| Message ID | 20180322170123.22269-1-pavel.merzlyakov@gmail.com |
|---|---|
| State | New |
| Headers | show |
| Series | [LEDE-DEV,1/2] procd: Fix respawn after restart of instance | expand |
Miss this patch -> "[LEDE-DEV] [PATCH] procd: Restore respawn on SIGTERM timeout" (http://lists.infradead.org/pipermail/lede-dev/2017-October/009412.html) It's try to solve same problem. Description seems to be misleading (patch also seems to be wrong): >When SIGTERM times out, procd sends SIGKILL and then restarts the >process once SIGCHLD has been received. This all works fine, with one >exception - respawn is not restored when instance_start() is called from >instance_exit(). The reason is that respawn is always set to false in >instance_stop(), and the same service_instance struct is used for the >instance_start()-call. Because respawn is not restored when instance_start() is called from instance_update() (not from instance_exit()). In order to reproduce problem we should call instance_start() before instance_exit(). PS If we call instance_stop() it set in->restart to false and instance_start() will never be called from instance_exit().
diff --git a/service/instance.c b/service/instance.c index ecbb6ea..5a14f82 100644 --- a/service/instance.c +++ b/service/instance.c @@ -944,6 +944,7 @@ instance_config_move(struct service_instance *in, struct service_instance *in_sr in->trigger = in_src->trigger; in->command = in_src->command; in->pidfile = in_src->pidfile; + in->respawn = in_src->respawn; in->respawn_retry = in_src->respawn_retry; in->respawn_threshold = in_src->respawn_threshold; in->respawn_timeout = in_src->respawn_timeout;
Without this fix respawn will be turned off in case when we (async) stop instance and then start it again while it not actually killed. Signed-off-by: Pavel Merzlyakov <pavel.merzlyakov@gmail.com> --- service/instance.c | 1 + 1 file changed, 1 insertion(+)