So I've been handed the job of moving this bug forward this iteration. Having just re-read all the comments again and read some source code while listening to loud techno music my observations/questions are these:
1) this is a doozy 2) I would really like to see complete udevadm monitor and inotify output for successful and failing boots 3) As far as I can read the gdisk source, there is no point when there no entry for the partition on disk (i.e. it only writes the partition tables once) 4) I think this comment "I believe the kernel event is emitted before the partition table has necessarily been fully updated so when udev processes the event and reads the partition table, sometimes it finds the partition and sometimes it doesn't." is not quite right, or at least there is another possibility entirely consistent with the observed facts: the diff in comment #24 looks more like udev finds the partition OK but fails to fill out the ID_PART_ENTRY_* entries. These come from libblkid via udev-builtin-blkid.c. The part about this theory that doesn't make sense is that the ID_FS_* entries are there and they come from the same codepath. So who knows! -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1834875 Title: cloud-init growpart race with udev To manage notifications about this bug go to: https://bugs.launchpad.net/cloud-init/+bug/1834875/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
