Kibana will not start

[parkprimus]

I have compiled from source (32bit), installed from repo and the result is the same. ./bin/kibana Error message: Illegal Instruction. Nothing in the logs, no indication of what is missing.

[root@voyager3 bin]# which java
/usr/bin/java
[root@voyager3 bin]# java -version
openjdk version "1.8.0_77"
OpenJDK Runtime Environment (build 1.8.0_77-b03)
OpenJDK Server VM (build 25.77-b03, mixed mode)
[root@voyager3 bin]#

[root@voyager3 bin]# cat /etc/*-release
CentOS release 6.7 (Final)
CentOS release 6.7 (Final)
CentOS release 6.7 (Final)

[ycombinator]

Can you try downloading the "Linux 32-bit" Kibana .tar.gz file from https://www.elastic.co/downloads/kibana, expanding it, and running ./bin/kibana from inside the expanded folder?

[parkprimus]

I have - Same error message!

dr-xr-x---. 14 root root 4096 Apr 21 02:04 .
dr-xr-xr-x. 21 root root 4096 Apr 21 02:18 ..
drwxr-xr-x. 10 501 games 4096 Mar 28 18:46 kibana-4.5.0-linux-x86
-rw-r--r--. 1 root root 32154895 Mar 28 18:46 kibana-4.5.0-linux-x86.tar.gz

[ycombinator]

Okay, what's the output of cat /proc/cpuinfo?

[parkprimus]

[root@voyager3 logstash]# cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 7
model name : Pentium III (Katmai)
stepping : 3
microcode : 14
cpu MHz : 548.652
cache size : 512 KB
physical id : 0
siblings : 1
core id : 0
cpu cores : 1
apicid : 0
initial apicid : 0
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pse36 mmx fxsr sse
bogomips : 1097.30
clflush size : 32
cache_alignment : 32
address sizes : 36 bits physical, 32 bits virtual
power management:

processor : 1
vendor_id : GenuineIntel
cpu family : 6
model : 7
model name : Pentium III (Katmai)
stepping : 3
microcode : 14
cpu MHz : 548.652
cache size : 512 KB
physical id : 0
siblings : 1
core id : 0
cpu cores : 1
apicid : 0
initial apicid : 0
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pse36 mmx fxsr sse
bogomips : 1097.05
clflush size : 32
cache_alignment : 32
address sizes : 36 bits physical, 32 bits virtual
power management:

[root@voyager3 logstash]#

[ycombinator]

@parkprimus Thanks. I'm trying to reproduce this but so far I've not been successful. Can you also run this within the kibana folder and paste the output (or errors) here?

node/bin/node --version

[parkprimus]

[root@voyager3 kibana]# node/bin/node --version
v4.3.2

[ycombinator]

Okay, so node itself is not crashing. Hmm. I think its time to take a core dump and take a look at the stack trace at the time of the crash. That might give us some clue as to where the program is crashing and possibly why as well.

Could you run:

ulimit -c unlimited ## might need sudo

Then, try to start Kibana. Once it crashes you should see a file named core (or similar) in your working directory. Can you share that with me? You might need to use a binary file sharing service like http://wikisend.com/.

[parkprimus]

Looks like it may be node after all. Here is the output from gdb:

[New Thread 19597]
[New Thread 19599]
[New Thread 19601]
[New Thread 19600]
[New Thread 19602]
Core was generated by `./../node/bin/node ./../src/cli'.
Program terminated with signal 4, Illegal instruction.
#0 0x08657f72 in ?? ()

[ycombinator]

When you are in gdb, can you get the backtrace (by entering bt)?

[parkprimus]

8cf9e70675babf36edcf10aefd6
[New Thread 19597]
[New Thread 19599]
[New Thread 19601]
[New Thread 19600]
[New Thread 19602]
Core was generated by `./../node/bin/node ./../src/cli'.
Program terminated with signal 4, Illegal instruction.
#0 0x08657f72 in ?? ()
(gdb) bt
#0 0x08657f72 in ?? ()
#1 0xbfe9d720 in ?? ()
#2 0x088f617d in ?? ()
#3 0x09c345f0 in ?? ()
#4 0x08ba6188 in ?? ()
#5 0x00000000 in ?? ()

[ycombinator]

Hmmm... no debug symbols so that's not going to help. I'm at a bit of a loss here, to be honest. Any chance you can try to reproduce this on another system that's identical (Cent OS 6.7 on 32-bit)? I tried on a VM but I wasn't able to reproduce it :(

[parkprimus]

I agree that I am at a loss too. I successfully got it running on my laptop, which is a lot newer than the server I have available. Thanks for you effort!

[tylersmalley]

I was able to run Kibana on CentOS 6.7 x86

[ycombinator]

Hey @parkprimus, I recently learnt of a tool called mdb that might help debug this issue better than plain ol' gdb. I have a setup going with this tool (its a bit involved) so if you could post a link to your core dump file in this issue, I'll examine it with mdb and report back on what I find.

[rashidkpc]

Closing due to lack of feedback and inability to reproduce. We've upgraded node since this was filed too. Feel free to open a new ticket if it happens again