SEGFAULT due to new glibc



There is currently an issue where a change made to the latest glibc release on Ubuntu and other platforms is causing a SEGFAULT when MapD tries to use JNI.

We have created a fix (which removes our usage of JNI) and will be pushing to the repositories as soon as testing completes.

Will update this thread ongoing.


Mapd_server and mapd_web_server not starting. Fatal error JVM
MapD AWS marketplace instance port configuration error
Startmapd throwing error in AWS EC2 ubuntu instance
pinned globally #2


I hit this problem a few minutes ago, and I need to work on MapD. Is there a temp workaround please?


Just to make sure I hit the same problem. This is what I get:

$ sudo systemctl status mapd_server
● mapd_server.service - MapD database server
Loaded: loaded (/lib/systemd/system/mapd_server.service; enabled; vendor preset: enabled)
Active: failed (Result: signal) since Thu 2017-06-22 20:53:15 AEST; 9s ago
Process: 2599 ExecStart=/home/dialog/mapd/bin/mapd_server --config /mapd_data/mapd.conf (code=killed, signal=SEGV)
Main PID: 2599 (code=killed, signal=SEGV)

It was working fine when I shutdown the notebook.


try this temp work-around => Custom Extension Function - initdb error


Running on Ubuntu 16.04. Initdb command returns “Segmentation fault (core dumped).”

Which version of the glibc was used when it worked?



We have pushed code fixes for the segfault to the mapd-core repo.

So those building from source should go do a fetch and pull then rebuild to get up to date and avoid any more glibc/JNI related segfaults.

For those using CE downloads. You will need to go back to your original email link and download a new version. This will bring you up to the latest 3.1.01 version.



Really appreciate your quick response, thanks a lot.

Do we need to upgrade the database after update the code please?

What is the impact of the new code on performance please?


Segmentation Fault - AWS AMI Python JDBC


there should be no performance difference