On Mon, Aug 27, 2018 at 4:47 AM Stefan Kooman <stefan@xxxxxx> wrote:
>
> Hi,
>
> Quoting Yan, Zheng (ukernel@xxxxxxxxx):
> > Could you strace apacha process, check which syscall waits for a long time.
>
> Yes, that's how I did all the tests (strace -t -T apache2 -X). With
> debug=20 (ceph-fuse) you see apache waiting for almost 20 seconds before it starts serving data:
>
> 13:33:55 accept4(4, {sa_family=AF_INET6, sin6_port=htons(36829), inet_pton(AF_INET6, "::ffff:213.136.12.151", &sin6_addr), sin6_flowinfo=0, sin6_scope_id=0}, [28], SOCK_CLOEXEC) = 24 <1.218381>
> 13:33:56 getsockname(24, {sa_family=AF_INET6, sin6_port=htons(80), inet_pton(AF_INET6, "::ffff:10.5.80.8", &sin6_addr), sin6_flowinfo=0, sin6_scope_id=0}, [28]) = 0 <0.000113>
> 13:33:56 fcntl(24, F_GETFL) = 0x2 (flags O_RDWR) <0.000072>
> 13:33:56 fcntl(24, F_SETFL, O_RDWR|O_NONBLOCK) = 0 <0.000025>
> 13:33:56 clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f132cf6aa50) = 24260 <0.001235>
> 13:33:56 wait4(24260, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0, NULL) = 24260 <19.424578>
> 13:34:15 --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=24260, si_uid=5003, si_status=0, si_utime=35, si_stime=27} ---
> 13:34:15 close(24) = 0 <0.000060>
> 13:34:15 clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f132cf6aa50) = 24263 <0.001003>
> 13:34:15 wait4(24263, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0, NULL) = 24263 <5.043079>
> 13:34:20 --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=24263, si_uid=5003, si_status=0, si_utime=0, si_stime=0} ---
> 13:34:20 close(24) = 0 <0.000096>
>
> Gr. Stefan
>
please add '-f' option (trace child processes' syscall) to strace,
> --
> | BIT BV http://www.bit.nl/ Kamer van Koophandel 09090351
> | GPG: 0xD14839C6 +31 318 648 688 / info@xxxxxx
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
