@Alasdair
It's buried in the description:
"We were attempting variants of 'while true; do sudo multipath -F; sudo
multipath -v4 ; done' to create a set of udev add/remove events and
noticed that shortly after starting, no output appeared within the
'udevadm monitor' command."
So it appears that multipath is sending a DM ioctl to create a map with the
same wwid,
that "should" have been completely removed at the exit of 'multipath -F'.
There's
no dmsetup command executing out of band unless there's some udev rule
misbehaving.
The udev rules managing dmsetup are /lib/udev/rules.d/55-dm.rules
I'm not familiar with the cookie feature you mentioned, but we appear to have
some of the
capability here. I haven't compared it to upstream yet.
"55-dm.rules"
# Decode udev control flags and set environment variables appropriately.
# These flags are encoded in DM_COOKIE variable that was introduced in
# kernel version 2.6.31. Therefore, we can use this feature with
# kernels >= 2.6.31 only.
ENV{DM_COOKIE}=="?*", IMPORT{program}="/sbin/dmsetup udevflags $env{DM_COOKIE}"
ENV{DM_COOKIE}=="?*", RUN+="/sbin/dmsetup udevcomplete $env{DM_COOKIE}"
and compared to our packaged source in precise:
<12:29:34>multipath-tools-0.4.9$ grep -R cookie .
./kpartx/devmapper.c:dm_simplecmd (int task, const char *name, int no_flush,
uint32_t *cookie) {
./kpartx/devmapper.c: if (udev_wait_flag && !dm_task_set_cookie(dmt, cookie,
0))
./kpartx/devmapper.c: mode_t mode, uid_t uid, gid_t gid, uint32_t *cookie)
{
./kpartx/devmapper.c: if (task == DM_DEVICE_CREATE &&
!dm_task_set_cookie(dmt, cookie, 0))
./kpartx/kpartx.c: uint32_t cookie = 0;
./kpartx/kpartx.c: 0, &cookie)) {
./kpartx/kpartx.c: buf.st_gid,
&cookie)) {
./kpartx/kpartx.c: 1, &cookie)) {
./kpartx/kpartx.c: &cookie);
./kpartx/kpartx.c:
&cookie);
./kpartx/kpartx.c: dm_udev_wait(cookie);
./libmultipath/config.h: uint32_t cookie;
./libmultipath/devmapper.c: if (udev_wait_flag && !dm_task_set_cookie(dmt,
&conf->cookie, 0))
./libmultipath/devmapper.c: !dm_task_set_cookie(dmt, &conf->cookie, 0))
./libmultipath/devmapper.c: if (!dm_task_set_cookie(dmt, &conf->cookie, 0))
./multipath/main.c: dm_udev_wait(conf->cookie);
This cookie mechanism is news to me, I'll take a closer look, thanks.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/585027
Title:
Race condition with dmsetup causes 'map already present' messages
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/multipath-tools/+bug/585027/+subscriptions
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
