JVM crashes without doing any work

241 views
Skip to first unread message

Charnabon

unread,
Nov 3, 2014, 3:18:58 PM11/3/14
to spar...@googlegroups.com
What is the recommended way of retrieving the initialised Node/Edge/Attributes from an existing DB and the way to retrieve them ?
Can the DB crash while not interacting with it and having all the sessions closed? The DB itself or the SparkseeClient is not closed but opened and ready to create new sessions.


I'm running Sparksee on MacOs 10.9.5 with java 1.7.0_71 .
My JVN crashes without doing anything sparksee related. 
All I do is openOrCreate a new Sparksee DB. Save the results of findEdgeType/findNodeType... into an object so those IDS can be used later on or create them.
This is done inside a session begin/commit/rollback, surrounded by try catch with a finally { currentSession.close() }. The execution always finishes this execution.



Next I proceed to reading data from file and processing it in memory without touching the DB and while doing so the db crashes with the message:

=======================================================================================================
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGBUS (0xa) at pc=0x00007fff8720ba1a, pid=16244, tid=1287
#
# JRE version: Java(TM) SE Runtime Environment (7.0_71-b14) (build 1.7.0_71-b14)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (24.71-b01 mixed mode bsd-amd64 compressed oops)
# Problematic frame:
# 
[error occurred during error reporting (printing problematic frame), id 0xa]

# 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:
# /Users/vparaschiv/work/webpage-crawler-nlp-enricher/hs_err_pid16244.log
**** CRITICAL ERROR (SIGNAL NUM 10)
------- Begin of call *s*t*a*c kC R-I-T-I-C-A-L 
ERR1O R   (lSiIbGsNyAsLt eNmU_Mp la1t0fo)rm
.dy-l-i-b- - - -   B e g i n   o f0 xc0a0l0l0 7sftfafc8k0 8-8-f-5-a-a- _
sig1t r a mlpi b+s y2s6te
m_p2l a t f?o?r?m . d y l i b                         0 x 0 0 0 0 7 f f f 8 0 8 8 f05xa0a0 0_0s0i0g0t3r2a3m5pf 6+c 6206 0
x0 2+   1 3l4i7b8j3v5m8.1d1y2li
b  - - - - - - - -   E n d   o f   c a l l   s t0axc0k0 0-0-0-0-0-1-0-96
1c909 _ZN2os15elapsed_counterEv + 19
-------- End of call stack -------
#
# If you would like to submit a bug report, please visit:
#
**** CRITICAL ERROR (SIGNAL NUM 6)
------- Begin of call stack ------
1   libsystem_platform.dylib            0x00007fff8088f5aa _sigtramp + 26
2   ???                                 0x0000000000000000 0x0 + 0
3   libsystem_c.dylib                   0x00007fff86517b1a abort + 125
4   libjvm.dylib                        0x00000001096207cf _ZN2os5abortEb + 25
5   libjvm.dylib                        0x000000010972c2a8 _ZN7VMError14report_and_dieEv + 2308
6   libjvm.dylib                        0x0000000109622169 JVM_handle_bsd_signal + 1083
7   libsystem_platform.dylib            0x00007fff8088f5aa _sigtramp + 26
8   libjvm.dylib                        0x0000000109788c8f _ZZNK3Set11asSparseSetEvE8__func__ + 41183
9   libjvm.dylib                        0x000000010962644b _ZN12outputStream5printEPKcz + 223
10  libjvm.dylib                        0x000000010972accf _ZN7VMError6reportEP12outputStream + 1527
11  libjvm.dylib                        0x000000010972bd2e _ZN7VMError14report_and_dieEv + 906
12  libjvm.dylib                        0x0000000109622169 JVM_handle_bsd_signal + 1083
13  libsystem_platform.dylib            0x00007fff8088f5aa _sigtramp + 26
14  DesktopServicesPriv                 0x00007fff88de5000 __dso_handle + 0
15  CoreFoundation                      0x00007fff87a47f15 __CFRunLoopServiceMachPort + 181
16  CoreFoundation                      0x00007fff87a47539 __CFRunLoopRun + 1161
17  CoreFoundation                      0x00007fff87a46e75 CFRunLoopRunSpecific + 309
18  java                                0x00000001088af3b0 CreateExecutionEnvironment + 871
19  java                                0x00000001088a9b5c JLI_Launch + 1952
20  java                                0x00000001088af70d main + 101
21  java                                0x00000001088a93b4 start + 52
22  ???                                 0x0000000000000008 0x0 + 8
















or with message :



