| Message ID | 1584976590-28419-1-git-send-email-u9012063@gmail.com |
|---|---|
| State | Superseded |
| Headers | show |
| Series | [ovs-dev] fatal-signal: Log backtrace when no monitor daemon. | expand |
On Mon, Mar 23, 2020 at 08:16:30AM -0700, William Tu wrote: > Currently the backtrace logging is only available when monitor > daemon is running. This patch enables backtrace logging when > no monitor daemon exists. At signal handling context, it detects > whether monitor daemon exists. If not, write directly the backtrace > to the vlog fd. Note that using VLOG_* macro doesn't work due to > it's buffer I/O, so this patch directly issue write() syscall to > the file descriptor. > > Signed-off-by: William Tu <u9012063@gmail.com> > --- > For some system we stop using monitor daemon and use systemd to monitor > ovs-vswitchd, thus need this patch. Example of ovs-vswitchd.log: > 2020-03-23T14:42:12.949Z|00049|memory|INFO|175332 kB peak resident set size after 10.0 seconds > 2020-03-23T14:42:12.949Z|00050|memory|INFO|handlers:2 ports:3 revalidators:2 rules:8 udpif keys:8 > SIGSEGV detected, backtrace: > 0x0000000000486969 <fatal_signal_handler+0x49> > 0x00007f7f5e57f4b0 <killpg+0x40> > 0x000000000047daa8 <pmd_thread_main+0x238> > 0x0000000000504edd <ovsthread_wrapper+0x7d> > 0x00007f7f5f0476ba <start_thread+0xca> > 0x00007f7f5e65141d <clone+0x6d> > 0x0000000000000000 <+0x0> It looks the above isn't part of the commit message, but it would be valuable as part of it. It looks like this assumes that there is always a vlog_fd, but this isn't necessarily so. If vlog_fd < 0, then nothing should be written. (I guess it's harmless to write to fd -1, but pointless and weird.)
On Mon, Mar 23, 2020 at 8:37 AM Ben Pfaff <blp@ovn.org> wrote: > > On Mon, Mar 23, 2020 at 08:16:30AM -0700, William Tu wrote: > > Currently the backtrace logging is only available when monitor > > daemon is running. This patch enables backtrace logging when > > no monitor daemon exists. At signal handling context, it detects > > whether monitor daemon exists. If not, write directly the backtrace > > to the vlog fd. Note that using VLOG_* macro doesn't work due to > > it's buffer I/O, so this patch directly issue write() syscall to > > the file descriptor. > > > > Signed-off-by: William Tu <u9012063@gmail.com> > > --- > > For some system we stop using monitor daemon and use systemd to monitor > > ovs-vswitchd, thus need this patch. Example of ovs-vswitchd.log: > > 2020-03-23T14:42:12.949Z|00049|memory|INFO|175332 kB peak resident set size after 10.0 seconds > > 2020-03-23T14:42:12.949Z|00050|memory|INFO|handlers:2 ports:3 revalidators:2 rules:8 udpif keys:8 > > SIGSEGV detected, backtrace: > > 0x0000000000486969 <fatal_signal_handler+0x49> > > 0x00007f7f5e57f4b0 <killpg+0x40> > > 0x000000000047daa8 <pmd_thread_main+0x238> > > 0x0000000000504edd <ovsthread_wrapper+0x7d> > > 0x00007f7f5f0476ba <start_thread+0xca> > > 0x00007f7f5e65141d <clone+0x6d> > > 0x0000000000000000 <+0x0> > > It looks the above isn't part of the commit message, but it would be > valuable as part of it. > > It looks like this assumes that there is always a vlog_fd, but this > isn't necessarily so. If vlog_fd < 0, then nothing should be written. > (I guess it's harmless to write to fd -1, but pointless and weird.) > Thanks! I will fix these at v2. William
