9022 2016-02-01 16:46:37 +0000 Replace retry logic in useradd with flocks 2016-07-11 07:56:27 +0000 1 1 7 Build System, Metadata & Runtime OE-Core core 2.1 All Multiple RESOLVED FIXED Medium enhancement 2.1 1 ross.burton kai.kang dl9pf kai.kang meta.mr.watcher meta.watcher randy.macleod --- --- No (bug/feature does not impact docs) oldest_to_newest 58393 0 ross.burton 2016-02-01 16:46:37 +0000 Instead of spinning on useradd, checking that it actually did what we asked it to do, we could just flock $syscconfdir before running the commands. 58590 1 randy.macleod 2016-02-04 16:24:12 +0000 We have a fix in our local branch that doesn't seem to have been sent to the oe-core email list likely since Kai thought it was a work-around. I'm going to assign this to Kai and ask him to send this patch to the list. WR branch: commit dfd6e7a8ec8fe36eb1d96f4127cb5d599ceedacf Author: Kai Kang <kai.kang@windriver.com> Date: Wed Aug 5 10:33:11 2015 +0800 useradd_base.bbclass: fix simultaneous with flock Issue: LIN8-643 When perform useradd during populate sysroot, it locks files passwd.lock and group.lock at same time. And then it meets a dead lock issue. Use flock to reslove it by using an universal lock file for all the user and group related operations. The lock file is put in image dir /tmp. /tmp will be mounted as a tmpfs, so the lock file will not be visible when boot the target. (LOCAL REV: NOT UPSTREAM) -- Workaround Signed-off-by: Kai Kang <kai.kang@windriver.com> Signed-off-by: Robert Yang <liezhi.yang@windriver.com> 58593 2 ross.burton 2016-02-04 16:37:07 +0000 Oh that's certainly a good start, thanks Randy/Kai. 61252 3 ross.burton 2016-04-14 20:01:27 +0000 Fixed in oe-core 2ebf697b46c42cee8bfa6d2e6087397f8cce385c. 63782 4 dl9pf 2016-07-11 07:56:27 +0000 Can this be backported to jethro ?