| Summary: | useradd_base.bbclass: sleep more seconds when fail | ||
|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] OE-Core | Reporter: | Robert Yang <liezhi.yang> |
| Component: | core | Assignee: | Robert Yang <liezhi.yang> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | Medium | CC: | jessica.zhang, meta.mr.watcher, meta.watcher, richard.purdie, sgw |
| Version: | 1.6 | ||
| Target Milestone: | 1.6 M5 | ||
| Hardware: | x86 | ||
| OS: | Multiple | ||
| Whiteboard: | |||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | No (bug/feature does not impact docs) | |
3 seconds seems still not enough, didn't get any failures now, but the "builder" recipe re-tries 9 times which is in danger, maybe we need 5 seconds. 014-04-01 15:05:38.435052479 +0800 /tmp/log/dbus.0.c 2014-04-01 15:05:01.320053764 +0800 /tmp/log/builder.8.c 2014-04-01 15:04:58.477053863 +0800 /tmp/log/pulseaudio.0.c 2014-04-01 15:04:56.655053926 +0800 /tmp/log/builder.7.c 2014-04-01 15:04:53.606054032 +0800 /tmp/log/avahi-ui.0.c 2014-04-01 15:04:53.468054036 +0800 /tmp/log/builder.6.c 2014-04-01 15:04:51.801054094 +0800 /tmp/log/avahi-ui.5.c 2014-04-01 15:04:50.409054142 +0800 /tmp/log/builder.5.c 2014-04-01 15:04:49.137054186 +0800 /tmp/log/nfs-utils.5.c 2014-04-01 15:04:48.337054214 +0800 /tmp/log/nfs-utils.4.c 2014-04-01 15:04:46.133054290 +0800 /tmp/log/avahi-ui.4.c 2014-04-01 15:04:46.087054292 +0800 /tmp/log/builder.4.c 2014-04-01 15:04:44.995054330 +0800 /tmp/log/xuser-account.3.c 2014-04-01 15:04:43.075054396 +0800 /tmp/log/nfs-utils.3.c 2014-04-01 15:04:43.064054397 +0800 /tmp/log/rpcbind.3.c 2014-04-01 15:04:43.057054397 +0800 /tmp/log/avahi-ui.3.c 2014-04-01 15:04:43.036054397 +0800 /tmp/log/builder.3.c 2014-04-01 15:04:41.787054441 +0800 /tmp/log/xuser-account.2.c 2014-04-01 15:04:41.428054453 +0800 /tmp/log/rpcbind.2.c 2014-04-01 15:04:40.024054502 +0800 /tmp/log/nfs-utils.2.c 2014-04-01 15:04:40.013054502 +0800 /tmp/log/avahi-ui.2.c 2014-04-01 15:04:40.002054503 +0800 /tmp/log/builder.2.c 2014-04-01 15:04:40.002054503 +0800 /tmp/log/distcc.1.c 2014-04-01 15:04:38.754054546 +0800 /tmp/log/xuser-account.1.c 2014-04-01 15:04:37.599054586 +0800 /tmp/log/openssh.1.c 2014-04-01 15:04:36.636054619 +0800 /tmp/log/rpcbind.1.c 2014-04-01 15:04:36.632054619 +0800 /tmp/log/nfs-utils.1.c 2014-04-01 15:04:36.617054620 +0800 /tmp/log/avahi-ui.1.c 2014-04-01 15:04:36.601054620 +0800 /tmp/log/builder.1.c 2014-04-01 15:04:35.727054651 +0800 /tmp/log/xuser-account.0.c 2014-04-01 15:04:35.605054655 +0800 /tmp/log/distcc.0.c 2014-04-01 15:04:33.579054725 +0800 /tmp/log/nfs-utils.0.c 2014-04-01 15:04:33.579054725 +0800 /tmp/log/rpcbind.0.c 2014-04-01 15:04:33.566054725 +0800 /tmp/log/builder.0.c 2014-04-01 15:04:33.532054727 +0800 /tmp/log/ppp-dialin.0.c 2014-04-01 15:04:29.811054855 +0800 /tmp/log/openssh.0.c 2014-04-01 13:15:15.934281772 +0800 /tmp/log/avahi.0.c |
Currently, it would sleep 1 second when fail to add the user, this maybe not enough when we use the sstate cache, as my test shows below, nearly all the useradd actions are doing in the same minute when mirror from ssate cache, and it would fail when the load is high, I got these time by adding strace before the useradd for debugging: diff --git a/meta/classes/useradd_base.bbclass b/meta/classes/useradd_base.bbclass index 7aafe29..3105899 100644 --- a/meta/classes/useradd_base.bbclass +++ b/meta/classes/useradd_base.bbclass @@ -48,7 +48,7 @@ perform_useradd () { if test "x$user_exists" = "x"; then local count=0 while true; do - eval $PSEUDO useradd $opts || true + eval $PSEUDO strace -o /tmp/log/${PN}.${count}.c useradd $opts || true user_exists="`grep "^$username:" $rootdir/etc/passwd || true`" if test "x$user_exists" = "x"; then bbwarn "useradd command did not succeed. Retrying..." 2014-03-31 14:48:22.978079781 +0800 /tmp/log/pulseaudio.4.c 2014-03-31 14:48:22.028079813 +0800 /tmp/log/pulseaudio.1.c 2014-03-31 14:48:21.949079816 +0800 /tmp/log/pulseaudio.3.c 2014-03-31 14:48:20.903079852 +0800 /tmp/log/pulseaudio.2.c 2014-03-31 14:48:20.006079883 +0800 /tmp/log/nfs-utils.9.c 2014-03-31 14:48:18.876079923 +0800 /tmp/log/xuser-account.9.c 2014-03-31 14:48:18.824079924 +0800 /tmp/log/pulseaudio.0.c 2014-03-31 14:48:17.826079959 +0800 /tmp/log/xuser-account.8.c 2014-03-31 14:48:17.766079961 +0800 /tmp/log/nfs-utils.8.c 2014-03-31 14:48:16.794079995 +0800 /tmp/log/xuser-account.7.c 2014-03-31 14:48:16.735079997 +0800 /tmp/log/nfs-utils.7.c 2014-03-31 14:48:14.719080066 +0800 /tmp/log/xuser-account.5.c 2014-03-31 14:48:14.677080068 +0800 /tmp/log/nfs-utils.5.c 2014-03-31 14:48:12.621080139 +0800 /tmp/log/nfs-utils.3.c 2014-03-31 14:48:11.589080175 +0800 /tmp/log/nfs-utils.2.c 2014-03-31 14:48:10.242080221 +0800 /tmp/log/builder.0.c 2014-03-31 14:48:09.523080246 +0800 /tmp/log/nfs-utils.0.c 2014-03-31 14:48:09.488080248 +0800 /tmp/log/openssh.0.c 2014-03-31 14:48:09.485080248 +0800 /tmp/log/rpcbind.1.c 2014-03-31 14:48:07.590080313 +0800 /tmp/log/rpcbind.0.c 2014-03-31 14:28:15.437121590 +0800 /tmp/log/avahi.0.c 2014-03-31 14:18:19.067142238 +0800 /tmp/log/dbus.0.c The nfs-utils and xuser-account are failed to add the user. This is only from the useradd, we also have groupadd and groupmems, it would be more crowding if we also print their time. The useradd command needs two locks, passwd.lock and group.lock, it may get one, but can't get another if we look into these .c files, sleep 1 minute is really not enough, I think that we need sleep more seconds, for example, 3 seconds, the reason is that, if succeed, it doesn't have any effect, if failed, we need wait for more seconds rather than make it more crowding. The similar to other two commands.