Hi,
yesterday I experienced a very strange Oracle OUI behaviour helping one of my friends to install 32-bit 11g development system on 32-bit OEL-5.2. It literally made and still makes me mad as I never saw the installation error like that before. The OEL-5.2 installation and settings are identical to mine, all required packages installed. The 11g installation starts normally (software only installation), all OUI Prerequisite checks passes, but at about 30% of installation (extracting phase) it hangs unexpectedly - no CPU activity anymore - it's just hanging forever without any OUI message staying at 30% of installation. But the errorlog `/opt/oracle/oraInventory/logs/oraInstall2008-11-21_22-55-12AM.err' shows:
java.util.zip.ZipException: invalid entry CRC (expected 0x482e3871 but got 0x1d296a66)
at java.util.zip.ZipInputStream.readEnd(ZipInputStream.java:378)
at java.util.zip.ZipInputStream.read(ZipInputStream.java:141)
at java.io.FilterInputStream.read(FilterInputStream.java:90)
at oracle.sysman.oii.oiix.OiixFileOps.copyStream(OiixFileOps.java:1470)
at oracle.sysman.oii.oiij.OiijFastJarExtracter.copyFileFromJar(OiijFastJarExtracter.java:307)
at oracle.sysman.oii.oiij.OiijFastJarExtracter.copyJarContents(OiijFastJarExtracter.java:227)
at oracle.sysman.oii.oiij.OiijFastJarExtracter.extract(OiijFastJarExtracter.java:146)
at oracle.sysman.oii.oiij.OiijJarExtractQueue$OiijJarExtractWorker.run(OiijJarExtractQueue.java:341)
Since that smelled like installation media file corruption, I checked cksum of the `linux_11gR1_database_1013.zip' and it was identical to the original on the OTN site, also "unzip -t <file.zip>" passed - no errors. Retrying installation (after deleting $ORACLE_HOME), doesn't help. It always hangs, but not at the same point in time. Sometimes it hangs at 5% of installation, sometimes at 20%, or 40%, but always during extracting phase, without OUI message, with similar errorlog.
Today I've tried it once again in order to figure out what's happening. It seemed I would have more luck today. The extracting and even linking phases succeeded, but, when it came to about 80% of installation (configuration pending), the installation suddenly aborted, OUI window disappeared from the screen, with error message visible only on console:
# An unexpected error has been detected by HotSpot Virtual Machine:
#
# SIGSEGV (0xb) at pc=0xb78d9725, pid=5690, tid=2767309712
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_11-b03 mixed mode)
# Problematic frame:
# V [libjvm.so+0x26c725]
Segmentation fault occured. No errorlog in /opt/oracle/oraInventoru/logs/, but java core file of 92 MB and java errorlog `hs_err_pid5690.log', showing stack traces, were generated in system's /tmp directory. The errorlog shows:
--------------- T H R E A D ---------------
Current thread (0x08a03c58): VMThread [id=5691]
siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x00000008
Registers:
EAX=0x00000000, EBX=0xb79f816c, ECX=0x000001fa, EDX=0x00000008
ESP=0xa4f1bbd8, EBP=0xa4f1bbe8, ESI=0xafde61c8, EDI=0xafde502c
EIP=0xb78d9725, CR2=0x00000008, EFLAGS=0x00210282
Top of Stack: (sp=0xa4f1bbd8)
0xa4f1bbd8: 000001fa 000001fa b79f816c afde4f98
0xa4f1bbe8: a4f1bc18 b78ed5ce afde4f98 a4f1bf40
0xa4f1bbf8: 00000001 b78ed581 ae758a48 afde7764
0xa4f1bc08: ae6051a4 b79f816c b79eb000 b7a06148
0xa4f1bc18: a4f1bc38 b78d9871 ae5f1138 afde4f78
0xa4f1bc28: 089b7d10 089b7d10 b79f816c b79efc44
0xa4f1bc38: a4f1bc58 b78d9b7c b79efc44 a4f1bca0
0xa4f1bc48: a4f1bc68 b796b16c a4f1bca0 b79f816c
Instructions: (pc=0xb78d9725)
0xb78d9715: 01 00 00 00 85 c9 75 1a 8b 46 04 51 51 8d 50 08
0xb78d9725: 8b 40 08 56 52 ff 50 5c 83 c4 10 8d 65 f8 5b 5e
Stack: [0xa4e9c000,0xa4f1d000), sp=0xa4f1bbd8, free space=510k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.so+0x26c725]
V [libjvm.so+0x2805ce]
V [libjvm.so+0x26c871]
V [libjvm.so+0x26cb7c]
... ... ...
... ... ...
Signal Handlers:
SIGSEGV: [libjvm.so+0x32a000], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
SIGBUS: [libjvm.so+0x32a000], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
SIGFPE: [libjvm.so+0x28e010], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
SIGPIPE: [libjvm.so+0x28e010], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
SIGILL: [libjvm.so+0x28e010], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
SIGUSR1: SIG_DFL, sa_mask[0]=0x00000000, sa_flags=0x00000000
SIGUSR2: [libjvm.so+0x290460], sa_mask[0]=0x00000000, sa_flags=0x10000004
SIGHUP: [libjvm.so+0x28fe90], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
SIGINT: [libjvm.so+0x28fe90], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
SIGQUIT: [libjvm.so+0x28fe90], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
SIGTERM: [libjvm.so+0x28fe90], sa_mask[0]=0x7ffbfeff, sa_flags=0x10000004
... ... ...
... ... ...
vm_info: Java HotSpot(TM) Client VM (1.5.0_11-b03) for linux-x86, built on Dec 15 2006 02:25:41 by java_re with gcc 3.2.1-7a (J2SE release)
These are only selected parts of the errorlog as it is too long to be presented in the whole.
What specially confuses me here is the fact that this OEL-5.2 installation is identical to mine, with same kernel, same kernel params, shared memory settings, etc. Also, this is obviously a java issue, but it is jdk1.5.0_11-b03 shipped with Oracle 11g. I've never had a problem with it.
Anyone has a clue what the heck is this and what causes those errors during Oracle 11g installation?
Thanks.
NJ