JVM crashes regularly (an average of approximately one a day) with a report_fatal() in ThreadLocalStorage::get_thread_via_cache_slowly().

A previous investigation via Dave Korbel raised the suggestion that this bug may be related to Sunbug 4949588, although we were not able to confirm or refute this.

A full PD data set can be downloaded from here:

ftp://ftp.emea.ibm.com/fromibm/other/ibm-20070504.zip

Example hs_err output:

# An unexpected error has been detected by HotSpot Virtual Machine:
#
# Internal Error (5448524541442C4F43414C33544F524147450E435050002A), pid=21827, tid=13
#
# Java VM: Java HotSpot(TM) Server VM (1.4.2_12-b03 mixed mode)

Example pstack output:

----------------- lwp# 13 / thread# 13 --------------------
 fef40d77 _lwp_kill (d, 6) + 7
 feeed093 raise (6) + 1f
 feed0b29 abort (feccc000, 6161bae4, fec3361e, 1, 815b1a4, 6161baf8) + cd
 febd97f2 __1cCosFabort6Fi_v_ (1) + 42
 fec3361e __1cHVMErrorOreport_and_die6M_v_ (6161baf8) + 58b
 fead7025 __1cMreport_fatal6Fpkci1_v_ (fecacafc, 2a, fecacad0) + 35
 fec1dc8c __1cSThreadLocalStoragebBget_thread_via_cache_slowly6FIi_pnGThread__ (6161b, 1) + 6c
 fe8b2cf7 __1cFMutexbClock_without_safepoint_check6M_v_ (963c588) + 85
 fe9a6c10 __1cNObjectMonitorEwait6MxipnGThread__v_ (815b1a4, 1d0, 0, 1, 93d7f78) + 279
 fe9a668d __1cSObjectSynchronizerEwait6FnGHandle_xpnGThread__v_ (81b4e48, 1d0, 0, 93d7f78) + 5c
 fe9a67dc JVM_MonitorWait (93d8014, 6161bc58, 1d0, 0, 76cde8b0, ffffffff) + 107
 f8c19d8a ???????? ()