Linux.com

Actually, I'd personally appreciate it (as a member of the Linux Community) if you would give the proper credit to PCLinuxOnline. It was first reported on it and then<nobr> <wbr></nobr>/. Now, I'm not associated with PCLO nor<nobr> <wbr></nobr>/. but read both. With all the SCO nonsense going on about internal checks then it seems that we should have some internal checking when stating a story is reported "first" somewhere<nobr> <wbr></nobr>... or maybe just giving the impression that it is reported "first" somewhere<nobr> <wbr></nobr>...

Regardless, thank you for including that you contacted MandrakeSoft. It adds to the value of the story. Hopefully MS<nobr> <wbr></nobr>.. err<nobr> <wbr></nobr>.. I mean MandrakeSoft will not be hurt by this glitch.

#

The article does not state where this was first reported, nor does it really matter.

#

Yeah, it mattered to the author to refer to<nobr> <wbr></nobr>/. therefore it matters. Otherwise remove that line or add in PCLinuxOnline. Now it isn't a BIG deal<nobr> <wbr></nobr>.. but it is something that should be dealt with none the less.

Gesh, show some respect.<nobr> <wbr></nobr>:-P

#

I'm the one who submitted the news to PCLO. The source is the Cooker mailing list.

lbbros

#

And before it was on cooker, it was on the maintainers list. And before that, it was reported by some users on the forum at MandrakeClub.

Who cares, as long as the author of this story took the time to get a statement from someone at Mandrakesoft (as they did), rather than just some posting on some newsgroup (as Slashdot did).

#

Actually, I saw it on Warped Systems before I saw it on PCLinuxOnline (I just checked the post times, must be a time difference somewhere) Should they get credit too? And I think it was on the mailing lists before that. What about the guy who had the first drive burn? Do we hunt him down and give him a cookie?

#

Nonetheless...

- "Let's see, Sir! You ordered a full set of ISOs, Mdk CD 1 to 3, one-year club membership with extra-early rights to download new versions, mmkay?"

- "Yes, ok..."

- "Now, do you want LG fries with that, Sir?"

-=oOo=- -=oOo=- -=oOo=- -=oOo=-

I use Mdk because it's more advanced... sometimes, living on edge can lead to falls...

#

The blame is on LG not on Mandrake.
Read ALL info before blah-blah-ting

#

Thanks to PCLO : "According to a post by Juan Quintela (member of MDKsoft's kernel team), the issue with the LG drives was found. Apparently, it's due to the buggy behavior of those CD-ROM drives. The command UPLOAD_FIRMWARE reuses the command FLUSH_CACHE. As you can imagine then, the result is a fried drive.
An updated kernel is on the QA queue. As for the ISOs, there have been talks of remastering, although it's not sure how the packs will be affected (they're already gone to press). "

Interestingly, some other distributions with kernel 2.4.22 are affected. I don't know yet for the mainstream distros.

#

An updated kernel is on the QA queue. As for the ISOs, there have been talks of remastering, although it's not sure how the packs will be affected (they're already gone to press).

I do hope they remaster and recall the packs already sold (if any). It would be expensive, but not doing it would probably be even more so!

Even though this appears to be (based on what I have seen so far on web forums) squarely LG's fault, it is MandrakeSoft who would take the rap (possibly even lawsuits) from customers, whose drive gets toasted by the package they have bought...

#

This is not mandrakes fault. Mandrake wich is a small company cannot check evry possible hardware combination. There are thousends off possibility's.
The fault is at LG-electronics. They refuse to work with the Free Software community. They don't test there hardware for GNU/Linux. Its a backward retarded vision.

If you have a drive that gets broken due installing Mandrake. Ask LG for a refund. Even if your warrenty is void. Its a design flaw.

Compare it to the car manufactory's. If they find a fatel flaw in model that is 7 years old the will recall the model and repair it. Its almost the same...

So write LG asking that they will refund a new drive when it gets broken. Explain them software never should be able to do that and that they are responsible!

Hell, in Amerika somebody could sue them!

#

and...how many kernel ticks will it take before someone write a win32-virus that uses this "feature" to fry your nextdoor-mostly-Dell-and-win32-only-dot-com?

p@sweden

#

Bah. Assuming the fault was with LG's firmware it is undeniably LG's fault. I say assuming because the above article update makes this unclear. Your logic though is a little backwards. I'm assuming some of these drives were very likely out before Mandrake 9.2, I believe the article says the suspect code in the kernel was introduced 8/15 and most likely Mandrake 9.2 was introduced after that<nobr> <wbr></nobr>:P. So your saying that all hardware companies should have to constantly retest their hardware on every software/hardware combination available.. even for just Linux this is a somewhat unrealistic task considering the frequency of releases and available (optional) patches. Now if you pay attention in class there are many more open source operating systems beyond Linux so should companies "embrace" open source your going to have to learn how to share. There are also several different platforms which support IDE CD-ROM drives. The variables are limitless and in an open-source world it falls upon the users to test the code. It may not be fair but its the only way things are going to get done. Mandrake needs to accept some responsibility and recall the products they've published.

#

A new petition is available online to boycott LG products. It can be found here --> http://www.thepetitionsite.com/takeaction/1904982<nobr>5<wbr></nobr> 3



Hardware manufacturers need to wake up and realize that they need to adaquately support their products for more than a single platform.



------

<A HREF="http://thekauze.cjb.net/" TITLE="cjb.net">Chickenman</a cjb.net>

#

Did you bother to read the headers of any of the other petitions they support? That one won't get any traction. If the intended petitonee looks to see who is actually fussing they will find it put on by folks that love to hear the sound of their own self importance. I am surprised. As left leaning as that site appears that it simply doesn't fall over.

