Bug 158075 - System freezes (UP and SMP) when use a stressed java application
System freezes (UP and SMP) when use a stressed java application
Status: CLOSED ERRATA
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: kernel (Show other bugs)
4.0
i386 Linux
medium Severity high
: ---
: ---
Assigned To: Larry Woodman
Brian Brock
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2005-05-18 10:03 EDT by Alexandre LABICHE
Modified: 2007-11-30 17:07 EST (History)
3 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2005-05-19 09:51:25 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)
/var/log/messages after Alt Sysrq T M (300.56 KB, text/plain)
2005-05-18 10:08 EDT, Alexandre LABICHE
no flags Details
/var/log/messages after Alt Sysrq T M (272.92 KB, text/plain)
2005-05-18 10:08 EDT, Alexandre LABICHE
no flags Details

  None (edit)
Description Alexandre LABICHE 2005-05-18 10:03:29 EDT
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; .NET CLR 1.1.4322)

Description of problem:
DL380 G2 uniproc
DL380 G3 SMP (2 Proc HT enabled or disabled)

Systems freezes erratically when use a Java Web Application server



Version-Release number of selected component (if applicable):
kernel-2.6.9-5.EL and kernel-2.6.9-5.0.5.EL

How reproducible:
Always

Steps to Reproduce:
1. Install Sun JDK1.5 and  a classic java stress tool like http://www.volano.com/pub/vmark2_5_0_9.class
2. If possible  use a memtester "value"  just enough to consume real memory and  leave 100 Mo real memory available (volano ./netserver.sh and ./netclient.sh already launched);  I think a little swapping system can produce crash very easily.  
3. launch a ps -ef after few minutes.  It doesn't return. Or  all is freezed.

Actual Results:  Alt Sysrq T shows  all java process including "ps -ef" not returning

ps            D 00000000  3000 26587      1               26494 (NOTLB)
c3227eec 00000082 ffffffff 00000000 ffffffff 00000000 00000000 00000000
00000000 c3227f3c c2835d60 00000001 00874612 97356883 0000025c f7e010b0
d8776d30 d8776e9c 00000000 00000000 e0c50001 f63c69b0 dcf91000 dcf91000
Call Trace:
[<c02c4de3>] rwsem_down_read_failed+0x143/0x162
[<c0127a58>] .text.lock.ptrace+0x7/0x23


Additional info:

if not freezed.  Reboot impossible ,  impossible to lauch several  programs . Only relaunch method is  to electrically shutdown/restart system.
Comment 1 Alexandre LABICHE 2005-05-18 10:08:05 EDT
Created attachment 114508 [details]
/var/log/messages  after Alt Sysrq T M
Comment 2 Alexandre LABICHE 2005-05-18 10:08:38 EDT
Created attachment 114509 [details]
/var/log/messages  after Alt Sysrq T M
Comment 4 Alexandre LABICHE 2005-05-18 12:21:42 EDT
After more investigations, whole  volano benchmark  use  2Go of memory. The net 
test works  great (10000 connexions after disabling ulimit restriction).  BUT 
if you have  less than 2 Go of real memory or  others processes consume extra 
memory (like memtester) , kernel is automatically dead, if  java processes  
begin to use some virtual memory.  It's very easy to reproduce.
Comment 5 Dave Jones 2005-05-18 23:22:25 EDT
Can you see how the U1 kernel fares on this ? You can grab a preview of it at
http://people.redhat.com/davej/kernels/RHEL4/
Comment 6 Alexandre LABICHE 2005-05-19 05:12:32 EDT
WHOOOO,

First results are great with kernel kernel-2.6.9-10  (UP  and SMP).  Now I can 
heavily  stress swap and thread(s) activity .  No process killed automatically, 
no freezes.  Just a  (normal) low response time because  of  heavy swap 
activity (kswapd)

First results are great !!!

Is it possible to know which patch  you have applied  because I have some 
others servers with custom 2.6 kernels (for ipvs version >= 2.6.10) ? 

Thanks, 
Alexandre LABICHE.
Comment 7 Larry Woodman 2005-05-19 09:24:05 EDT
Great, lots of fixes went into U1.  Please let me know if this bug can be
closed.

Thanks, Larry Woodman
Comment 8 Alexandre LABICHE 2005-05-19 09:33:02 EDT
Yes the bug can be closed.  All this day  I made  a lot of  stress tests 
without necessity to reboot any machine .  



