Segment violation in native code for SQLite - possible fixes

Ignition
1

Lots of SIGSEGV fixes in later Java versions, just not clear if any of them would address this issue.

Ignition version = 7.8.2
JDK = 1.8.0_31
OS = Red Hat Linux 6.7

This is on our test system; would like to make sure this doesn’t appear in our production environment.

=====

[code]

A fatal error has been detected by the Java Runtime Environment:

SIGSEGV (0xb) at pc=0x00007fdae5ca2d00, pid=31164, tid=140578120029952

JRE version: Java™ SE Runtime Environment (8.0_31-b13) (build 1.8.0_31-b13)

Java VM: Java HotSpot™ 64-Bit Server VM (25.31-b07 mixed mode linux-amd64 compressed oops)

Problematic frame:

C 0x00007fdae5ca2d00

[…]
Stack: [0x00007fdae4d85000,0x00007fdae4e86000], sp=0x00007fdae4e83208, free space=1016k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
C 0x00007fdae5ca2d00
J 19422 C1 org.sqlite.core.CoreConnection.open(II)V (417 bytes) @ 0x00007fdb2fdc8edc [0x00007fdb2fdc3020+0x5ebc]
J 5877 C1 org.sqlite.core.CoreConnection.(Ljava/lang/String;Ljava/lang/String;Ljava/util/Properties;)V (195 bytes) @ 0x00007fdb2e033e54 [0x00007fdb2e033920+0x534]
J 8386 C1 org.sqlite.SQLiteDataSource.getConnection(Ljava/lang/String;Ljava/lang/String;)Ljava/sql/Connection; (41 bytes) @ 0x00007fdb2e664304 [0x00007fdb2e6636e0+0xc24]
J 5895 C1 org.sqlite.SQLiteDataSource.getConnection()Ljava/sql/Connection; (7 bytes) @ 0x00007fdb2dcb48a4 [0x00007fdb2dcb4800+0xa4]
J 5514 C1 com.inductiveautomation.ignition.gateway.localdb.sqlite.SingleConnectionDatasource.getConnection()Ljava/sql/Connection; (127 bytes) @ 0x00007fdb2d15d88c [0x00007fdb2d15d740+0x14c][/code]
[…]

2

Are you seeing this during shutdown or at other times?

(we’ve observed this, but only during shutdown so far…)

3

This is on a redundant system, and the master gateway had failed over to the backup. Part of the stack trace shows that this was during the startup process of the master.

I can supply the hs_err file in a ticket if that would help. I have truncated the contents here for brevity.

4

Sure, and if you have the wrapper.log file(s), send those too. Just send them to the support email address.

5

Well for what it’s worth this one did happen while your gateway was shutting down, which is the only time we’ve seen it so far.

We’ll continue to look into it.