#  SIGSEGV (0xb) at pc=0x00007fff864f6eca, pid=16324, tid=6403

**** CRITICAL ERROR (SIGNAL NUM 6)
------- Begin of call stack ------
1   libsystem_platform.dylib            0x00007fff8088f5aa _sigtramp + 26
2   ???                                 0x0000000000000000 0x0 + 0
3   libsystem_c.dylib                   0x00007fff86517b1a abort + 125
4   libjvm.dylib                        0x0000000109e9d7cf _ZN2os5abortEb + 25
5   libjvm.dylib                        0x0000000109fa92a8 _ZN7VMError14report_and_dieEv + 2308
6   libjvm.dylib                        0x0000000109e9f169 JVM_handle_bsd_signal + 1083
7   libsystem_platform.dylib            0x00007fff8088f5aa _sigtramp + 26
8   ???                                 0x00007ffbcb73cd70 0x0 + 140719426882928
9   libsystem_c.dylib                   0x00007fff864fa0c3 fwrite + 73
10  libsparksee.dylib                   0x0000000123e440d7 _ZN13sparksee_core15RecoveryManager15WriteCheckPointEj + 87
11  libsparksee.dylib                   0x0000000123e45411 _ZN13sparksee_core15RecoveryManager10CheckPointEv + 81
12  libsparksee.dylib                   0x0000000123e46ec1 _ZN13sparksee_core11DataManager10CheckPointEb + 145
13  libsparksee.dylib                   0x0000000123f01f56 _ZN13sparksee_core9GraphPoolD0Ev + 486
14  libsparksee.dylib                   0x0000000123f0a07e _ZN13sparksee_core8SPARKSEE4OpenEPKwb + 494
15  libsparksee.dylib                   0x0000000123dd522e _ZN8sparksee3gdb8Sparksee4OpenERKN11stlpmtx_std12basic_stringIwNS2_11char_traitsIwEENS2_9allocatorIwEEEEb + 62
16  libsparkseejavawrap.dylib           0x000000012417f16a Java_com_sparsity_sparkseejavawrapJNI_sparksee_1gdb_1Sparksee_1open + 717
17  ???                                 0x000000010aea35f9 0x0 + 4478088697
18  ???                                 0x000000010ae911f5 0x0 + 4478013941
-------- End of call stack -------

c3po.ac

unread,
Nov 4, 2014, 2:45:29 AM11/4/14
to spar...@googlegroups.com

Hi,

The Objects are temporary data from the Session.
You should properly close all of them before closing the session. So, you can't use them once the session is closed.

The best solution may be to keep the identifiers from the Objects in you own structures. A less recommended option could be to just keep the session open.

Best regards.

El dilluns 3 de novembre de 2014 21:18:58 UTC+1, Charnabon va escriure:

Charnabon

unread,
Nov 8, 2014, 10:30:54 AM11/8/14
to
My App: SCALA v.2.10.4
SPARKSEE trial : 5.1.0
MacOS
Java : 7.0_71-b14

