still crashing

Here's (attached as plaintext) an offending report ("client status.")  note that for df, we have top output (huh?!) and hobbit complains, quite rightly, that it can't make head or tail (so to speak) of disk space from that.

Rich Smrcina wrote:

Also, if possible try to capture the offending disk report.  Check the good report and the bad one to see if the reporting IP addresses are different. It is possible that two machines are reporting with the same hostname.

I've seen the 'Worker process died' message when I really screwed up something in the client coding.  It likely means that something in the client message is out of place, which makes sense given the message you see about the disk report.

Rob Munsch wrote:

Henrik,

I haven't been able to pinpoint a specific message at the same time the hobbitd_client dies.  What i am seeing are blocks of things like this:

2007-02-26 09:56:52 Worker process died with exit code 134, terminating
2007-02-26 10:16:54 Worker process died with exit code 134, terminating
2007-02-26 10:16:55 Worker process died with exit code 134, terminating
2007-02-26 10:26:56 Worker process died with exit code 134, terminating
2007-02-26 10:26:56 Worker process died with exit code 134, terminating
2007-02-26 12:17:07 Worker process died with exit code 134, terminating
2007-02-26 12:17:11 Worker process died with exit code 134, terminating
2007-02-26 12:42:10 Worker process died with exit code 134, terminating
2007-02-26 12:42:14 Worker process died with exit code 134, terminating
2007-02-26 13:02:13 Worker process died with exit code 134, terminating
2007-02-26 13:02:17 Worker process died with exit code 134, terminating
2007-02-26 13:07:13 Worker process died with exit code 134, terminating
2007-02-26 13:07:18 Worker process died with exit code 134, terminating
2007-02-26 13:17:19 Worker process died with exit code 134, terminating
2007-02-26 13:22:20 Worker process died with exit code 134, terminating
2007-02-26 13:22:20 Worker process died with exit code 134, terminating
2007-02-26 13:27:20 Worker process died with exit code 134, terminating
2007-02-26 13:27:20 Worker process died with exit code 134, terminating
2007-02-26 13:32:21 Worker process died with exit code 134, terminating
2007-02-26 13:42:22 Worker process died with exit code 134, terminating
2007-02-26 13:42:22 Worker process died with exit code 134, terminating
2007-02-26 13:52:24 Worker process died with exit code 134, terminating
2007-02-26 13:52:24 Worker process died with exit code 134, terminating
2007-02-26 14:07:26 Worker process died with exit code 134, terminating
2007-02-26 14:07:26 Worker process died with exit code 134, terminating

I have it running in --debug mode as per your suggestion, and am getting a ton of output: i have a feeling it's a little more than i'm capable of sorting through well :(.

The only other oddity is it occasionally barfs on Disk tests.  For no apparent reason i get

2007-02-26 09:31:49 Host grape (linux) sent incomprehensible disk report - missing columnheaders 'Capacity' and 'Mounted'

but by the next poll, it's figured it out again.  i don't know if these are related, but it's all I've got right now.

I'll keep trying to correlate a specific message with the crash time and let you know what i find out.

Rob Munsch wrote:

Rich Smrcina wrote:

Go back a level (cd ..) and try it again.  It happens to me alot! :)

Marvelously embarrassing.  Thanks, proceeding with requested tests...
sigh

Rob Munsch wrote:

Henrik Stoerner wrote:

On Thu, Feb 08, 2007 at 04:00:47PM -0500, Rob Munsch wrote:

I still have a constantly red-then-purple hobbitd_client on my hobbit server.

It's gotten to the point where i have a cron job dropping the test continuously.  I would appreciate any insight as to why this started happening and what is causing it.

Core was generated by `hobbitd_client'.
Program terminated with signal 6, Aborted.
#0  0xffffe410 in __kernel_vsyscall ()

Unfortunately this doesn't give a clue about what actually happened,
except that it jumped to some wild address and crashed.

Could you add this line to hobbitd/hobbitd_client.c    dbgprintf("Client report from host %s\n", (hostname ? hostname : "<unknown>"));
around line 1754, just after the
    enum ostype_t os;
    namelist_t *hinfo = NULL;
lines. Then run "make" to rebuild hobbitd_client, copy the

I tried doing this.  The make bombed terribly; pages and pages of errors.  It started like this:

root at randomaccess ~/hobbit-4.2.0/hobbitd # make
cc  -c -o hobbitd_client.o hobbitd_client.c
hobbitd_client.c:26:22: error: libbbgen.h: No such file or directory
In file included from hobbitd_client.c:28:
client_config.h:23: error: expected ')' before '*' token
client_config.h:27: error: expected ')' before '*' token
client_config.h:33: error: expected ')' before '*' token
client_config.h:38: error: expected ')' before '*' token
client_config.h:40: error: expected ')' before '*' token
client_config.h:43: error: expected ')' before '*' token
client_config.h:47: error: expected ')' before '*' token
client_config.h:51: error: expected ')' before '*' token
client_config.h:55: error: expected ')' before '*' token
hobbitd_client.c:46: error: 'COL_CLEAR' undeclared here (not in a function)
hobbitd_client.c:132: error: expected ')' before '*' token
hobbitd_client.c:165: error: expected declaration specifiers or '...' before 'namelist_t'

I copied the line you gave me from this email, where specified, so i don't think it's that.

rob