If the system reaches a defined limit of processes in running state, udevd
starts to count its own processes in running state from its session (all
forked hotplug child processes, subprocesses and callouts) and throttles
further process forking if the limit is reached.
This should help setups with hundreds of events emitted hotplug events
in parallel with hundreds of processes in "R" state. which makes the machine
unresponsible.
I placed a 100% cpu time consuming program in /etc/hotplug.d/ which runs for 5
seconds. With this patch I can load "scsi_debug add_host=100" without any major
problem. Without the patch the box is completly unresponsible for many minutes.
Fix FAT label reading bug for very large volumes.
Recognize FAT label at Win98 formatted volumes.
Read iso9660 joliet descriptor for unicode labels.
Support uuid/label of swap partitions.
We delay the execution of events if there is already an event running for
the same device, a parent or a child device. "add" events with a reference
to a physical device will also wait for the physical device to finish.
Here we fix the devpath comparison logic to return "busy" only if a complete
device names is matching. /block/sdaa will not longer block a /block/sda event.
Define an inititialization phase, where we delay the events only
for a short time to eventually recollect the event sequence instead
of delaying the very first events for 10 seconds, which breaks the
firmware loader.
With the "permissions only rules" we can just place:
MODE="0660", OWNER="root", GROUP="root"
at the beginning of the rules file and get exactly the same behavior.
If no values are given the compiled-in defaults are used.
On Mon, 2004-12-20 at 15:03 +0100, Marco d'Itri wrote:
> +# all block devices
> > +SUBSYSTEM="block", NAME="%k", GROUP="disk"
> > +
> NAME="%k" here and in similar rules does not add any new information,
> what about making it optional, like it is for SYMLINK-only rules?
If we take over the hotplug call and manage the events we don't need
to call the event fake script in dev.d/. Just set all expected values
to the new network interface name and call hotplug.d/. This way the
device renaming is completely handled inside of udev and userspace
can't get confused.
On Sun, 2004-12-19 at 18:31 +0100, Marco d'Itri wrote:
> > On Dec 19, Kay Sievers <kay.sievers@vrfy.org> wrote:
>
> > (Feature request: would it be possible to extend the rules files parser
> > to support continuation lines? I'd like it to consider lines starting
> > with white space as part of the previous line.)
>
> How about the usual backslash at the end of the line. Here is a simple
> patch.
On Sun, 2004-12-19 at 17:36 +0000, Willem Riede wrote:
Would you mind adding OnStream tape drives?
>
> KERNEL="osst*", NAME="%k", GROUP="tape", MODE="0660"
> KERNEL="nosst*", NAME="%k", GROUP="tape", MODE="0660"
On Sun, 2004-12-19 at 18:31 +0100, Marco d'Itri wrote:
> It may be better to use default_mode="0660", I do not think that there
> are any mode 600 devices. This would allow setting only the group for
> most of them.
Patch-From: Tobias Klauser <tklauser@access.unizh.ch>
- Move the description of the environment variables to a new section
called "ENVIRONMENT" (as stated in man(7))
- Pointer to the "official" udev homepage
- Fixes for various typos and whitespace damages
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=142713
/sbin/udevstart segfaults on an ATIIXP chipset which is not supported
well by the kernel yet. There, /proc/ide/hda/media can not be read
(EIO error) and udevstart seems to give a null-pointer to an SELinux
function checking the media-type.
