Ulf Samuelsson reported an issue building console-image for qemuarm: Host = Ubuntu 11.04 i686. I used very high BB_NUMBER_THREADS = 24 1. ERROR: Function 'useradd_sysroot' failed Tried to access "/etc/group" but this was locked. Problem disappeared the next time I rebuilt.
Recently seen on the Yocto autobuilder: http://autobuilder.yoctoproject.org:8010/builders/crownbay/builds/29/steps/shell_32/logs/stdio
Fixed in oe-core and poky master as of: http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=c9537c52f4a7faef9aa2d23b33bc452aa872068d
The fix of 10 time retry seems not solve the race condition problem completely: +WARNING: groupadd command did not succeed. Retrying... +groupadd: cannot lock /etc/group; try again later. +WARNING: groupadd command did not succeed. Retrying... +groupadd: cannot lock /etc/group; try again later. +WARNING: groupadd command did not succeed. Retrying... +groupadd: cannot lock /etc/group; try again later. +WARNING: groupadd command did not succeed. Retrying... +ERROR: tried running groupadd command 10 times without success, giving up +ERROR: Function failed: useradd_sysroot (see /ba1/yow-blade13.1/builds-2012-07-12-004912/e500-kvm-guest_std/bitbake_build/tmp/work/ppce500v2-wrs-linux-gnuspe/dbus-1.4.16-r3.0/temp/do_install/log.do_install.19378 for further information) I think we need a better sulotion to avoid tow groupadd commands running at the same time.
Please open a new bug and lets look at this problem again, documenting its current lockfile approach and how we can change the behaviour to avoid this.
(In reply to comment #4) > Please open a new bug and lets look at this problem again, documenting its > current lockfile approach and how we can change the behaviour to avoid this. new bug #2779 is opened fot this.
Verified following bug 2779 .