[nSLUG] kernel setup stack overlaps lilo

Paul Boudreau pbx at eastlink.ca
Fri Mar 12 11:59:41 AST 2004

Donald Teed wrote:

>I detect some sarcasm in this response left over from
>the previous debate.  Gentoo or any distro will provide
>people will a way to shoot themselves in the foot.
>But usually the way it happens is tricky for
>the developers to solve easily and often unique
>to the distribution.
>All I'm pointing out is that Slackware seems to
>have encountered an oops of the sort that was
>easily preventable given that this exact experience
>happened years ago in at least one major distro.
>If Redhat deserved shame for their oops, it is
>because they should have the resources to QA this.
>If Slackware deserves shame for this opps, it is
>because it is a simple mistake that has happened
>before and someone should have caught it.
>Note that I say "seems to" - I do not know for certain
>as I have not seen if there is at least a warning that
>Vikram missed.
>Sure you don't loose your system, but suppose you had
>cron'ed that update script like those poor Redhat 7.0
>users - you would be looking at unexpected downtime
>(for which reason may not be obvious) and in the land
>of machine rooms, downtime can be serious.
>People generally do consider Slackware to be very robust,
>so this is an exception to the rule, but it might
>be a sign of the future as it enters some growing
>pains of advancing features.  Having a larger
>developer base would help if Patrick wants to
>enhance things beyond the Toyota Tercel of Linuxes.
>On Thu, 11 Mar 2004, Paul Boudreau wrote:
>>Donald Teed wrote:
>>>>># Added Linux*
>>>>>And thats it.  I have my desktop back.  Thanks.
>>>Redhat got into deep doo-doo with users back around 7.0
>>>when the up2date utility would apply a new kernel and bust
>>>the system.  I don't remember whether it was just a question
>>>of rerunning lilo back then or not but that seems to be
>>>all this required.  Is Slack trying to catch up to the
>>>modern distros and making similar mistakes?  A big "tsk, tsk"
>>>if there wasn't even a warning produced by the new package
>>>management system.
>>>Whenever the linux kernel file is overwritten, it is not
>>>guaranteed to be at the same last physical location on disk,
>>>and thus lilo must be run.  This doesn't apply for grub or silo (sparc).
>>>nSLUG mailing list
>>>nSLUG at nslug.ns.ca
>>Yes, I'm sure nothing like this would ever happen on one
>>of the more modern distros, like Gentoo for instance, I
>>almost never see anyone having problems with files over-
>>writing others. And of course, you'd never have security
>>updates screwing up your Debian testing distro either.
>>Ah, you have to admire these Slackware die-hards and
>>the way they carry on in the face of adversity, with no
>>decent package manager, with no clue at all what they're
>>up against, when people who obviously know better try to
>>tell them otherwise. I wouldn't let it deter you from trying
>>to enlighten them regardless, someone has to do it. Poor
>>nSLUG mailing list
>>nSLUG at nslug.ns.ca
>nSLUG mailing list
>nSLUG at nslug.ns.ca
I'm just glad that we have a choice of numerous versions
of Linux, all of which do some things better than others.
In a world of "Free" software where SuSE sells for $250
million U.S., and Red Hat is a billion dollar company and
Oracle and IBM are spending wads of money, I find it
refreshing that Slackware has never sold out and remain
true to the spirit of the whole thing. People like Patrick
Volkerding, Klaus Knopper and others are what drives
this whole engine, not the lawyers and salesmen at IBM.
Perhaps if I was educated, knew more than I did and was
hoping to get a job and make big money off Linux that my
attitude would be different. Until then, I remain skeptical
and sarcastic. I find the money thing very distasteful.

More information about the nSLUG mailing list