Nota: I taked  srpms  2.6.9.10 sources  to test  this  on  other production 
machines  ( i can't  install now  a fresh RHEL 4 on these critical machines)

Thanks a lot.
Comment 9 Albert Graham 2005-05-19 16:29:09 EDT
f
Comment 10 Albert Graham 2005-05-19 17:16:29 EDT
Hello, 

Could Redhat please check for this bug against RHE3 Update 3/4 (Also was this
bug ported from RHEL3 to RHEL4 ?), We have had a simular problem with Java
causing Panics/Oops and JVM crashes on RHE3 for more than a year now, I've
reported them many many times and the general response I get from my
subscription support is "we don't support Sun Java!"

Basically under load, Java 1.4.2_xx thru 1.5_xxlatest will crash on all kernels
from > 2.4.21-15.04 thru 2.4.21-27.0.4. (2.4.21-15.04 does NOT crash Java
specifically but it does have lots of other Panic/opps issues) I only use the
SMP kernels and don't know if this happens with UP kernels ?

The problem with my setup is I cannot produce this bug at will ?, but it does
seem to happen when memory is full (8G of memory).

When the machine crashes it is either a kernel panic/oops or SIG 11 JVM crash
(usually crashes WHILE INSIDE garbage collection - which should not be
interrupted!).

Becuase I need a reliable Java environment, I am stuck in limbo land on kernel
24.21-15.04 and have to suffer many other kernel issues/panics that I know have
been fixed in later kernels.

I think this problem is kswapd/vm related but am not 100% sure.

Lastly, Alexandre in post #6 above asked "Is it possible to know which patch 
you have applied" ? could you please answer this question ?

My RHN subscription is once again up for renewal ?, I'm not sure that I should
continue with RHN having paid $1500 last year and the previous year ?, If I
renew again I would have paid $4500 ? and I'm still at first base ? - a not very
happy RH customer :(


Thank you in advance.

Albert Graham.

Comment 11 Alexandre LABICHE 2005-05-20 04:59:17 EDT
Hello,

Thanks to all contributors...

I confirm  Albert Graham's problems on redhat 3.  I think, this awful behavior 
affects Oracle database too when  a swap activity begins. We abandon linux 
platform for oracle (same problems with debian 2.4 kernels)

After analysing Changelog (I don't understand all things and  I don't know 
howto to recompile redhat 2.6.9.10 
(http://people.redhat.com/davej/kernels/RHEL4/)
 from scratch so I can't insert each patch to find which of them (or 
combinaison of patchs) is the GOD patch).

Some new patchs seem  to be very interesting (affects swap behavior)

linux-2.6.9-vm-dirty_ratio-initialisation-fix.patch
linux-2.6.9-vm-dma-zone-exhaustion.patch
linux-2.6.9-vm-improve-scanning.patch
linux-2.6.9-vm-mincore.patch
linux-2.6.9-vm-oomkiller-debugging.patch
linux-2.6.9-vm-oomkiller-tweak.patch
linux-2.6.9-vm-pageout-throttling.patch
linux-2.6.9-vm-page-writeback.patch
linux-2.6.9-vm-sc-congested.patch
linux-2.6.9-vm-swaptoken-null-mm.patch
linux-2.6.9-vm-tame-oomkiller.patch
linux-2.6.9-vm-total-scanned.patch


all of them (It's in a file,  but  are they really applied ???)