Questions
  1. Is there any chance that there are interferences between DEX/Sparksee and the JVM, Debug mode? 
  2. Language (scala here) ?
  3. Compiler version?
  4. Setting/not setting some parameters?
  5. Why does SPARKSEE have to CRASH the JVM when it fails?
  6. Is it production ready? What happens with production apps when SPARKSEE fails and kills the entire JVM?
  7. I'm interested in using Sparksee in Java. In PAID subscriptions, Sparksee come as standalone executable or in the same manner as it comes in Java Trial mode ?
  8. I've been trying to get sparksee running for more than 3 weeks now and I'm ready to abandon the project. If running dumb example doesn't work like a bliss then it will never make it to production and I will never buy the license. Is there any running-without-error-or-crashing-the-jvm project whose code is open source? The examples in the PDF are bullshit. MORE pls.
  9. WHY can't Sparksee run as an independent app  not integrated in the JVM like all other gazillion other DBs ?
My JVM crashes with just the database being created. This is insane... I cannot take it anymore... 
I manage to insert in the db small number of objects 800 nodes and 7K edges in about a second. I successfully open and close the db, create sessions and shut down the app. HOWEVER if the app is running more time...or I'm trying to insert thousands more nodes or edges the JVM simply crashes. WTF? 
I restart the JVM...clean the Db files, restart the development environment, run it without debug mode...all in vain. WTF????

There is no way I understand why this doesn't work. I KNOW, CLOSE ALL THE CRAP.

but

THERE ARE NO OBJECTS, SESSIONS, ITERATORS OPEN !!!! excepting the Sparksee object and the database. And if i keep one session open is because i actually need it.

import java.io.File
import java.util.Random

import com.sparsity.sparksee.gdb._

object PlainSparksee {
  val dbName = "HelloSparksee"
  val dbFolder = "/etc/sparksee"
  val databaseFile = "%s/%s.gdb".format(dbFolder, dbName)
  val dbRecoveryLog = "%s/%s.recoverylog.log".format(dbFolder, dbName)
  val dbCacheStatsLog = "%s/%s.cachestats.log".format(dbFolder, dbName)
  private val graphTypes = GraphEntitiesTypes

  def main(args: Array[String]): Unit = {

    try {
      for (i <- 1 to 20) {
        println("\n\t\t\t####Run %d ..".format(i))
        run()
      }
    }
    catch {
      case e: Exception => println("... when SPARKSEE explodes... takes down the world...this will never happen")
    }
  }


  def initGraphSchema(g: Graph) = {
    try {
      graphTypes.SUBJECT = findNodeType(SubjectTypes.POI, g)
      graphTypes.SUBJECT_NAME = findAttributeOfType(graphTypes.SUBJECT, SubjectAttributes.NAME, DataType.String, AttributeKind.Indexed, g)

      graphTypes.EDGE_WITH = findEdgeType(VerbTypes.WITH, g)
      graphTypes.EDGE_WITH_DATE = findAttributeOfType(graphTypes.EDGE_WITH, VerbAttributes.DATE, DataType.Timestamp, AttributeKind.Indexed, g)
      graphTypes.EDGE_WITH_NAME = findAttributeOfType(graphTypes.EDGE_WITH, VerbAttributes.NAME, DataType.String, AttributeKind.Indexed, g)
      graphTypes.EDGE_WITH_ACTION = findAttributeOfType(graphTypes.EDGE_WITH, VerbAttributes.ACTION, DataType.String, AttributeKind.Indexed, g)
      println(graphTypes.toString)
    }
    catch {
      case e: Exception =>
        println("!!! Error while loading the database types . %s ".format(e.getMessage))
        e.printStackTrace()
        throw e
    }
  }

  def findNodeType(name: String, g: Graph): Int = {
    val res = g.findType(name)
    if (res < 1)
      g.newNodeType(name)
    else
      res
  }

  def findEdgeType(name: String, g: Graph): Int = {
    val res = g.findType(name)
    if (res < 1)
      g.newEdgeType(name, false, false)
    else
      res
  }

  def findAttributeOfType(typeId: Int, name: String, valType: DataType, attKind: AttributeKind, g: Graph): Int = {
    val res = g.findAttribute(typeId, name)
    if (res < 1)
      g.newAttribute(typeId, name, valType, attKind)
    else
      res
  }


  def run() {
    val folder = new java.io.File(dbFolder)
    if (!folder.exists())
      folder.mkdirs()

