs/super.c : "Self-destruct in 5 seconds. Have a nice day...\n",
Offline