COPYING.modules
genkey
kernel-2.6.9-i686.config
kernel-2.6.9-i686-hugemem.config
kernel-2.6.9-i686-smp.config
kernel-2.6.9-ia64.config
kernel-2.6.9-ppc64.config
kernel-2.6.9-ppc64iseries.config
kernel-2.6.9-ppc.config
kernel-2.6.9-s390.config
kernel-2.6.9-s390x.config
kernel-2.6.9-x86_64.config
kernel-2.6.9-x86_64-smp.config
linux-2.4.0-nonintconfig.patch
linux-2.4.0-test11-vidfail.patch
linux-2.6.0-4g4g.patch
linux-2.6.0-compile.patch
linux-2.6.0-devmem.patch
linux-2.6.0-exec-shield.patch
linux-2.6.0-must_check.patch
linux-2.6.0-sleepon.patch
linux-2.6.10-ac-selected-bits.patch
linux-2.6.10-CAN-2005-0867-sysfs-signedness.patch
linux-2.6.10-net-3c59x-reload-EEPROM.patch
linux-2.6.10-net-e1000-update.patch
linux-2.6.10-net-e100-update.patch
linux-2.6.10-net-tg3-update.patch
linux-2.6.10-s390-cio-fix.patch
linux-2.6.10-s390-dasd_io_error-fix.patch
linux-2.6.10-s390-qdio_packet_loss-fix.patch
linux-2.6.10-s390-qdio_time_delay-fix.patch
linux-2.6.10-sata-updates.patch
linux-2.6.10-scsi-cciss-clustering-fix.patch
linux-2.6.10-scsi-midlayer-updates.patch
linux-2.6.10-scsi-qla2xxx-update.patch
linux-2.6.10-sysfs-update.patch
linux-2.6.10-x86-dma_declare_coherent_memory-kmalloc-args.patch
linux-2.6.11-s390-cio-vary_off-fix.patch
linux-2.6.11-s390-qeth_fake_ll-fix.patch
linux-2.6.11-sys_ipc-fix.patch
linux-2.6.1-s390-compile.patch
linux-2.6.2-ppc64-build.patch
linux-2.6.2-tux.patch
linux-2.6.3-crash-driver.patch
linux-2.6.3-ia64-build.patch
linux-2.6.3-printopen.patch
linux-2.6.4-stackusage.patch
linux-2.6.5-ext3-online-resize.patch
linux-2.6.5-ext3-reservations.patch
linux-2.6.5-x86-nostack.patch
linux-2.6.7-aic7xxx-diskdump.patch
linux-2.6.7-devmem.patch
linux-2.6.7-early-schedule.patch
linux-2.6.7-iobitmap.patch
linux-2.6.7-modsign-core.patch
linux-2.6.7-modsign-crypto.patch
linux-2.6.7-modsign-include.patch
linux-2.6.7-modsign-ksign.patch
linux-2.6.7-modsign-mpilib.patch
linux-2.6.7-modsign-script.patch
linux-2.6.7-netdump.patch
linux-2.6.7-scsi-whitelist.patch
linux-2.6.7-voluntary-preemption.patch
linux-2.6.8-4g4g-backout.patch
linux-2.6.8-crashdump-common.patch
linux-2.6.8-diskdump-3.patch
linux-2.6.8-diskdump-scsi-3.patch
linux-2.6.8-execshield-vaspace.patch
linux-2.6.8-ext3-reservations-update.patch
linux-2.6.8-ipr-diskdump.patch
linux-2.6.8-lockd-racewarn2.patch
linux-2.6.8-mptfusion-diskdump.patch
linux-2.6.8-netconsole.patch
linux-2.6.8-ppc64-netboot.patch
linux-2.6.8-print-fatal-signals.patch
linux-2.6.8-sym53c8xx-diskdump.patch
linux-2.6.9-4g4g-hugemem-warning.patch
linux-2.6.9-4g4g-maxtasksize.patch
linux-2.6.9-4g4g-noncachable.patch
linux-2.6.9-80211-update.patch
linux-2.6.9-80332-IOP-hotplug.patch
linux-2.6.9-ac-backouts.patch
linux-2.6.9-acpi-breakpoint-nop.patch
linux-2.6.9-acpi-debug-level.patch
linux-2.6.9-acpi-lequal-less-strict.patch
linux-2.6.9-acpi-reset-mechanism.patch
linux-2.6.9-agp-missing-cacheflushes.patch
linux-2.6.9-agp-posting-bugs.patch
linux-2.6.9-alsa-intel-hd-driver.patch
linux-2.6.9-ASPM-workaround-PCIE.patch
linux-2.6.9-autofs-leak.patch
linux-2.6.9-autofs-recognise-map-update.patch
linux-2.6.9-block-blkdev_get_blocks-EOF.patch
linux-2.6.9-block-cciss-id-updates.patch
linux-2.6.9-block-cciss-ioctl-returncode.patch
linux-2.6.9-blockfixes.patch
linux-2.6.9-block-__getblk_slow-hang.patch
linux-2.6.9-block-segment-coalesce.patch