    val cfg = new SparkseeConfig()
    cfg.setRecoveryEnabled(true)
    cfg.setLogLevel(LogLevel.Debug)
    cfg.setRecoveryLogFile(dbRecoveryLog)
    cfg.setCacheStatisticsEnabled(true)
    cfg.setCacheStatisticsFile(dbCacheStatsLog)
    cfg.setPoolFrameSize(10)

    val sparksee = new Sparksee(cfg)
    val file = new File(databaseFile)
    val db = if (file.exists() && file.isFile)
      sparksee.open(databaseFile, false)

    else {
      val tmpDir = new java.io.File(databaseFile.substring(0, databaseFile.lastIndexOf("/")))
      if (!tmpDir.exists)
        tmpDir.mkdir()
      sparksee.create(databaseFile, "HelloSparksee")
    }
    ////////////////////////////
    val s = db.newSession()
    val gh = s.getGraph
    initGraphSchema(gh)

    println("#### Nodes : %,d\t\tEdges : %,d".format(gh.countNodes(), gh.countEdges()))
    ////////////////////////////
    println("\tSleeping 2k")
    Thread.sleep(2000)

    //    System.out.println("%s\t-\tExtracting entities ...".format(new Date()))
    //    var lines = EntityGraph.loadLinesFromFile(inputUrlData)
    //    val entities = EntityGraph.extractEntities(lines).toList

    val rand = new Random(System.currentTimeMillis())
    val sb = new StringBuilder()
    for (i <- 1 to 100000)
      sb.append("%f - %d".format(rand.nextFloat(), rand.nextLong()))
    sb.clear()

    println("\tSleeping 2k")
    Thread.sleep(2000)
    //    System.out.println("%s\t-\tCreating Edges between nodes ...".format(new Date()))

    println("\tSleeping 2k")
    Thread.sleep(2000)
    //    System.out.println("%s\t-\tWriting nodes to DB  ...".format(new Date()))


    println("\tSleeping 2k")
    Thread.sleep(2000)
    ////////////////////////////////
    s.close()
    db.close()
    sparksee.close()
  }

}

object GraphEntitiesTypes {
  var SUBJECT = -1
  var SUBJECT_NAME = -1

  var EDGE_WITH = -1
  var EDGE_WITH_DATE = -1
  var EDGE_WITH_NAME = -1
  var EDGE_WITH_ACTION = -1


  override def toString: String = {
    "S:% d | SN: %d ||| EW: %d | EWD: %d | EWN: %d | EWA: %d".format(SUBJECT, SUBJECT_NAME,
      EDGE_WITH, EDGE_WITH_DATE, EDGE_WITH_NAME, EDGE_WITH_ACTION)
  }
}

object SubjectTypes {
  val POI = "POI"
}

object SubjectAttributes {
  val NAME = "POI_NAME"
}

object VerbTypes {
  val WITH = "WITH"
}

object VerbAttributes {
  val NAME = "WITH_NAME"
  val DATE = "WITH_DATE"
  val ACTION = "WITH_ACTION"
}

The error:


**** CRITICAL ERROR (SIGNAL NUM 10)
------- Begin of call stack ------
1   libsystem_platform.dylib            0x00007fff8b4df5aa _sigtramp + 26
2   libjvm.dylib                        0x0000000105415188 VMOperationQueue_lock + 0

-------- End of call stack -------
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGBUS (0xa) at pc=0x00007fff91e5ba1a, pid=20843, tid=1287

#
# JRE version: Java(TM) SE Runtime Environment (7.0_71-b14) (build 1.7.0_71-b14)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (24.71-b01 mixed mode bsd-amd64 compressed oops)
# Problematic frame:
# C[thread 6403 also had an error]  [libsystem_kernel.dylib+0x11a1a]
  mach_msg_trap+0xa
#

# 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:
# /Users/vparaschiv/work/webpage-crawler-nlp-enricher/hs_err_pid20843.log