#

I hope the Redmond vultures don't swoop down on this and use it to accelerate their FUD campaign.
This should also give pause to the people in the Open Source software in government efforts. Maybe before they put forth these blanket (must consider Open Source) bills, they should consider first putting forth simple, separate, and with unimpeachable intent, legislation that would ban future government purchases of any hardware that is not compliant with all viable alternative PC software and operating systems. Thereby avoiding future (CDROM like) problems, insure governments ability to adopt open source solutions at any future point. And would force hardware manufacturers like LG electronics to not only test all of their products with OSS software, but also provide alternative OSS device drivers (even binary if they must) for them as well. (what mfg would allow themselves to become non compliant or ineligible for government contracts?)
Of course everyone sespects that Microsoft is behind the device driver sand bagging.
This point, and line of thought, might provide the open source community a very good counter strategy to any FUD storms that may develop from this unfortunate Mandrake situation. Actually giving us a legitemate cause to bring to the forefront the lack of manufacturers compliant OSS hardware and driver suport.
The lack of available, fully compliant hardware with their essential device drivers, are a huge exploitable gray area loophole in the must consider Open Source bills as they stand now anyway.

#

What the government and everyone else should be insisting on is that CD and DVD ATAPI devices conform to the SCSI MMC standard.

It sounds to me like LG Electronics invented their own command for downloading firmware instead of using (or at least not overloading) the commands defined in the MMC standard.

One of the nice things about having the MMC standard is that testing against it goes a long way towards ensuring a device will work with any OS and driver combination, present or future.

Laws requiring OSS compliance are bound to be shot down in court, if they have a chance of passing at all. For just one example, it could be argued that requiring vendors to test against all existing OSes would be an unfair burden. To require testing against a subset of OSes would be giving unfair competitive advantage to those OSes.

Instead we should be insisting that the government gives preference to software and hardware that complies with open standards when those standards exist (and by open, I also mean unencumbered by non-free patents, copyrights, etc.)

#

Burdens are only unfair if they are applied to a single entity or are not applied accross the board. By requiring only the purchase of hardware that could also be used with viable OSS software, all harware manufaturers would be in the same position. As a result they would have to routinely, as a matter of regular course, support current viable OSS solutions.
If you re-read the post, you will find I mentioned (all "viable" alternative OS)these could be identified in an amendment wich should include your standards based criteria.
Your premise implies, any bills that require compliant hardware have no chance of passage or of being upheld in the courts. A single, separate bill focused solely on hardware compliance, has a far better chance of being passed and upheld, than these blanket must consider open source bills which are flowery rags filled with grey area loopholes and open to corruption (Yeh, We considered Open source software for our department, but we had one Laptop with a centrino chip on it that isn't supported, so we went with Microsoft for everything."excuse","cash graft check").
Similarly, a single, separate bill focused solely on open document standards should be put forward. "Take it One step at a time!"

#

LG says that any complaint should be adressed at the your local hardware shop or DELL...
Hmm lets write DELL about this

#

Well, I know that I won't be purchasing LG equipment anymore. Their response has not been what I would consider acceptable.

There are several comparatively low-cost drive manufacturers available. Dell and others should simply switch to a manufacturer that takes responsibility for their error. Equating "Flush Cache" and "Upload Firmware"??? That can't have been an intentional act, so it's a defect.

This isn't a Linux issue; if Microsoft had issued a service pack that used the same command, the result would have been wholesale, widespread destruction (and probably a class-action lawsuit). As it is, LG had the opportunity to correct their error, at low cost, and look like heros. Instead, they turned their backs on impacted customers.

If they are not going to take responsibility for their product, why buy them?

#

You rock man,
But all those outsiders "layman" of the IT won't know the diffrence and blame Mandrake and GNU/Linux.

#

"Instead, they turned their backs on impacted customers."

It seems to me that their "impacted customers" should have seen a proctologist instead.<nobr> <wbr></nobr>....sorry, couldn't resist.<nobr> <wbr></nobr>:::getting more coffee:::

#

I've always been a fan of "good hardware" solutions...uP's by AMD, MoBo's by MSI, scanners by Epson, CD/DVD-anything by HP....that kind of thing.
I can certainly empathize with folks who are stuck with orphaned hardware (I'm still the semi-proud owner of a Teac 2.88MB floppy drive.); but, seriously, folks, LG drives are THE bottom-of-the-barrel.
I've never run into a more problematic drive in my life.

Sure HP/Compaq, Dell, Gateway put LG drives in their systems. It's the default, buck-and-a-half, margin-boosting, CD writer that's just barely good enough for those who don't know the difference between hardware and a compiler directive. Most mfrs provide a means of specifying something OTHER THAN LG in the system configuration order--for a ridiculously low price differential.

Hopefully this will not play into M$'s FUD goals (though I can well imagine Steve Ballmer is busy crafting his next speech about how Linux will physically nuke your motherboard and processors), and will serve as a wake-up call to Mfrs/OEMs/integrators...don't use crappy hardware components.

If you've got one of these paper weights in your own system, and it's still under warranty, contact your Mfr/OEM/Integrator to discuss the possibility of swapping the offending device out for something substantially better. If your warranty is dead, then bite the bullet and replace it (the drive, not the machine...unless you looking for an excuse) outright.
In either case, ask about components on all future purchases, and insist on the highest-quality hardware that your budget will allow.

#