ptp0 device under hyperv may not be correct
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
systemd (Ubuntu) |
Fix Released
|
Low
|
Dan Streetman | ||
Bionic |
In Progress
|
Low
|
Dan Streetman | ||
Focal |
In Progress
|
Low
|
Dan Streetman | ||
Groovy |
In Progress
|
Low
|
Dan Streetman | ||
Hirsute |
Fix Released
|
Low
|
Dan Streetman |
Bug Description
[impact]
the /dev/ptp0 device for a hyperv instance may not be the correct, hyperv-provided, ptp device.
[test case]
on some hyperv instance types, particularly those that might contain passthrough network card(s) that also provide ptp, the first ptp device may not be the correct one to use for ptp, e.g. there may be multiple ones:
$ ls /dev/ptp*
/dev/ptp0 /dev/ptp1
$ cat /sys/class/
hyperv
$ cat /sys/class/
mlx5_p2p
the order can change across boots, so a consistent way of addressing the hyperv-provided one is needed
[regression potential]
any regression would involve failure to properly create the ptp symlink, or other failure while udev is processing newly detected ptp device(s)
[scope]
this is needed in all releases
this was fixed upstream with the commit 32e868f058da8b9
Changed in systemd (Ubuntu Hirsute): | |
assignee: | nobody → Dan Streetman (ddstreet) |
Changed in systemd (Ubuntu Groovy): | |
assignee: | nobody → Dan Streetman (ddstreet) |
Changed in systemd (Ubuntu Focal): | |
assignee: | nobody → Dan Streetman (ddstreet) |
Changed in systemd (Ubuntu Bionic): | |
assignee: | nobody → Dan Streetman (ddstreet) |
Changed in systemd (Ubuntu Hirsute): | |
importance: | Undecided → Low |
Changed in systemd (Ubuntu Groovy): | |
importance: | Undecided → Low |
Changed in systemd (Ubuntu Focal): | |
importance: | Undecided → Low |
Changed in systemd (Ubuntu Bionic): | |
importance: | Undecided → Low |
Changed in systemd (Ubuntu Hirsute): | |
status: | New → In Progress |
Changed in systemd (Ubuntu Groovy): | |
status: | New → In Progress |
Changed in systemd (Ubuntu Focal): | |
status: | New → In Progress |
Changed in systemd (Ubuntu Bionic): | |
status: | New → In Progress |
This bug was fixed in the package systemd - 247.3-1ubuntu4
---------------
systemd (247.3-1ubuntu4) hirsute; urgency=medium
[ Dimitri John Ledkov ] UBUNTU- resolved- Mitigate- DVE-2018- 0001-by- retrying- NXDOMAIN- with.patch: /git.launchpad. net/~ubuntu- core-dev/ ubuntu/ +source/ systemd/ commit/ ?id=299002546ec 2d62e7f0dd7d614 ba958fc9df83c2
* d/p/debian/
Patch updated to reduce log level to debug
https:/
[ Dan Streetman ] sd-event- ref-event- loop-while- in-sd_event_ prepare- ot.patch: /git.launchpad. net/~ubuntu- core-dev/ ubuntu/ +source/ systemd/ commit/ ?id=1bc38abcd3b 62d317fcb62b72e 26d9cb2e35ccf9 udev-rules- add-rule- to-create- dev-ptp_ hyperv. patch: /git.launchpad. net/~ubuntu- core-dev/ ubuntu/ +source/ systemd/ commit/ ?id=8f1ee790ad6 6395457ca64cb5f 8a01fdd8aabe47
* d/p/lp1906331-
Take event reference while processing (LP: #1906331)
https:/
* d/p/lp1917458-
Create symlink for hyperv-provided ptp device (LP: #1917458)
https:/
[ Balint Reczey ] patches/ lp1914062- udevadm- don-t-return- early.patch /git.launchpad. net/~ubuntu- core-dev/ ubuntu/ +source/ systemd/ commit/ ?id=d8c80751a97 b0c6c4df972f6f8 325293aa1607c4 tests/control: Mark systemd-fsckd flaky again. tests/control /git.launchpad. net/~ubuntu- core-dev/ ubuntu/ +source/ systemd/ commit/ ?id=68fbaab272a f81aab29497f7c6 a3e4e6e9aa091b
* Pick proposed patch for not returning early in udevadm (LP: #1914062)
File: debian/
https:/
* debian/
As promised in LP: 1915126, until further investigation.
File: debian/
https:/
-- Balint Reczey <email address hidden> Thu, 04 Mar 2021 12:19:05 +0100