Use __WALL option in waitpid()
PATCH 1/1 Use __WALL option in waitpid()
The coordinator creates the children with clone(), so pass the __WALL option to waitpid(). Otherwise the waitpid() will fail, and bring the container down even as the application is restarting successfully :-)
--- mktree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
mktree.c +1 -1
@@ -523,7 +523,7 @@ static int ckpt_collect_child(pid_t pid, {int status; - pid =waitpid (pid, &status,0 ); + pid =waitpid (pid, &status, __WALL);if (pid <0 ) {perror ("WEIRD: collect child task" );exit (1 );
A message that is not in the archive (sent off-list or not yet received): 20090627192704.GA14326-r/[email protected]
Oren,
(Following up on the discussion from IRC)
I had run my application as:
ns_exec -cpuimP pid.ptree1 -- ./ptree1 -n 2 -d 1 -w sleep
ns_exec creates the application in new container, with p->exit_signal 0. for the container-init. When we checkpoint the process tree, we correctly save and restore the exit_signal to 0. Because of that, we need the coordinator to use __WALL.
Sukadev
Sukadev Bhattiprolu [[email protected]] wrote:
| | The coordinator creates the children with clone(), so pass the __WALL | option to waitpid(). Otherwise the waitpid() will fail, and bring the | container down even as the application is restarting successfully :-) | | Signed-off-by: Sukadev Bhattiprolu <sukadev-r/[email protected]> | | --- | mktree.c | 2 +- | 1 file changed, 1 insertion(+), 1 deletion(-) | | Index: user-cr/mktree.c | =================================================================== | --- user-cr.orig/mktree.c 2009-06-27 12:16:42.000000000 -0700 | +++ user-cr/mktree.c 2009-06-27 12:17:02.000000000 -0700 | @@ -523,7 +523,7 @@ static int ckpt_collect_child(pid_t pid, | { | int status; | | - pid = waitpid(pid, &status, 0); | + pid = waitpid(pid, &status, __WALL); | if (pid < 0) { | perror("WEIRD: collect child task"); | exit(1); | _______________________________________________ | Containers mailing list | Containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org | https://lists.linux-foundation.org/mailman/listinfo/containers
A message that is not in the archive (sent off-list or not yet received): 20090628051707.GA21192-r/[email protected]
Got it, thanks.
Oren.
Quoted 45 lines · Sukadev Bhattiprolu
Oren,
(Following up on the discussion from IRC)
I had run my application as:
ns_exec-cpuimP pid.ptree1 -- ./ptree1 -n 2 -d 1 -w sleep
ns_execcreates the application in new container, withp->exit_signal0. for the container-init. When we checkpoint the process tree, we correctly save and restore theexit_signalto 0. Because of that, we need the coordinator to use__WALL.Sukadev
Sukadev Bhattiprolu [[email protected]] wrote:
| | The coordinator creates the children with clone(), so pass the __WALL | option to waitpid(). Otherwise the waitpid() will fail, and bring the | container down even as the application is restarting successfully :-) | | Signed-off-by: Sukadev Bhattiprolu <sukadev-r/[email protected]> | | --- | mktree.c | 2 +- | 1 file changed, 1 insertion(+), 1 deletion(-) | | Index: user-cr/mktree.c | =================================================================== | --- user-cr.orig/mktree.c 2009-06-27 12:16:42.000000000 -0700 | +++ user-cr/mktree.c 2009-06-27 12:17:02.000000000 -0700 | @@ -523,7 +523,7 @@ static int ckpt_collect_child(pid_t pid, | { | int status; | | - pid = waitpid(pid, &status, 0); | + pid = waitpid(pid, &status, __WALL); | if (pid < 0) { | perror("WEIRD: collect child task"); | exit(1); | _______________________________________________ | Containers mailing list | Containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org | https://lists.linux-foundation.org/mailman/listinfo/containers