|Summary:||Kernel 2.6.11-1.14_FC3 hangs|
|Product:||[Fedora] Fedora||Reporter:||Ed Hill <ed>|
|Component:||kernel||Assignee:||Dave Jones <davej>|
|Status:||CLOSED CANTFIX||QA Contact:||Brian Brock <bbrock>|
|Version:||3||CC:||florin, ilaurie, pembo13, pfrields, wtogami|
|Fixed In Version:||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||2005-10-03 00:28:57 UTC||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
Description Ed Hill 2005-05-02 17:56:43 UTC
Description of problem: Kernel 2.6.11-1.14_FC3 hangs with no errors on-screen or in /var/log/messages. Version-Release number of selected component (if applicable): 2.6.11-1.14_FC3 How reproducible: Every few days or perhaps as often as once a day. Steps to Reproduce: 1. update FC3 to Kernel 2.6.11-1.14_FC3 2. wait for lockup 3. Actual results: Expected results: Additional info: Please see the thread starting at: https://www.redhat.com/archives/fedora-list/2005-May/msg00129.html where several people have described the same (or a very similar) problem. The symptoms are very real and hopefully someone will come forward with some helpful debugging info or tips to track down the cause(s).
Comment 1 Ed Hill 2005-05-02 18:13:59 UTC
The systems that have crashed for me with the 2.6.11-1.14_FC3 kernel are a ThinkPad T42p (model 2373-HTU) and a generic Athlon system built on an Asus A7V600 motherboard. The chashes were both complete lockups and neither of the systems were reachable through SSH connections. Both systems were running the unmodified FC3 kernel and there were no 3rd-party kernel modules installed (for instance, both are running the standard, updated FC3 Xorg RPMs and not any of gthe ATi or nVidia drivers). And in contrast, the 2.6.10-1.770_FC3 kernel has been completely stable for me.
Comment 2 Florin Andrei 2005-05-03 18:31:50 UTC
Something's very wrong with this kernel version. See bug #145258 for an instance where ntpd is killed suddenly. Also I've seen weird Postfix issues, but I cannot tell for sure whether they're linked to this kernel or not (I only looked at the logs _after_ upgrading to the new kernel, and i cannot re-test now). Maybe a devel version of the kernel tree must be branched off after all... ;-)
Comment 3 George König 2005-05-04 18:31:32 UTC
Actually I'm experiencing the lockups (generic Athlon system) simply waiting for some time (a few hours) or, simply plugging in two USB mass storage devices at the same time. I can't keep on this, reverting kernel to 2.6.10-1.770_FC3.
Comment 4 Ian Laurie 2005-05-06 06:07:40 UTC
I am also experiencing lockups with this kernel. In addition to updating the kernel, at about the same time I started using the "owner" module in iptables with syntax like "... -m owner --cmd-owner .....". Naturally I initially tended to blame the "owner" module in iptables. After backdating my firewall, the kernel has been more stable (no lockups for a few days now). I've tried to investigate as a prelude to making an intelligent bug report but the issue is so random I've not been able to gather reliable data. Also no hint of trouble in the log files. But.... on at least one lock-up, I could ping the box even though I could not telnet in, ssh in, or get a responce from the web server. So while the box appeared to be dead, the IP stack was still working, at least for a while.
Comment 5 Sitsofe Wheeler 2005-05-06 18:16:08 UTC
Comment #3 sounds a lot like #155472. Ed, are you using usb keys?
Comment 6 Ed Hill 2005-05-07 00:19:05 UTC
Hi Sitsofe, yes I was using USB keys sporadically back when the crashes occurred. I don't remember using the USB key often or at exactly the same time that the crashes happened but yes I was occasionally using a USB key. Also, I've stopped using 2.6.11-1.14_FC3 because the hangs were simply too disruptive. I've reverted back to a 2.6.10-1.770_FC3-based kernel and have had a week of blissful stability. I will try a newer kernel at some point but I hate to loose work to known-flaky kernels. :-/
Comment 7 Warren Togami 2005-05-07 00:42:18 UTC
People are complaining about several different things here with no commonality. The only way a problem can be solved is if it is isolated. This bug has no chance of that happening.
Comment 8 Ed Hill 2005-05-07 00:56:59 UTC
Yes, Warren, the posts here are probably not all the same thing. Granted. But declaring "no chance" is neither helpful nor informative. So, do you have any constructive suggestions on what can be done to help "isolate" and/or gather information?
Comment 9 Arthur Pemberton 2005-05-14 00:43:19 UTC
I am also encountering these lockups. I however have Fedora on two machines. One which I use as a general purpose server, the other I use as my desktop. Thus I am encountering these lockups at a very disheartening frequency. [SERVER] CPU Type: Intel(R) Celeron(R) CPU 2.70GHz CPU MHz: 2700.630 RAM Type: DDR RAM Amount: 512MB Run Level: 5 Number or lockups: 2 [DESKTOP] CPU Type: Pentium III (Coppermine) CPU MHz: 751.798 RAM Type: SDRAM RAM Amount: 512MB Run Level: 5 Number or lockups: 5 The lockups seem totally random. I've never noticed any warnign signs such as CPU at 100% before the lockups. I have notice no warnign signs. I have however noticed, at least on the server, that TCP/IP is the last to go, as ip forwarding can still occur even after IO has failed. I am still using this kernel, 2.6.11-1.14_FC3, hoping that I might be of some assistance in data collection on the issue.
Comment 10 Sitsofe Wheeler 2005-05-15 14:51:28 UTC
If you aren't Ed please don't add further comments to this bug - file a bug of your own! Ed: Did you take a look at bug #155472 ? Does using a newer kernel solve the lock up problem (assuming you used and removed a key before the lockup)?
Comment 11 Dave Jones 2005-07-15 19:48:21 UTC
An update has been released for Fedora Core 3 (kernel-2.6.12-1.1372_FC3) which may contain a fix for your problem. Please update to this new kernel, and report whether or not it fixes your problem. If you have updated to Fedora Core 4 since this bug was opened, and the problem still occurs with the latest updates for that release, please change the version field of this bug to 'fc4'. Thank you.
Comment 12 Dave Jones 2005-10-03 00:28:57 UTC
This bug has been automatically closed as part of a mass update. It had been in NEEDINFO state since July 2005. If this bug still exists in current errata kernels, please reopen this bug. There are a large number of inactive bugs in the database, and this is the only way to purge them. Thank you.