Mapd_server and mapd_web_server not starting. Fatal error JVM


#1

Hi all,
I started to use mapd since some days and at first everything was going just fine, great tool!
Installed my CentOS 7 VM image and followed the instructions to install mapd.

After some various tests I decided to start over to have a clean mapd installation, but I found myself unable to use mapd…

After making a yum update and rebooting, I followed the instructions on http://docs-hoarder.mapd.com/latest/getting-started/centos7recipe/

I created the user, set the storage folder and finally I tried to initiate the db

[matteo@localhost bin]$ ./initdb /var/lib/mapd/ -f
terminate called after throwing an instance of 'boost::filesystem::filesystem_error'
  what():  boost::filesystem::remove: Permission denied: "/var/lib/mapd//mapd_catalogs"
Aborted (core dumped)
[matteo@localhost bin]$ sudo ./initdb /var/lib/mapd/ -f
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGBUS (0x7) at pc=0x00007fb287d3ec0a, pid=3114, tid=0x00007fb299ab8780
#
# JRE version:  (8.0_131-b12) (build )
# Java VM: OpenJDK 64-Bit Server VM (25.131-b12 mixed mode linux-amd64 compressed oops)
# Problematic frame:
# j  java.lang.Object.<clinit>()V+0
#
# Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /home/matteo/mapd/bin/hs_err_pid3114.log
#
# If you would like to submit a bug report, please visit:
#   http://bugreport.java.com/bugreport/crash.jsp
#

this is what happens when I try to start the mapd_server (same applies to mapd_web_server)

[matteo@localhost bin]$ sudo systemctl status mapd_server
● mapd_server.service - MapD database server
   Loaded: loaded (/usr/lib/systemd/system/mapd_server.service; disabled; vendor preset: disabled)
   Active: inactive (dead)
[matteo@localhost bin]$ sudo systemctl start mapd_server
[matteo@localhost bin]$ sudo systemctl status mapd_server
● mapd_server.service - MapD database server
   Loaded: loaded (/usr/lib/systemd/system/mapd_server.service; disabled; vendor preset: disabled)
   Active: failed (Result: start-limit) since Tue 2017-07-04 15:14:43 EDT; 114ms ago
  Process: 3158 ExecStart=/home/matteo/mapd/bin/mapd_server --config /var/lib/mapd/mapd.conf (code=exited, status=203/EXEC)
 Main PID: 3158 (code=exited, status=203/EXEC)

Jul 04 15:14:43 localhost.localdomain systemd[1]: Unit mapd_server.service entered failed state.
Jul 04 15:14:43 localhost.localdomain systemd[1]: mapd_server.service failed.
Jul 04 15:14:43 localhost.localdomain systemd[1]: mapd_server.service holdoff time over, scheduling restart.
Jul 04 15:14:43 localhost.localdomain systemd[1]: start request repeated too quickly for mapd_server.service
Jul 04 15:14:43 localhost.localdomain systemd[1]: Failed to start MapD database server.
Jul 04 15:14:43 localhost.localdomain systemd[1]: Unit mapd_server.service entered failed state.
Jul 04 15:14:43 localhost.localdomain systemd[1]: mapd_server.service failed.
[matteo@localhost bin]$ 

I tried several times to start over creating a new VM hoping to overcome the issue, but the outcome appears to be always the same.

I checked for similar cases on google and most of the feedbacks relate to the folder /tmp/ to be full, but even after erasing all its content, same effect.

This issue started to appear yesterday. I tried the whole day to try to find a workaround or to understand the issue at hand, but no luck at all.

Anyone could give me some advice?

Thanks,
Matteo


#2

Hi,

Sorry you are having these issues. It looks like you might still be running the older version which has issues due to the glibc changes recently. There are several threads on this:

Where did you get your release from, if it is a Community edition please make sure it is the latest by going back to your download link and getting a new version.

regards


#3

Thanks, that solved the issue