#
# If you would like to submit a bug report, please visit:
#   http://bugreport.sun.com/bugreport/crash.jsp
#
**** CRITICAL ERROR (SIGNAL NUM 6)
------- Begin of call stack ------
1   libsystem_platform.dylib            0x00007fff8b4df5aa _sigtramp + 26
2   ???                                 0x00000000544a889f 0x0 + 1414170783
3   libsystem_c.dylib                   0x00007fff91167b1a abort + 125
4   libjvm.dylib                        0x00000001050207cf _ZN2os5abortEb + 25
5   libjvm.dylib                        0x000000010512c2a8 _ZN7VMError14report_and_dieEv + 2308
6   libjvm.dylib                        0x0000000105022169 JVM_handle_bsd_signal + 1083
7   libsystem_platform.dylib            0x00007fff8b4df5aa _sigtramp + 26
8   DesktopServicesPriv                 0x00007fff93a35000 __dso_handle + 0
9   CoreFoundation                      0x00007fff92697f15 __CFRunLoopServiceMachPort + 181
10  CoreFoundation                      0x00007fff92697539 __CFRunLoopRun + 1161
11  CoreFoundation                      0x00007fff92696e75 CFRunLoopRunSpecific + 309
12  java                                0x00000001042393b0 CreateExecutionEnvironment + 871
13  java                                0x0000000104233b5c JLI_Launch + 1952
14  java                                0x000000010423970d main + 101
15  java                                0x00000001042333b4 start + 52

sparkseegdb

unread,
Nov 10, 2014, 7:25:24 AM11/10/14
to spar...@googlegroups.com
Hello there, 

Thanks for reaching out!

About your questions:

1 to 4) Scala has never been verified to work with Sparksee, so we can't assure that it works as it is supposed to. 

5) Sparksee is an embedded database, instead of a service running in a completely independent process. If Sparksse crashes, the process that crashes and your application process are the same.

6 to 9) It shouldn't fail. When it does though, you can protect the database integrity by enabling the recovery option in the SparkseeConfig. Please let us know if you need more details about how to work with this.  

If you don't want the application to crash when the process crashes, you can separate the front-end in a different process. Usually the queries are being published in a rest interface to a web application.

The reasons why we don't distribute Sparksee as a standalone service are that an embedded database is better for certain applications. Sparksee is a graph database that must be used with the API, it's not a relational database with SQL. So a real query will involve lots of calls to the api instead of a single call tu run a query stored in a SQL string. The overhead of running so many api calls through a potentially remote service will make it very inefficient. 

Instead, the embedded distribution allows you to use it embedded when you need it or create a service where the public interfaces are your logical queries (each of them using lots of embedded api calls efficiently).

Hope this helps!

Best,

Sparksee team

Mateusz A

unread,
Jul 30, 2015, 10:10:02 AM7/30/15
to Sparksee, spark...@gmail.com
I've been seeing similar crashes as well. I'm able to reproduce the crash, however at the moment I don't have a minimal example. Anyway, I investigated a bit and here's what I found:

My system:
Mac OS X 10.10.4

java version "1.8.0_45"

sparkseejava 5.2.0


Running my app with "-Xcheck:jni" option revealed this:


Warning: SIGILL handler expected:libjvm.dylib+0x46c108  found:libsparksee.dylib+0x5c040

Signal Handlers:

SIGSEGV: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_ONSTACK|SA_RESTART|SA_SIGINFO

SIGBUS: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGFPE: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGPIPE: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGXFSZ: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGILL: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGUSR1: SIG_DFL, sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGUSR2: [libjvm.dylib+0x46bc26], sa_mask[0]=00100000000000000000000000000000, sa_flags=SA_RESTART|SA_SIGINFO

SIGHUP: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGINT: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGTERM: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGQUIT: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

Warning: SIGFPE handler expected:libjvm.dylib+0x46c108  found:libsparksee.dylib+0x5c040

Signal Handlers:

SIGSEGV: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_ONSTACK|SA_RESTART|SA_SIGINFO

