8.04 restarts every hour.

[brianmrowe]

My box started to reboot every hour after I installed 2GB of memory bringing it to 4GB.
cat syslog | grep restart outputs:
Feb 11 01:34:52 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 02:35:07 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 03:35:26 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 04:35:44 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 05:36:01 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 06:36:19 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 07:36:37 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 08:37:00 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 09:37:24 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 10:37:30 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 11:37:48 saturn syslogd 1.5.0#1ubuntu1: restart.
Feb 11 12:38:07 saturn syslogd 1.5.0#1ubuntu1: restart.

Pretty consistent. I went to look at each entry before them the cron hourly runs then about 10mins later the restart happens.

Here are the statements before each restart:
Feb 11 03:09:01 saturn /USR/SBIN/CRON[6947]: (root) CMD ( [ -x /usr/lib/php5/maxlifetime ] && [ -d /var/lib/php5 ] && find /var/lib/php5/ -type f -cmin +$(/usr/lib/php5/maxlifetime) -print0 | xargs -r -0 rm)
Feb 11 03:17:01 saturn /USR/SBIN/CRON[6956]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly)

I noticed a few things after the restart that I thought were interesting, but not sure they are the problem:

Feb 11 03:35:26 saturn kernel: [ 0.000000] Linux version 2.6.24-23-generic (buildd@vernadsky) (gcc version 4.2.3 (Ubuntu 4.2.3-2ubuntu7)) #1 SMP Mon Jan 26 00:13:11 UTC 2009 (Ubuntu 2.6.24-23.48-generic)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-provided physical RAM map:
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 0000000000000000 - 000000000009ec00 (usable)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 000000000009ec00 - 00000000000a0000 (reserved)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 0000000000100000 - 00000000cf420000 (usable)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf420000 - 00000000cf422000 (reserved)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf422000 - 00000000cf4ee000 (usable)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf4ee000 - 00000000cf5ea000 (ACPI NVS)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf5ea000 - 00000000cf5ee000 (usable)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf5ee000 - 00000000cf5f3000 (ACPI data)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf5f3000 - 00000000cf5f4000 (usable)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf5f4000 - 00000000cf5ff000 (ACPI data)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf5ff000 - 00000000cf600000 (usable)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000cf600000 - 00000000d0000000 (reserved)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000f0000000 - 00000000f8000000 (reserved)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 00000000fff00000 - 0000000100000000 (reserved)
Feb 11 03:35:26 saturn kernel: [ 0.000000] BIOS-e820: 0000000100000000 - 000000012c000000 (usable)
Feb 11 03:35:26 saturn kernel: [ 0.000000] Warning only 4GB will be used.
Feb 11 03:35:26 saturn kernel: [ 0.000000] Use a HIGHMEM64G enabled kernel.

I'm researching these items now. After I installed the 4gig I only get ~3.2GB, which I was fine with, instead of messing up the machine. Of course rebooting every hour is pretty messed up for a box that hadn't needed a restart except for hardware and system upgrades.

I just can't figure out how to debug what is causing the reboot. It thought it might be temp at first but its runs cool and has 3 fans.
I'm running 8.04 with 4GB ram on Intel Quad.

I just can't figure out what to do next to fix it.
Any help would be great.

[Fire_Chief]

Although this may not fix the rebooting issue, you may want to install the PAE enabled kernel instead of the standard one. That would allow you to use all 4 GB of RAM.

The reboot issue looks too consistent to be a fluke. Check over the stuff that executes in the hourly cron and in the user cron jobs (if any).

Cheers!

[brianmrowe]

Although this may not fix the rebooting issue, you may want to install the PAE enabled kernel instead of the standard one. That would allow you to use all 4 GB of RAM.

The reboot issue looks too consistent to be a fluke. Check over the stuff that executes in the hourly cron and in the user cron jobs (if any).

Cheers!

Is it easy to install a PAE enabled kernel, or is there something bad I should know before I try that. I was going to just try to move to 64-bit Ubuntu, but I knew that would have some new challenges. I'm most worried about nvidia driver support for dual monitors.
Thank you.

[Fire_Chief]

Well, the move to 64-bit would be a re-install. Can't upgrade from 32-bit :(

As far as the PAE kernel install...have a look at this page (http://www.cyberciti.biz/faq/ubuntu-linux-4gb-ram-limitation-solution/#comments)

There may be some gotchas with hardware since the PAE enabled kernel is also designed for server installations but it may work perfectly.

Perhaps you could try the PAE approach first and then if it does not work like you expect, do a 64-bit install instead?

Cheers!

[brianmrowe]

Its interest issues.. I can't figure this out.
the is showing more than 4GB? Should 4Gb stopped at 4096?