linux-2.6.9-CAN-2004-1056-drm-insufficient-locking.patch
linux-2.6.9-CAN-2004-1137-igmp-flaws.patch
linux-2.6.9-CAN-2004-1235-do_brk.patch
linux-2.6.9-CAN-2005-0001-expand-stack-race.patch
linux-2.6.9-CAN-2005-0135-ia64-unwind.patch
linux-2.6.9-CAN-2005-0136-ia64-ptrace.patch
linux-2.6.9-CAN-2005-0176-shmlockperms.patch
linux-2.6.9-CAN-2005-0204-outs-iobitmap.patch
linux-2.6.9-CAN-2005-0207-NFSD-O_DIRECT.patch
linux-2.6.9-CAN-2005-0209-dst-leak.patch
linux-2.6.9-CAN-2005-0384-ppp-dos.patch
linux-2.6.9-CAN-2005-0400-ext2-infoleak.patch
linux-2.6.9-CAN-2005-0449-ip_defrag.patch
linux-2.6.9-CAN-2005-0531-size_t.patch
linux-2.6.9-CAN-2005-0736-epoll-overflow.patch
linux-2.6.9-CAN-2005-0749-elfloader-kfree.patch
linux-2.6.9-CAN-2005-0750-bluetooth-rangecheck.patch
linux-2.6.9-CAN-2005-0767-drm-radeon-race.patch
linux-2.6.9-CAN-2005-0815-isofs.patch
linux-2.6.9-CAN-2005-0839-N_MOUSE.patch
linux-2.6.9-CAN-2005-1263-binfmt_elf.patch
linux-2.6.9-CAN-2005-1264-raw-blkdev_ioctl.patch
linux-2.6.9-cciss-getluninfo-fix.patch
linux-2.6.9-compat-F_GETLK.patch
linux-2.6.9-cpufreq-silence-warnings.patch
linux-2.6.9-crashdump-fix-reboot-failure.patch
linux-2.6.9-dellserial.patch
linux-2.6.9-diskdump-badmsg.patch
linux-2.6.9-diskdump-condition.patch
linux-2.6.9-diskdump-dienmi.patch
linux-2.6.9-diskdump-export_state.patch
linux-2.6.9-diskdump-gendisk.patch
linux-2.6.9-diskdump-mdelay.patch
linux-2.6.9-diskdump-megaraid.patch
linux-2.6.9-diskdump-mem.patch
linux-2.6.9-diskdump-sata.patch
linux-2.6.9-diskdump-system_state.patch
linux-2.6.9-diskdump-wce.patch
linux-2.6.9-dm-64bit-fixes.patch
linux-2.6.9-dm-add-dm_dev-name.patch
linux-2.6.9-dm-avoid-bdget.patch
linux-2.6.9-dm-details-recordrestore.patch
linux-2.6.9-dm-dm_target_msg-ioctl.patch
linux-2.6.9-dm-export-mapinfo.patch
linux-2.6.9-dm-fix-mirror-log-refcount.patch
linux-2.6.9-dm-kprintf-tidy.patch
linux-2.6.9-dm-multipath-ioctl-ref-by-devno.patch
linux-2.6.9-dm-multipath.patch
linux-2.6.9-dm-multipath-suspend-requeueing.patch
linux-2.6.9-dm-raid1-deadlock-fix.patch
linux-2.6.9-dm-suspendhook.patch
linux-2.6.9-do_wait-hang-fix.patch
linux-2.6.9-dump_smp_call_function-3.patch
linux-2.6.9-dump_smp_call-i386.patch
linux-2.6.9-dump_smp_call-ia64.patch
linux-2.6.9-dump_smp_call-ppc64.patch
linux-2.6.9-dump_smp_call-x86_64.patch
linux-2.6.9-edd-config.patch
linux-2.6.9-emulex-lpfc-80163.patch
linux-2.6.9-emulex-lpfc-80166.patch
linux-2.6.9-emulex-lpfc-80166x2.patch
linux-2.6.9-emulex-lpfc.patch
linux-2.6.9-execshield-iret.patch
linux-2.6.9-exit-deadtask-nodentry-cache.patch
linux-2.6.9-exports.patch
linux-2.6.9-ExpressCard-hotplug-ICH6M.patch
linux-2.6.9-ext2-mbcache.patch
linux-2.6.9-ext3-cleanup-abort.patch
linux-2.6.9-ext3-file-limit.patch
linux-2.6.9-ext3-handle-bitmapdel.patch
linux-2.6.9-ext3-handle-double-revoke.patch
linux-2.6.9-ext3-mbcache.patch
linux-2.6.9-ext3-release-race.patch
linux-2.6.9-ext3-umount-leak.patch
linux-2.6.9-futex-disable-warning.patch
linux-2.6.9-futex-mmap_sem-deadlock.patch
linux-2.6.9-gpt-partition-noprobe.patch
linux-2.6.9-hotplug-msi-update.patch
linux-2.6.9-hugetlb_get_unmapped_area-fix.patch
linux-2.6.9-ia64-cpu-relax.patch
linux-2.6.9-ia64-cyclone-timer-fix.patch
linux-2.6.9-ia64-ia32_signal-memset-correctness.patch
linux-2.6.9-ia64-init-trigger-switch4.patch
linux-2.6.9-ia64-irq-routing-maxcpus.patch