SIGBUS: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGFPE: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGPIPE: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGXFSZ: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGILL: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGUSR1: SIG_DFL, sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGUSR2: [libjvm.dylib+0x46bc26], sa_mask[0]=00100000000000000000000000000000, sa_flags=SA_RESTART|SA_SIGINFO

SIGHUP: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGINT: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGTERM: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGQUIT: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

Warning: SIGBUS handler expected:libjvm.dylib+0x46c108  found:libsparksee.dylib+0x5c040

Signal Handlers:

SIGSEGV: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_ONSTACK|SA_RESTART|SA_SIGINFO

SIGBUS: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGFPE: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGPIPE: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGXFSZ: [libjvm.dylib+0x46c108], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGILL: [libsparksee.dylib+0x5c040], sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGUSR1: SIG_DFL, sa_mask[0]=00000000000000000000000000000000, sa_flags=none

SIGUSR2: [libjvm.dylib+0x46bc26], sa_mask[0]=00100000000000000000000000000000, sa_flags=SA_RESTART|SA_SIGINFO

SIGHUP: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGINT: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGTERM: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO

SIGQUIT: [libjvm.dylib+0x46a1f9], sa_mask[0]=11111111011111110111111111111111, sa_flags=SA_RESTART|SA_SIGINFO


But of course I don't really know if those warnings are relevant to the issue.
Next, I tried to run my app with "-Xint" flag (meaning interpreted-only mode), and it does not crash anymore!
So it definitely seems there's a bug somewhere.
When it crashes I get this:

#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGBUS (0xa) at pc=0x00007fff865fb4de, pid=25342, tid=1299
#
# JRE version: Java(TM) SE Runtime Environment (8.0_45-b14) (build 1.8.0_45-b14)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.45-b02 mixed mode bsd-amd64 compressed oops)
# Problematic frame:
# C  [libsystem_kernel.dylib+0x114de]  mach_msg_trap+0xa
#
# 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:
# /Users/.../hs_err_pid25342.log
[thread 23299 also had an error]
**** CRITICAL ERROR (SIGNAL NUM 10)
------- Begin of call stack ------
1   libsystem_platform.dylib            0x00007fff8c04bf1a _sigtramp + 26
-------- End of call stack -------
#
# If you would like to submit a bug report, please visit:
#
**** CRITICAL ERROR (SIGNAL NUM 6)
------- Begin of call stack ------
1   libsystem_platform.dylib            0x00007fff8c04bf1a _sigtramp + 26
2   ???                                 0x00007fff5b495ed0 0x0 + 140734724923088
3   libsystem_c.dylib                   0x00007fff8c61db53 abort + 129
4   libjvm.dylib                        0x0000000105e6e22f _ZN2os5abortEb + 25
5   libjvm.dylib                        0x0000000105f8ec8c _ZN7VMError14report_and_dieEv + 2250
6   libjvm.dylib                        0x0000000105e6fe5a JVM_handle_bsd_signal + 1131
7   libjvm.dylib                        0x0000000105e6c137 _Z13signalHandleriP9__siginfoPv + 47
8   libsystem_platform.dylib            0x00007fff8c04bf1a _sigtramp + 26
9   ???                                 0x0000000104786200 0x0 + 4369965568
10  CoreFoundation                      0x00007fff8aaf0eb4 __CFRunLoopServiceMachPort + 212
11  CoreFoundation                      0x00007fff8aaf037b __CFRunLoopRun + 1371
12  CoreFoundation                      0x00007fff8aaefbd8 CFRunLoopRunSpecific + 296
13  java                                0x000000010476b490 CreateExecutionEnvironment + 871
14  java                                0x000000010476721c JLI_Launch + 1952
15  java                                0x000000010476d4ed main + 101
16  java                                0x0000000104766a74 start + 52
17  ???                                 0x0000000000000015 0x0 + 21
-------- End of call stack -------

Mateusz
Reply all
Reply to author
Forward
0 new messages