#cat /proc/mtrr
reg00: base=0x00000000 ( 0MB), size=2048MB: write-back, count=1
reg01: base=0x80000000 (2048MB), size=1024MB: write-back, count=1
reg02: base=0xc0000000 (3072MB), size= 256MB: write-back, count=1
reg03: base=0xcf800000 (3320MB), size= 8MB: uncachable, count=1
reg04: base=0xcf600000 (3318MB), size= 2MB: uncachable, count=1
reg05: base=0x100000000 (4096MB), size= 512MB: write-back, count=1
reg06: base=0x120000000 (4608MB), size= 128MB: write-back, count=1
reg07: base=0x128000000 (4736MB), size= 64MB: write-back, count=1

Anyway - I did a shutdown -r now and the system did not reboot in the last hour. Also there are no entries in my /etc/cron.hourly

I'm installing a Video card tonight. Then I will disable the onboard video and check this output again.

Thanks

[brianmrowe]

This just stopped happening after I did the reboot. Ubuntu magic, I guess...

[Fire_Chief]

Did you apply any updates prior to the reboot or something?
I guess as long as the problem does not reappear... :)

Glad it's working for you now.

Cheers!

[cariboo]

Your system isn't restarting every hour syslogd is. Read what it says in the log


Feb 11 01:34:52 saturn syslogd 1.5.0#1ubuntu1: restart.

This is normal and there is nothing wrong with your system.

Jim

[brianmrowe]

While I agree with your comment about reading syslog, I totally disagree, about the system reboot. I just here while it shutdown and did a reboot again. Not sure why this happening. It just restarts.
Any help on this would be great. The syslog entry looks like a distraction not the answer. But the system still stops and restarts about every hour.

[brianmrowe]

:D
oh.. Just rebooted again. so this morning 8:58 and 9:59 it rebooted, by my watch. No warnings, just shutdown and restart.
Syslog did restart at 8:59 and 10:00, so maybe syslog restart entries are happing just after the system is starting??
Any ideas how I figure out why this happening??

Brian

[Fire_Chief]

You said this did not start happening until you installed the extra 2 GB of RAM. You may want to temporarily remove it and see how the box performs. Also if you have a spare system that uses that kind of RAM (or if not, use the server), may want to put it in there by itself and run a memory test on it to make sure it's good.

Cheers!

[chewbie]

Hi,

I have a very similar problem and I don't get it either...
If someone have any solution please help ^^

[dgermann]

Hi--

Have a look at this thread, particularly post #37 for a band-aid:

http://ubuntuforums.org/showthread.php?p=6830123#post6830123

Not an answer, but others are having this issue too. Even people outside Ubuntu http://www.linuxquestions.org/questions/linux-hardware-18/reboots-at-60-minutes-661688/page4.html I first discovered this band-aid from a person running red hat.

Is it time to post this as a bug? Where? To Ubuntu? To kernel people?

[Alwayslost]

Hi guys I am having the same problem as well on Ubuntu 8.10. I am not sure if this is actually the problem, but what happens in my case is it seems to be after I upgrade the kernel. The reason I say this if I run kernel 2.6.24-21-generic and every thing else is up to date I don't have the problem. But if I run kernels,
Ubuntu 8.10, kernel 2.6.27-11-generic,
Ubuntu 8.10, kernel 2.6.27-9-generic
Ubuntu 8.10, kernel 2.6.24-22-generic
The problem occurs. I have tried these kernels with and without my Nvidia driver installed with out any luck, but haven't tried the tune2fs method.
Hope this helps a little.

[dgermann]

Alwayslost--

Interesting. I'm on 8.04.1 on this box, so the kernel I am running is 2.6.24-23-generic. But it suggests I watch more closely to see if the problem is related to upgrades. Although I don't recall it being tied to kernel upgrades.

Try tune2fs and see what that does for you, when you have a chance.

If we get enough people trying stuff, maybe we'll get this figured out.

[dusan.saiko]

I would definitly reboot into memory test and let it run over a night.

(memory test should be in your grub menu or you can run it from most of the live distro CDs)

[dgermann]

dusan.saiko--

Thanks!

Interesting suggestion. You think it might be a glitch in memory? What leads you to that thought?

Anybody else see this as a strong clue? I do because I let this machine with the problem run continuously without rebooting for months. On the other hand, rebooting itself should cure a memory problem, yes? But the box keeps repeatedly rebooting till I can do the tune2fs thing.

Something to try the next time....

[Alwayslost]

I just tried the tune2fs method and so far touch wood no problems, been running 8 hours now. This problem has happened to me about 4 or 5 times over the last 12 months. My solution was to try different kernels until the problem went away. The tune2fs is a much easier solution. Thanks.

[dgermann]

Alwayslost--

You are welcome. Glad it worked for you.