linux-2.6.9-ia64-mmtimer-sn2-fix.patch
linux-2.6.9-ia64-pci-sn2-fix.patch
linux-2.6.9-ia64-qla1280-sn2-fix.patch
linux-2.6.9-ia64-rx1600-pdh-console-fix.patch
linux-2.6.9-ia64-sba_iommu-size.patch
linux-2.6.9-ia64-sgiioc4-sn2-fix.patch
linux-2.6.9-ia64-sgiioc-ide-workaround.patch
linux-2.6.9-ia64-sn2-update.patch
linux-2.6.9-ia64-sn-console-sn2-fix.patch
linux-2.6.9-ia64-snsc-sn2-fix.patch
linux-2.6.9-ia64-sys_waitid.patch
linux-2.6.9-ia64-tr_info-hang.patch
linux-2.6.9-ia64-zx2-idents.patch
linux-2.6.9-ide-blacklist-update.patch
linux-2.6.9-ide-cd-early-EOF.patch
linux-2.6.9-ide-cd-panic.patch
linux-2.6.9-ide-csb6-raid.patch
linux-2.6.9-idefloppy-suppress-noise.patch
linux-2.6.9-ide-supress-error-msg.patch
linux-2.6.9-ieee80211.patch
linux-2.6.9-intel8x0-sound-ids.patch
linux-2.6.9-invalidate-page-race-fix.patch
linux-2.6.9-ipw2100.patch
linux-2.6.9-ipw2100-update.patch
linux-2.6.9-ipw2200.patch
linux-2.6.9-ipw2200-update.patch
linux-2.6.9-irqaffinity-disable-E7xxx.patch
linux-2.6.9-kern_exit-race.patch
linux-2.6.9-lockd-block-nosleep.patch
linux-2.6.9-lockd-callbacks.patch
linux-2.6.9-lockd-NLM-length.patch
linux-2.6.9-lockd-reclaims.patch
linux-2.6.9-lockd-svc-reclaims.patch
linux-2.6.9-md-bi_max_vecs-fix.patch
linux-2.6.9-md-multipath-assemly.patch
linux-2.6.9-md-nr_pending.patch
linux-2.6.9-md-resync-bugs.patch
linux-2.6.9-module_version.patch
linux-2.6.9-net-b44-4g4g.patch
linux-2.6.9-net-bonding-panic.patch
linux-2.6.9-net-cmsg_signedness.patch
linux-2.6.9-net-compat-missing-security.patch
linux-2.6.9-netconsole-tg3-oops.patch
linux-2.6.9-net-e1000-64k-align-check-dma.patch
linux-2.6.9-net-e1000-avoid-sleep-in-timer-context.patch
linux-2.6.9-net-e1000-erratum23.patch
linux-2.6.9-net-e1000-flush-rmmod.patch
linux-2.6.9-net-e1000-post-mature-writeback.patch
linux-2.6.9-net-e1000-rx-mini-jumbo-inval.patch
linux-2.6.9-net-e100-e100_tx_timeout-workqueue.patch
linux-2.6.9-net-e100-fix-NAPI-state-machine.patch
linux-2.6.9-net-e100-ich7.patch
linux-2.6.9-net-e100-xmit-timeout-enable-interrupts.patch
linux-2.6.9-net-forcedeth-class-quirk.patch
linux-2.6.9-net-forcedeth-rx-csum.patch
linux-2.6.9-net-fragment-corruption.patch
linux-2.6.9-net-ftp_conntrack_leak.patch
linux-2.6.9-net-ip_options_leak.patch
linux-2.6.9-net-ipsec-sa-sequence-collision.patch
linux-2.6.9-net-ipv6-fix-mtu-calculation.patch
linux-2.6.9-netpoll-oops.patch
linux-2.6.9-net-procroute-stale-pointer.patch
linux-2.6.9-net-s2io-update.patch
linux-2.6.9-net-sctp-recv-accounting.patch
linux-2.6.9-net-sctp-sendbuffer-accounting.patch
linux-2.6.9-net-SIOCGIFHWADDR-NULL-dev_addr.patch
linux-2.6.9-net-sk98lin-module_device_table.patch
linux-2.6.9-net-sk_forward_alloc-BUG.patch
linux-2.6.9-net-tcp-bic-fix.patch
linux-2.6.9-net-tg3-fiber-autoneg-bounces.patch
linux-2.6.9-net-tr-irqlock-fix.patch
linux-2.6.9-net-tulip-waitdmastop.patch
linux-2.6.9-net-via-rhine-devinit.patch
linux-2.6.9-net-vlan-change_mtu-success.patch
linux-2.6.9-net-xfrm-fixes.patch
linux-2.6.9-NFS-auth-oops.patch
linux-2.6.9-NFS-cthon-rename.patch
linux-2.6.9-NFSD-add_set_client.patch
linux-2.6.9-NFSD-locallock-oopsfix.patch
linux-2.6.9-NFSD-non-null-getxattr.patch
linux-2.6.9-NFSD-portwarning-dottedquads.patch
linux-2.6.9-NFSD-putrootfh-return.patch
linux-2.6.9-NFSD-use_set_client4.patch
linux-2.6.9-NFS-locking-oops.patch
