Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 02 Jul 1995 05:01:55 -0700
From:      David Greenman <davidg@Root.COM>
To:        "Rashid Karimov." <rashid@haven.ios.com>
Cc:        freebsd-hackers@freebsd.org
Subject:   Re: Random Lockups 
Message-ID:  <199507021201.FAA05455@corbin.Root.COM>
In-Reply-To: Your message of "Tue, 27 Jun 95 12:59:04 EDT." <199506271659.MAA13692@haven.ios.com> 

next in thread | previous in thread | raw e-mail | index | archive | help
>> On one of our 486's and now my Micron, the system will lock up solid causing
>> a hard reboot in the 486 case.  In the P100 case the system has to be reset
>> by the button.  Has anyone else experienced this?  This has happened quite
>> frequently with the 2.0.5R, never with the 0210-SNAP or 04-SNAP

   That's interesting.

>	Well, allegedly the lock-up could be caused by QUOTAs
>	code or "rlogin" command under cirtain circumstances
>	( it was reorted here a day ago) .

   The only fix that affected quotas was made on 3/10/95 and was to fix a
panic:

---
revision 1.4
date: 1995/03/10 22:06:39;  author: davidg;  state: Exp;  lines: +4 -3
Patch to fix quota panic from Mike Karels:

allow Q_SYNC regardless of "target" uid, we allow it with -1;
fix bug that caused all ops to refer to user quotas, not group.
---

>	If any1 here thinks about switching to 205 to run  the
>	server - DO NOT DO IT ! The system is not stable :( Alas :(

   Actually, 2.0.5R has proven to be quite stable for thousands of people.

>	Nodoby apparently is working on those mysterious lock-ups.
>	What's the position of the chief architect - Jordan ? 

   Well, I'm the chief architect... If there is a lock-up problem that can be
described in specific terms or one that I can reproduce here, I will see to it
that it gets fixed. Any bug which is undescribable and unreproducible is one
that we likely can not fix. If you would like to help with the debugging
effort, then there is just a chance that we might be able to fix it for you in
the next release.

>	Should we really be concerned about bleeding edge instead
>	of robustness ??? 1% enhancements and not the way to smoothly
>	upgrade the system with new version ? How can FreeBSD get 
>	its share of market w/o it ? Who will be willing to run 
>	the system but computer geeks ? 

   2.0.5R wasn't intended to be a "robust" release. We always strive to make
quality releases...2.0.5R is intended to get the new technology out to people
via a full release (including CDROM). 2.1R (which we are currently working on
along with 2.2) is intended to be a stable system. That has been our goal
since early this year.
   I'm sorry to hear that you are having problems. I don't know why you see a
problem on your systems while most people don't see it on theirs. If you
wouldn't mind providing some information about your systems and describe what
you do with them, and perhaps even provide some diagnostic information (add
DDB to your kernel, for example, and look at the states of processes when the
system is hung), then I can have something more to go by than just "my system
hangs and requires a hard reset". Do you really think this is very useful? It
isn't.

-DG



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199507021201.FAA05455>