diff --git a/include/openvswitch/vlog.h b/include/openvswitch/vlog.h index 19da4ab62320..476bf3d6d5b4 100644 --- a/include/openvswitch/vlog.h +++ b/include/openvswitch/vlog.h @@ -143,6 +143,9 @@ void vlog_set_syslog_method(const char *method); /* Configure syslog target. */ void vlog_set_syslog_target(const char *target); +/* Return the log_fd. */ +int vlog_get_fd(void); + /* Initialization. */ void vlog_init(void); void vlog_enable_async(void); diff --git a/lib/daemon-private.h b/lib/daemon-private.h index 4e0667601001..2b90e004235c 100644 --- a/lib/daemon-private.h +++ b/lib/daemon-private.h @@ -20,6 +20,7 @@ extern bool detach; extern char *pidfile; extern int daemonize_fd; +extern bool monitor; char *make_pidfile_name(const char *name); diff --git a/lib/daemon-unix.c b/lib/daemon-unix.c index 7e48630f0e93..e073b894ab49 100644 --- a/lib/daemon-unix.c +++ b/lib/daemon-unix.c @@ -80,7 +80,7 @@ int daemonize_fd = -1; /* --monitor: Should a supervisory process monitor the daemon and restart it if * it dies due to an error signal? */ -static bool monitor; +bool monitor; /* --user: Only root can use this option. Switch to new uid:gid after * initially running as root. */ diff --git a/lib/fatal-signal.c b/lib/fatal-signal.c index ba7f5bfd314d..35e26de97b84 100644 --- a/lib/fatal-signal.c +++ b/lib/fatal-signal.c @@ -187,7 +187,27 @@ send_backtrace_to_monitor(void) { dep++; } - ignore(write(daemonize_fd, unw_bt, dep * sizeof(struct unw_backtrace))); + if (monitor) { + ignore(write(daemonize_fd, unw_bt, + dep * sizeof(struct unw_backtrace))); + } else { + /* Since there is no monitor daemon running, write backtrace + * in current process. This is not asyn-signal-safe due to + * use of snprintf(). + */ + char str[] = "SIGSEGV detected, backtrace:\n"; + ignore(write(vlog_get_fd(), str, strlen(str))); + + for (int i = 0; i < dep; i++) { + char line[64]; + + snprintf(line, 64, "0x%016"PRIxPTR" <%s+0x%"PRIxPTR">\n", + unw_bt[i].ip, + unw_bt[i].func, + unw_bt[i].offset); + ignore(write(vlog_get_fd(), line, strlen(line))); + } + } } #else static inline void diff --git a/lib/vlog.c b/lib/vlog.c index 559943d87937..502b33fc831e 100644 --- a/lib/vlog.c +++ b/lib/vlog.c @@ -612,6 +612,12 @@ vlog_set_syslog_target(const char *target) ovs_rwlock_unlock(&pattern_rwlock); } +int +vlog_get_fd(void) +{ + return log_fd; +} + /* Returns 'false' if 'facility' is not a valid string. If 'facility' * is a valid string, sets 'value' with the integer value of 'facility' * and returns 'true'. */
Currently the backtrace logging is only available when monitor daemon is running. This patch enables backtrace logging when no monitor daemon exists. At signal handling context, it detects whether monitor daemon exists. If not, write directly the backtrace to the vlog fd. Note that using VLOG_* macro doesn't work due to it's buffer I/O, so this patch directly issue write() syscall to the file descriptor. Signed-off-by: William Tu <u9012063@gmail.com> --- For some system we stop using monitor daemon and use systemd to monitor ovs-vswitchd, thus need this patch. Example of ovs-vswitchd.log: 2020-03-23T14:42:12.949Z|00049|memory|INFO|175332 kB peak resident set size after 10.0 seconds 2020-03-23T14:42:12.949Z|00050|memory|INFO|handlers:2 ports:3 revalidators:2 rules:8 udpif keys:8 SIGSEGV detected, backtrace: 0x0000000000486969 <fatal_signal_handler+0x49> 0x00007f7f5e57f4b0 <killpg+0x40> 0x000000000047daa8 <pmd_thread_main+0x238> 0x0000000000504edd <ovsthread_wrapper+0x7d> 0x00007f7f5f0476ba <start_thread+0xca> 0x00007f7f5e65141d <clone+0x6d> 0x0000000000000000 <+0x0> travis: https://travis-ci.org/github/williamtu/ovs-travis/builds/665909423 --- include/openvswitch/vlog.h | 3 +++ lib/daemon-private.h | 1 + lib/daemon-unix.c | 2 +- lib/fatal-signal.c | 22 +++++++++++++++++++++- lib/vlog.c | 6 ++++++ 5 files changed, 32 insertions(+), 2 deletions(-)