linux-2.6.9-NFS-mmap-corruption.patch
linux-2.6.9-NFS-mounthangfix.patch
linux-2.6.9-NFS-nlmcreds.patch
linux-2.6.9-NFS-stackoverflow.patch
linux-2.6.9-nonpower2sectorsize.patch
linux-2.6.9-overlapping-vma.patch
linux-2.6.9-pagevec-alignment.patch
linux-2.6.9-panic_on_oops-default.patch
linux-2.6.9-pci_mmcfg_write-flush_error.patch
linux-2.6.9-pcix-hotplug-fixes.patch
linux-2.6.9-ppc64-alloc-consistent-order.patch
linux-2.6.9-ppc64-cpu-hotplug-destroy_sched_domains.patch
linux-2.6.9-ppc64-cpu-hotplug-map-cpu-node.patch
linux-2.6.9-ppc64-cpu-hotplug-notifier.patch
linux-2.6.9-ppc64-cpu-hotplug-reinit-scheddomains.patch
linux-2.6.9-ppc64-cpu-hotplug-sched-domains.patch
linux-2.6.9-ppc64-cpu-hotplug-use-notifier.patch
linux-2.6.9-ppc64-eeh-recover.patch
linux-2.6.9-ppc64-eeh-reset-state2.patch
linux-2.6.9-ppc64-ensure-irqs-not-hard-disabled.patch
linux-2.6.9-ppc64-getpurr.patch
linux-2.6.9-ppc64-hvsi-hangup.patch
linux-2.6.9-ppc64-hvsi-reset.patch
linux-2.6.9-ppc64-hvsi-udbg.patch
linux-2.6.9-ppc64-ibmveth-getlink.patch
linux-2.6.9-ppc64-ibmveth-starve.patch
linux-2.6.9-ppc64-ibmvscsi-race-fix.patch
linux-2.6.9-ppc64-icom-driver.patch
linux-2.6.9-ppc64-idle-setup.patch
linux-2.6.9-ppc64-iseries-veth-mod-race.patch
linux-2.6.9-ppc64-lparcfg-paca-align.patch
linux-2.6.9-ppc64-numa-memhole.patch
linux-2.6.9-ppc64-numa-setup.patch
linux-2.6.9-ppc64-pci-hostbridge-hotplug.patch
linux-2.6.9-ppc64-prom-init.patch
linux-2.6.9-ppc64-purr.patch
linux-2.6.9-ppc64-rpaclose.patch
linux-2.6.9-ppc64-sighandler-stackalign.patch
linux-2.6.9-ppc64-signal-backtrace.patch
linux-2.6.9-ppc64-sigsuspend-regstomping.patch
linux-2.6.9-ppc64-singlestep.patch
linux-2.6.9-ppc64-tce-table-space.patch
linux-2.6.9-ppc64-tiocgicount32.patch
linux-2.6.9-ppc64-viocd-2.patch
linux-2.6.9-ppc64-viocd.patch
linux-2.6.9-ppc64-vpa-init.patch
linux-2.6.9-ppc64-vscsi.patch
linux-2.6.9-ppc64-xmon-early.patch
linux-2.6.9-procfs-deadtask-dereference.patch
linux-2.6.9-procfs-getpid-fix.patch
linux-2.6.9-ptrace-fixes.patch
linux-2.6.9-ptrace-sched-race.patch
linux-2.6.9-pty-smp-race.patch
linux-2.6.9-random-sysctl-overflow.patch
linux-2.6.9-rlimit_memlock-bypass.patch
linux-2.6.9-rpc-autherr-retry.patch
linux-2.6.9-s390-autoraid.patch
linux-2.6.9-s390-config_watchdog.patch
linux-2.6.9-s390-core_dump-fix.patch
linux-2.6.9-s390-dasd-fixed-buffer.patch
linux-2.6.9-s390-lcs_seq_numbers.patch
linux-2.6.9-s390-lcs_startup-fix.patch
linux-2.6.9-s390-no_hz_timer-fix.patch
linux-2.6.9-s390-qeth_addr-fix.patch
linux-2.6.9-s390-qeth-fake_ll-fix.patch
linux-2.6.9-s390-qeth_hipersocket-fix.patch
linux-2.6.9-s390-zfcp_port-fix.patch
linux-2.6.9-s390-zfcp-stackframe.patch
linux-2.6.9-sata-ahci-update.patch
linux-2.6.9-sata-lba48-max-sectors.patch
linux-2.6.9-sata-nth-page.patch
linux-2.6.9-sata.patch
linux-2.6.9-sata-updates.patch
linux-2.6.9-scsi-aacraid-dead-param.patch
linux-2.6.9-scsi-aac-remove-handle-aif.patch
linux-2.6.9-scsi-aic-hostraid.patch
linux-2.6.9-scsi-aic-leakplug.patch
linux-2.6.9-scsi-aic-oops-nohardware.patch
linux-2.6.9-scsi-bounce-limit.patch
linux-2.6.9-scsi-full-status-sg_io.patch
linux-2.6.9-scsi-inverted-refcounting.patch
linux-2.6.9-scsi-ioctl-overflow.patch
linux-2.6.9-scsi-ips-update.patch
linux-2.6.9-scsi-megaraid-kioc.patch
linux-2.6.9-scsi-megaraid-update.patch
linux-2.6.9-scsi-megaraid-warning-fixes.patch