If you get any other clues about what is going on here, please let us know!

[CGH]

hi, i've been having the same problem, the only thing i could get as an error was:
[code]
org.apache.commons.modeler.Registry registerComponent SEVERE: Error registering Catalina:t
ype=Valve,name=StandardContextValve,path=/jsp-examples,host=localhost javax.management.MBeanException: Cannot instantiate ModelMBean of class org.apache.comm
ons.modeler.BaseModelMBean ^Iat org.apache.commons.modeler.ManagedBean.createMBean (ManagedBean.java:385) ^Iat org.apache.commons.modeler.Registry.registerCom
ponent(Registry.java:835) ^Iat org.apache.catalina.core.StandardPipeline.register Valve(StandardPipeline.java:302) ^Iat org.apache.catalina.core.StandardPipel
ine.start(StandardPipeline.java:234) ^Iat org.apache.catalina.core.StandardContext.start(Sta ndardContext.java:4140) ^Iat org.apache.catalina.core.ContainerBa
se.addChildInternal(ContainerBase.java:760) ^Iat org.apache.catalina.core.ContainerBase.access$0(Co ntainerBase.java:744) ^Iat org.apache.catalina.core.Contai
nerBase$PrivilegedAddChild.run(ContainerBase.java: 144) ^Iat java.security.AccessController.d
May 15 12:54:32 dell-desktop jsvc.exec[5923]: talina.core.ContainerBase.addChild(ContainerBase.j ava:73 ^Iat org.apache.catalina.core.StandardHost.addChild(
StandardHost.java:544) ^Iat org.apache.catalina.startup.HostConfig.deployDescr iptor(HostConfig.java:626) ^Iat org.apache.catalina.startup.HostConfig.deployDe
scriptors(HostConfig.java:553) ^Iat org.apache.catalina.startup.HostConfig.deployApps( HostConfig.java:48 ^Iat org.apache.catalina.startup.HostConfig.start(
HostConfig.java:113 ^Iat org.apache.catalina.startup.HostConfig.lifecycleEv ent(HostConfig.java:311) ^Iat org.apache.catalina.util.LifecycleSupport.fireLife
cycleEvent(LifecycleSupport.java:120) ^Iat org.apache.catalina.core.ContainerBase.start(Conta inerBase.java:1022) ^Iat org.apache.catalina.core.StandardHost.s
tart(StandardHost.java:736) ^Iat org.apache.catalina.core.ContainerBase.start(Conta inerBase.java:1014) ^Iat org.apache.catalina.core.StandardEngine.start(Sta
ndardEngine.java:443) ^Iat org.apache.catalina.core.StandardService.start(Sta ndardService.ja
May 15 12:54:32 dell-desktop jsvc.exec[5923]: rver.start(StandardServer.java:700) ^Iat org.apache.catalina.startup.Catalina.start(Catalin a.java:552) ^Iat sun
.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ^Iat sun.reflect.NativeMethodAccessorImpl.invoke(Native MethodAccessorImpl.java:39) ^Iat sun.reflect.
DelegatingMethodAccessorImpl.invoke(DelegatingMeth odAccessorImpl.java:25) ^Iat java.lang.reflect.Method.invoke(Method.java:597) ^Iat org.apache.catalina.star
tup.Bootstrap.start(Bootstrap.java:295) ^Iat sun.reflect.NativeMethodAccessorImpl.invoke0(Nativ e Method) ^Iat sun.reflect.NativeMethodAccessorImpl.invoke(Nat
iveMethodAccessorImpl.java:39) ^Iat sun.reflect.DelegatingMethodAccessorImpl.invoke(De legatingMethodAccessorImpl.java:25) ^Iat java.lang.reflect.Method.invok
e(Method.java:597) ^Iat org.apache.commons.daemon.support.DaemonLoader.sta rt(DaemonLoader.java:177) Caused by: java.security.AccessControlException: access d
enied (java.io.FilePermission /usr/share/tomcat5.5-webapps/jsp-examples/WEB-INF/classes/logg
[\code]

but i don't use tomcat, maybe that's the error

Saludos!

[dgermann]

CGH--

Hmmm....

What do you make of those messages? They have lost me.

I do not have apache nor tomcat on my system, at least according to Synaptic. I have a program I was testing once called jbilling, and it has some or all of tomcat connected with it. I'd ought to see how I can go about uninstalling that....

[dgermann]

Hi all--

Might I suggest that we move this conversation over to the longer thread on this same issue, here (http://ubuntuforums.org/showthread.php?t=970006)

(Is there some way to get a sysop to mark this thread closed?) ):P

[stardvd555]

There may be some gotchas with hardware since the PAE enabled kernel is also designed for server installations but it may work perfectly