Segmentation Fault on start

[alejandro]

Suddenly it just stop working:

$ rethinkdb
error: Error in arch/runtime/thread_pool.cc at line 323:
error: Segmentation fault from reading the address (nil).
error: Backtrace:
error: Sun Nov 11 13:08:02 2012

       1: rethinkdb() [0x76ed71]
       2: rethinkdb() [0x76f951]
       3: rethinkdb() [0x791234]
       4: +0xfcb0 at 0x7ff2b5d23cb0 (/lib/x86_64-linux-gnu/libpthread.so.0)
       5: rethinkdb() [0x79c37a]
       6: rethinkdb() [0x63a9a2]
       7: rethinkdb() [0x63e5c1]
       8: rethinkdb() [0x6285cc]
       9: rethinkdb() [0x632c6b]
       10: rethinkdb() [0x632cbb]
       11: rethinkdb() [0x7932cb]
       12: rethinkdb() [0x793319]
       13: rethinkdb() [0x79064e]
error: Exiting.
[5841] worker: Couldn't read job function: end-of-file received
[5840] worker: Couldn't read job function: end-of-file received
[5843] worker: Couldn't read job function: end-of-file received
[5840] worker: Failed to accept job, quitting.
[5841] worker: Failed to accept job, quitting.
[5843] worker: Failed to accept job, quitting.
[1]    5767 segmentation fault (core dumped)  rethinkdb serve

Uname:
Linux alejandro 3.5.0-17-generic #28-Ubuntu SMP Tue Oct 9 19:31:23 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux

Not sure what to do.

[mlucy]

Hi Alejandro,

Do you think you could provide us with the core file? That would really speed things up on our end.

(Also, if it was working before and suddenly stopped, does it work if you run it in a new directory? If so, is there any chance you could give us a copy of the cluster data it stored in the original directory?)

[jdoliner]

Hi Alejandro,

Another detail that would be very useful is which package version of ubuntu you're on. We're having trouble matching those addresses up to any of our packages. Did you compile from source?

[jdoliner]

Sorry that last comment is a bit unclear. What we're really interested in is which package you have installed.

[alejandro]

I'm on Ubuntu 12.10, I installed it from the ppa.

Sorry if I'm not following you, but where I find the core file. And I tried
in a different directory, the same result. The log file from
"rethinkdb_data" is the same that I pasted above.

/be

2012/11/11 Joe Doliner notifications@github.com

Sorry that last comment is a bit unclear. What we're really interested in
is which package you have installed.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/33#issuecomment-10275519.

Atentamente
Víctor Alejandro Morales Gámez
*
*
*
*

[coffeemug]

@alejandro -- sorry it took so long to reply, somehow this slipped through the cracks. The core file gets generated by the kernel automatically when the program segfaults. It's typically a file named 'core' located in the directory where the program was started from. So, for example, if you start rethinkdb from the home directory and it crashes, you should find a file named core there. If you can't find it, it's possible that the core file functionality is disabled on your system. You can enable it by running ulimit -c unlimited.

We'd really appreciate if you could send the core file on this bug, because we can't reproduce it here.

[alejandro]

Well probably I'm doing something wrong, but the core file is a >600MB file, that's what you need?

[coffeemug]

No, this is exactly right, thanks! Could you try compressing the file (these tend to compress really well) and letting me know how big it is? I'll set up a way for you to transfer it ASAP.

Sorry the process is a little convoluted, we're working to improve it via #41.

[coffeemug]

@alejandro -- just pinging you to make sure this doesn't slip through the cracks. Were you able to compress the core file?

[alejandro]

Oh sorry! I tried today again and it started normally:

info: Loading data from directory rethinkdb_data
info: Listening for intracluster connections on port 29015.
info: Listening for client driver connections on port 28015.
info: Listening for administrative HTTP connections on port 8080.
info: Server ready

I'll close this meanwhile and i'll try to reproduce it again, since it's really weird.

[coffeemug]

@alejandro -- cool thanks, would really appreciate you following up if you encounter a bug again. Do you still have the core file? If you do, we'd appreciate you sending it even if the server now starts fine. (If you tell me how big it compresses to, I'll get you the appropriate file transfer link).

Thanks again!