linux-2.6.9-scsi-NULL-iterate-devices.patch
linux-2.6.9-scsi-qla-fix-hw-segment-counting.patch
linux-2.6.9-scsi-refcount-cmd-allocation.patch
linux-2.6.9-scsi-reset-ULDs.patch
linux-2.6.9-scsi-sg_cmd_done-sg_release-race.patch
linux-2.6.9-scsi-silence-sg_io-warning.patch
linux-2.6.9-scsi-test_unit_ready-cdrom.patch
linux-2.6.9-scsi-usb-forced-remove-oops.patch
linux-2.6.9-selected-ac-bits.patch
linux-2.6.9-selinux-avc-deadlock.patch
linux-2.6.9-selinux-avc-rcu.patch
linux-2.6.9-selinux-avc_update_node-spinlock-oops.patch
linux-2.6.9-selinux-destroy-avtab-node-cache.patch
linux-2.6.9-selinux-mediate-send_sigurg.patch
linux-2.6.9-selinux-netif-fixes.patch
linux-2.6.9-selinux-setscheduler-deadlock.patch
linux-2.6.9-selinux-setxattr-daccheck.patch
linux-2.6.9-selinux-sidtab-locking-fix.patch
linux-2.6.9-selinux-xattr-rework-tmpfs-mm.patch
linux-2.6.9-sha1.patch
linux-2.6.9-signal-handling-dr7.patch
linux-2.6.9-speedtouch.patch
linux-2.6.9-spinlock-debug-panic.patch
linux-2.6.9-statm-combined.patch
linux-2.6.9-sx8-sysfs.patch
linux-2.6.9-sys_io_setup-unwritable-addr.patch
linux-2.6.9-taint-badpage.patch
linux-2.6.9-taint-force-rmmod.patch
linux-2.6.9-taint-mce.patch
linux-2.6.9.tar.bz2
linux-2.6.9-timer-barrier.patch
linux-2.6.9-tmpfs-truncate-BUG.patch
linux-2.6.9-tty-locking-fix.patch
linux-2.6.9-usb-edgeport-overflows.patch
linux-2.6.9-usb-storage-reload.patch
linux-2.6.9-vc-resizing-overflow.patch
linux-2.6.9-vesafb-probe-error.patch
linux-2.6.9-via-apic-quirk-devinit.patch
linux-2.6.9-vm-dirty_ratio-initialisation-fix.patch
linux-2.6.9-vm-dma-zone-exhaustion.patch
linux-2.6.9-vm-improve-scanning.patch
linux-2.6.9-vm-mincore.patch
linux-2.6.9-vm-oomkiller-debugging.patch
linux-2.6.9-vm-oomkiller-tweak.patch
linux-2.6.9-vm-pageout-throttling.patch
linux-2.6.9-vm-page-writeback.patch
linux-2.6.9-vm-sc-congested.patch
linux-2.6.9-vm-swaptoken-null-mm.patch
linux-2.6.9-vm-tame-oomkiller.patch
linux-2.6.9-vm-total-scanned.patch
linux-2.6.9-waitid-bogus-ECHILD.patch
linux-2.6.9-x86_64-amd-dualcore.patch
linux-2.6.9-x86_64-change_page_attr-flush-fix.patch
linux-2.6.9-x86_64-clustered-apic.patch
linux-2.6.9-x86_64-configure-oops-stackdump.patch
linux-2.6.9-x86_64-copy_user_generic-exception.patch
linux-2.6.9-x86_64-flexmmap.patch
linux-2.6.9-x86_64-hit-hpet-twice.patch
linux-2.6.9-x86_64-hugetlb.patch
linux-2.6.9-x86_64-lost-edge-triggered-irqs.patch
linux-2.6.9-x86_64-mga-dri.patch
linux-2.6.9-x86_64-missing-compat-ioctls.patch
linux-2.6.9-x86_64-ni_syscall-overrun.patch
linux-2.6.9-x86_64-nmi-switch.patch
linux-2.6.9-x86_64-panic_timeout.patch
linux-2.6.9-x86_64-phys_proc_id-only-when-initialised.patch
linux-2.6.9-x86_64-srat-numa.patch
linux-2.6.9-x86_64-switch_to-missinglock.patch
linux-2.6.9-x86_64-syscall_signal-restart.patch
linux-2.6.9-x86_64-task_size-32bit.patch
linux-2.6.9-x86-enhanced-speedstep.patch
linux-2.6.9-x86-intel-ich7-ids.patch
linux-2.6.9-x86-share-cachedescriptors.patch
linux-2.6.9-x86-show_trace-irq-context.patch
linux-2.6.9-x86-sysenter-ebp.patch
linux-2.6.9-x86-sysrq-b-oops.patch
linux-2.6.9-x86-vsyscall-sysenter-unwind-info.patch
linux-2.6.9-xtime-correctness.patch
patch-2.6.9-ac11.bz2
Comment 12 Albert Graham 2005-05-20 06:25:39 EDT
Thank you Alexandre, I am extremely happy that you have confirmed this bug on
RHEL 3, thank you very much. 

I also tried Java 1.5x on my machine and get:

# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75c5820, pid=547, tid=2777164720
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_03-b07 mixed mode)
# Problematic frame:
# C  [libpthread.so.0+0x6820]  __pthread_mutex_trylock+0x10
#
# An error report file with more information is saved as hs_err_pid547.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#

within  a day of uptime.

Alexandre, Can you re-open this bug please as I don't have permission to do so.

I think this problem is related to: 

https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=144680

and

https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=132639

Also the "God Patch" was added after kernel 2.4-21.15.04 

I can also confirm that oracle will crash on all kernels >2.4-21.15.04 which I
also think is related to this bug ? - a real Show stopper for oracle users :(

If someone can produce a Beta Kernel for RHEL 3, I will test.

Thank you.

Albert.






Comment 13 Alexandre LABICHE 2005-05-20 10:08:21 EDT
Hello

I can't reopen this bug for redhat 3. For the moment it is better with this 
this beta RHEL4 kernel 2.6.9.10 .  And I definitly abandon 2.4 kernels after 
spending months of debugging of my (your? problem). 

For the fun  I have compiled  standard kernel 2.6.11-10 and compare with beta 
kernel.  I think the oom-killer is more aggressive (2.6.11-10). And if oom-
killer try to shoot  a java application server, I don't know why but system 
freezes particulary on java. oom-killer can't kill properly java processes 
(perhaps because of lot of threads inside java programs)


A good swap test and killing OS:

launch  your normal java application  (if oom-killer try to shoot java it's 
dead)

Recipe very easy to reproduce (example You have a 3 Go system you choose to 
launch every 5 minutes a memtester of 250 Mo)




In a crontab

*/5 * * * * /usr/bin/memtester 250

So kernel oom-killer after a while begins to shoot extra memtester
(normally ...) .

You leave this system one day.

Bye and good luck.
Comment 14 Alexandre LABICHE 2005-05-20 10:08:53 EDT
Hello

I can't reopen this bug for redhat 3. For the moment it is better with this 
this beta RHEL4 kernel 2.6.9.10 .  And I definitly abandon 2.4 kernels after 
spending months of debugging of my (your? problem). 

For the fun  I have compiled  standard kernel 2.6.11-10 and compare with beta 
kernel.  I think the oom-killer is more aggressive (2.6.11-10). And if oom-
killer try to shoot  a java application server, I don't know why but system 
freezes particulary on java. oom-killer can't kill properly java processes 
(perhaps because of lot of threads inside java programs)


A good swap test and killing OS:

launch  your normal java application  (if oom-killer try to shoot java it's 
dead)

Recipe very easy to reproduce (example You have a 3 Go system you choose to 
launch every 5 minutes a memtester of 250 Mo)




In a crontab

*/5 * * * * /usr/bin/memtester 250

So kernel oom-killer after a while begins to shoot extra memtester
(normally ...) .

You leave this system one day.

Bye and good luck.
Comment 15 Albert Graham 2005-05-20 11:52:02 EDT
On thing I have learnt from using RHEL is that you should wait for at least a
month or two before installing new Kernels regardless of the secrity implications.

I really like Kernel 2.6 but I am not in a position to use it in our environment
  at this time.

I have opened the same ticket refering to this bug for RHEL 3 so it at least
gets fixed (hopefully in Update 5 or 6)

https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=158328
Comment 16 Albert Graham 2005-05-23 15:50:10 EDT
Alexandre, Could I ask a huge favor of you ?

Could you run your crash test on RHEL 3 Update 5, I need to confirm that this
problem still exists in RHEL 3-U5, which I think it must, because RHELU5 was
released about the same time you first created this thread. ?

I'm not sure that this is going to be fixed unless it can be proven that it
exists in U5. 

Could you also report back either of these threads:

RHEL 4 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=158075

RHEL 3 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=158328

Thanks in advance.

Albert Graham.


Thanks in advance,
Albert.









Note You need to log in before you can comment on or make changes to this bug.