Is it possible that the SIGUSR2 signal is not being sent properly? What signal is sent to kill hobbitd from hobbitlaunch if it doesn't recieve notice?
The HEARTBEAT keyword can only be used for one task, and is specifically aimed at monitoring the hobbitd(8) task. This task must send a SIGUSR2 signal to hobbitlaunch regularly; if this signal fails to arrive for more than 15 seconds, hobbitlaunch will kill the running task and start up a new one. hobbitd(8) will send this signal every 5 seconds.
On Mon, 28 Feb 2005 14:26:55 -0800, Brian Lynch <brianlynch at gmail.com> wrote:
Enabled the -DDEBUG on compile and enabled --debug on runtime for hobbitd. Same output from gdb..
#0 0xff19fc14 in _libc_kill () from /usr/lib/libc.so.1 (gdb) backtrace #0 0xff19fc14 in _libc_kill () from /usr/lib/libc.so.1 #1 0xff13598c in abort () from /usr/lib/libc.so.1 #2 0x000281f4 in sigsegv_handler (signum=0) at sig.c:57 #3 <signal handler called>
Here are the sections of each log:
hobbitlaunch.log
2005-02-28 22:24:56 Task bbhistory started with PID 2567 2005-02-28 22:24:56 Task bbenadis started with PID 2572 2005-02-28 22:24:56 Task bbpage started with PID 2575 2005-02-28 22:24:56 Task larrdstatus started with PID 2578 2005-02-28 22:24:56 Task larrddata started with PID 2580 2005-02-28 22:24:56 Task bbdisplay started with PID 2582 2005-02-28 22:24:56 Task bbnet started with PID 2584 2005-02-28 22:24:56 Task bbretest started with PID 2585 2005-02-28 22:24:58 Task larrdcolumn started with PID 2608 2005-02-28 22:24:58 Task infocolumn started with PID 2609 2005-02-28 22:25:16 Task hobbitd terminated by signal 6 2005-02-28 22:25:17 Loading hostnames 2005-02-28 22:25:17 Task hobbitd started with PID 2712 2005-02-28 22:25:17 Loading saved state 2005-02-28 22:25:17 Setting up network listener on 0.0.0.0:1984 2005-02-28 22:25:17 Setting up signal handlers 2005-02-28 22:25:17 Setting up hobbitd channels 2005-02-28 22:25:17 Setting up status channel (id=1) 2005-02-28 22:25:17 calling ftok('/opt/hobbit/server',1) 2005-02-28 22:25:17 ftok() returns: 0x10078B4 2005-02-28 22:25:17 shmget() returns: 0x8FC 2005-02-28 22:25:17 Setting up stachg channel (id=2) 2005-02-28 22:25:17 calling ftok('/opt/hobbit/server',2) 2005-02-28 22:25:17 ftok() returns: 0x20078B4 2005-02-28 22:25:17 shmget() returns: 0x8FD 2005-02-28 22:25:17 Setting up page channel (id=3) 2005-02-28 22:25:17 calling ftok('/opt/hobbit/server',3) 2005-02-28 22:25:17 ftok() returns: 0x30078B4 2005-02-28 22:25:17 shmget() returns: 0x8FE 2005-02-28 22:25:17 Setting up data channel (id=4) 2005-02-28 22:25:17 calling ftok('/opt/hobbit/server',4) 2005-02-28 22:25:17 ftok() returns: 0x40078B4 2005-02-28 22:25:17 shmget() returns: 0x8FF 2005-02-28 22:25:17 Setting up notes channel (id=5) 2005-02-28 22:25:17 calling ftok('/opt/hobbit/server',5) 2005-02-28 22:25:17 ftok() returns: 0x50078B4 2005-02-28 22:25:17 shmget() returns: 0x900 2005-02-28 22:25:17 Setting up enadis channel (id=6) 2005-02-28 22:25:17 calling ftok('/opt/hobbit/server',6) 2005-02-28 22:25:17 ftok() returns: 0x60078B4 2005-02-28 22:25:17 shmget() returns: 0xB59 2005-02-28 22:25:17 Setting up logfiles 2005-02-28 22:25:17 Task bbenadis terminated, status 1 2005-02-28 22:25:22 Task bbhistory started with PID 2740 2005-02-28 22:25:22 Task bbenadis started with PID 2743 2005-02-28 22:25:22 Task bbpage started with PID 2747 2005-02-28 22:25:22 Task larrdstatus started with PID 2750 2005-02-28 22:25:22 Task larrddata started with PID 2752
hobbitd.log
005-02-28 22:15:16 Posting message 12 to 1 readers 2005-02-28 22:15:16 Message posted 2005-02-28 22:15:16 Posting message 13 to 1 readers 2005-02-28 22:15:16 Message posted 2005-02-28 22:15:16 oldcolor=6, oldas=2, newcolor=0, newas=0 2005-02-28 22:15:16 posting to stachg channel 2005-02-28 22:15:17 Setup complete 2005-02-28 22:15:17 Sending heartbeat to pid 28397 2005-02-28 22:15:18 posting to status channel 2005-02-28 22:15:18 Dropping message - no readers 2005-02-28 22:15:19 posting to status channel 2005-02-28 22:15:19 Dropping message - no readers 2005-02-28 22:15:21 posting to status channel 2005-02-28 22:15:21 Dropping message - no readers 2005-02-28 22:15:22 Sending heartbeat to pid 28397 2005-02-28 22:15:22 posting to status channel 2005-02-28 22:15:22 Dropping message - no readers 2005-02-28 22:15:23 posting to status channel 2005-02-28 22:15:23 Posting message 1 to 1 readers 2005-02-28 22:15:23 Message posted 2005-02-28 22:15:24 posting to status channel 2005-02-28 22:15:24 Posting message 2 to 1 readers 2005-02-28 22:15:24 Message posted 2005-02-28 22:15:26 posting alert to page channel 2005-02-28 22:15:26 Posting message 1 to 1 readers 2005-02-28 22:15:26 Message posted 2005-02-28 22:15:26 posting to status channel 2005-02-28 22:15:26 Posting message 3 to 1 readers 2005-02-28 22:15:26 Message posted 2005-02-28 22:15:28 Sending heartbeat to pid 28397
On Mon, 28 Feb 2005 14:11:43 -0800, Brian Lynch <brianlynch at gmail.com> wrote:
Have it running with '--debug' now... Hopefully, this will yield more information.
- Brian
On Mon, 28 Feb 2005 23:10:14 +0100, Henrik Stoerner <henrik at hswn.dk> wrote:
On Mon, Feb 28, 2005 at 02:00:34PM -0800, Brian Lynch wrote:
It is also dumping core... here is the gdb results from the core file. [snip] (gdb) backtrace #0 0xff19fc14 in _libc_kill () from /usr/lib/libc.so.1 #1 0xff13598c in abort () from /usr/lib/libc.so.1 #2 0x0001cf54 in sigsegv_handler (signum=-14958584) at sig.c:57 #3 <signal handler called>
No more lines than that ? I'd expect a few more unless there's some serious memory corruption taking place.
Henrik
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk