----- Original Message ----- From: "Warner Losh via cctalk" <cctalk at classiccmp.org> To: "Murray McCullough" <c.murray.mccullough at gmail.com>; "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org> Sent: Thursday, January 04, 2018 1:05 PM Subject: Re: Spectre & Meltdown -------- Finally, an excuse to use all those old 486 boxes... m

On 1/4/2018 12:34 PM, Ed Sharpe via cctalk wrote: There is no difference between them and any other intel x86 or x64 processor as far as the flaw involved. Though they are not mentioning it, I suspect one can target P3 and P4 equally well with the exploit.? It has been around that long. thanks Jim

oh oh.... ? hopefully software ?fix ?soon? ? Ed# ? In a message dated 1/4/2018 1:54:43 PM US Mountain Standard Time, cctalk at classiccmp.org writes: ? From the exploit homepage (https://spectreattack.com/) , it seems like the Meltdown vulnerability affects all out-of-order executing Intel *branded* CPUs (from the P6 onward), and the Spectre vulnerability potentially impacts all superscalar processors of...all brands potentially :( Sophie On Thu, Jan 4, 2018 at 3:44 PM, jim stephens via cctalk < cctalk at classiccmp.org> wrote:

I misspoke - Spectre potentially affects all processors that use *pipelining and speculative execution*, not just superscalar ones (I mis-parsed "all modern processors capable of keeping many instructions in flight"). There's been ongoing patches to the Linux kernel for Meltdown (and for other OSes, though we can't read their mailing lists). For Spectre, though, it seems like solutions might take longer to distribute. It's kind of fascinating to run in to a cross-platform vulnerability like this! Is anyone else aware of similar vulnerabilities from history that also affected multiple processors, but relied on their implementation details? On Thu, Jan 4, 2018 at 4:03 PM, Ed Sharpe <couryhouse at aol.com> wrote:

Here here! I hope this is a wakeup call to all the people out there with all the unnecessary connected "lives". Forget all the social media BS but also the cloud storage, streaming everything (and not really having control of anything you "own"), IoT, and of course software as a service. As I understand it the exploits only work if run on the machine locally - which can occur if you run a malicious JavaScript through your browser.... So if you don't have to go on the internet just to run your email program to check your mail you (i.e. webmail or outlook online vs. POP/IMAP access) then you are more secure... I've gone out of my way to make sure I buy only equipment that I can connect to directly and is not dependent on some ephemeral cloud service whenever humanly possible. It may be less convenient or more technical but at the end it is always more secure (well not always but you get the idea). -Ali

You forgot "Outer Limits". I put that show in the same category. Wayne Sudol Riverside PressEnterprise A DigitalFirst Media Newspaper. On Thu, Jan 4, 2018 at 3:53 PM, Fred Cisin via cctalk <cctalk at classiccmp.org

Here's a PDP-10 subreddit: https://www.reddit.com/r/PDP10/

On Thu, 4 Jan 2018, TeoZ wrote: I'd be interested in hearing about opinions of the 100GB "M-disc". I've heard that they have decent longevity, and, the "low" capacity ones are interchangeable with conventional DVDs. I can still put 20 100GB DVDs (2017 technology) on a 2TB 2.5" Thin SATA. However, I'm also looking for multi-terabyte storage. Are higher capacity DVDs on their way? Howzbout multi-TearByte SSDs? I still want one of the ones that Kieth Hensen designed. Converting it suitable form factors and loading options) -- Grumpy Ol' Fred cisin at xenosoft.com

100GB M-Discs are dual layer BlueRay media correct (not readable on a DVD player)? I actually have a BDXL BR burner. I also have the M-Disc capable DVD burners but never tried that media on them. -----Original Message----- From: Fred Cisin via cctalk Sent: Thursday, January 04, 2018 9:38 PM To: General Discussion: On-Topic and Off-Topic Posts Subject: Large discs (Was: Spectre & Meltdown On Thu, 4 Jan 2018, TeoZ wrote: I'd be interested in hearing about opinions of the 100GB "M-disc". I've heard that they have decent longevity, and, the "low" capacity ones are interchangeable with conventional DVDs. --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus

On Jan 4, 2018 22:17, "TeoZ via cctalk" <cctalk at classiccmp.org> wrote: 100GB M-Discs are dual layer BlueRay media correct (not readable on a DVD player)? I actually have a BDXL BR burner. They are three-layer, and will ONLY work on BDXL drives, not older BD drives.

You're one of the first people I've heard quote that. Do you know where that is said? Years ago several friends and myself all picked up 64mb usb thumb drives so we could have multiple backups of a game and few other projects we were coding.? Maybe it was an extended period of time (we ended up switching to compatible removable drive bays) but 2 out of 3 of us lost all the data on our thumb drives around the same time.? I haven't heard may others share the concern but i wouldn't use ssd as a tech unless im forced to for that reason. -------- Original message --------From: Mazzini Alessandro via cctalk <cctalk at classiccmp.org> Date: 1/5/18 7:15 AM (GMT-06:00) I'm? not sure I would use SSD for long term "secure" storage, unless maybe using enterprise level ones. Consumer level SSD are, by specifics, guaranteed to retain data for 6 months if unpowered... any more time means being lucky. Would suck to save, store, and after some years find (over snipped)

On Fri, 5 Jan 2018, Mazzini Alessandro wrote: Yep! SSD would be very unsuitable for archiving. But, it is a nice fast medium for short-term uses. AND, it MIGHT be the first to get a unit larger than 2TB that will fit in a thin 2.5" form factor. Probably better SHORT-TERM reliability than the Seagate 2TB thin SATA spinning rust. What is the archival life of a BDXL, other than M-disc? M-disc media is a bit expensive. It looks like an excellent medium for data collections a tenth the size of what I'm playing with. It seems that it is still necessary to maintain multiple copies (geographically separate - we had a 4.4 quake yesterday morning), on multiple different media, and make new copies on a regular basis. -- Grumpy Ol' Fred cisin at xenosoft.com

That's curious. Then again, end of life for enterprise SSDs is many thousands of write passes over the full disk (or the same amount of writes to smaller address ranges thanks to remapping). Under high but not insane loads that takes 5-7 years. So presumably the retention while fairly new (not very worn) is much better. Still it's surprising to see a number that small. I don't know if the "cold storage" SSD stuff is going anywhere. But in any case, it seems to aim at high density at the expense of low endurance. I don't remember hearing retention discussed at all, higher or unchanged. Having drives with limited retention seems quite problematic. And "unpowered" suggests that leaving the power on would help -- but I don't see why that would be so. As for writable DVDs and such, do they have any useful retention specs? paul

On 01/05/2018 01:01 PM, Mazzini Alessandro via cctalk wrote: Here's a government-(Canada) sponsored paper on the subject: https://www.canada.ca/en/conservation-institute/services/conservation-prese… In any case, my DVDs will last longer than I will. --Chuck

On 01/05/2018 02:37 PM, Fred Cisin via cctalk wrote: I suppose that I should at least catalog my "alien" media--there's a lot of it. One of these days. It's hard enough to keep the stuff roughly alphabetized. At any rate, the latest study I can find on the longevity of optical media is this 2005 NIST paper: https://www.nist.gov/publications/nistlibrary-congress-loc-optical-media-lo… --Chuck

On Fri, Jan 05, 2018 at 09:33:47PM +0000, Bill Gunshannon via cctalk wrote: It was bad enough with the BBC Domesday project. Paper. Paper is the only way. acid free paper. -- - db at FreeBSD.org db at db.net http://www.db.net/~db

On Fri, 5 Jan 2018, Mazzini Alessandro via cctalk wrote: let's start with: "How old is the oldest surviving M-disc?" "Is there a form of machine readable data older than 150 years?" "Have they completely figured out Stonehenge?" [Don't you hate rhetorical questions?] They can postulate theoretical life. They can bake them, freeze them, hold them up in a tornado, use them as Starbucks coasters, etc., and then measure how much wear, damage, and aging they detect, and extrapolate from there. Have they successfully figured out what factors will shorten the life? Are there any factors that they missed? (such as a previously unnoticed air pollutant that suddenly becomes common (Sontaran Atmos?) that destroys the binder layer, or an imaginary wireless signal that corrupts all m-discs within range?) Every new development since, and including, CD has been hyped as unbelievably long lasting, such as "50 years!". Then, over time, they find out more about the risk factors, and the retention time gets shortened. Some of those "many decades" things ended up going bad in a few years. If you buy M-Disc, and it doesn't last 1000 years, . . . 1) Will Verbatim be around to honor a warranty? 2) Is that a full replacement warranty, or is it "pro-rated"? 3) How do you store a drive to last that long? 4) Is any of OUR data worth keeping long-term? Actual data about data retention time certainly exists. But, I couldn't trivially Google it. [Yes, "Google" as a verb resulted in a case claiming that once it becomes a verb, that the trademark becomes invalid! http://blogs.findlaw.com/supreme_court/2017/08/scotus-to-consider-whether-g… https://arstechnica.com/tech-policy/2017/10/supreme-court-wont-nullify-goog… ] 'course you will get a firehose worth of anecdotal crap, such as "I bought some discs on eBay, and they went bad" and "They last FOREVER! The discs that I bought in high school 3 years ago are still good!" I was surprised that M-disc.com didn't put up links (other than a bar graph without a scale!) providing the retention data for comparison. Does that mean that it is BETTER than we thought? But, M-Disc DOES seem like the best long-term form. So, I will get an M-disc compatible BDXL, and a stack of media (100G M-Disc are EXPENSIVE!) That action on my part, will surely ensure the immediate development and release of cheaper, more compact, faster, longer lasting storage systems, such as $5 10,000 year 10GB/S PetaByte thumb-drives ("The LottaByte NSA Utah Data Center https://nsa.gov1.info/utah-data-center/ will fit in your pocket") (or my death, and my heirs will instantaneously dumpster everything that I have) -- Grumpy Ol' Fred cisin at xenosoft.com

On Fri, Jan 5, 2018 at 1:44 PM, Paul Koning <paulkoning at comcast.net> wrote: There's method in this madness... The enterprise drives are trading retention for speed and endurance. When you design NAND storage devices, you basically get to pick two of the following to optimize for: speed, retention, endurance. Consumer drives are optimized for less endurance, better retention and decent speed. Enterprise drives are optimized for maximum speed, good endurance and OK retention. I've seen briefings that suggest higher retention for that class of product, but I've not seen the JEDEC profile for it yet. That's simple: powered drives have firmware running. The firmware can manage the NAND such that it reads the data in just enough time to ensure it can do it reliably before rewriting it to another part of the drive. Unpowered, it can't do that. No clue there. I worked for 3 years on a NMVE flash drive so got to learn all about this junk there since my job focused on NAND reliability and ensuring we met our benchmark numbers for speed, endurance and retention in the different configuration we sold. Warner

On 01/04/2018 05:50 PM, TeoZ via cctalk wrote: In my case, it's mostly a matter of the following: What can't I afford to lose? This falls into the following general ares: 1. Customer data (kept offline anyway, as it can be sensitive) 2. Code I have written and would be hard put to reconstruct. 3. Documentation that would be difficult to find again. 4. Emails. 5. Business records The rest I can afford to lose and could be duplicated if needed. This model has served me well for at least 40 years. --Chuck

Chuck, You do have to repower your machine. You never know what is still laying around in memory. If the byte are anywhere in memory, there is always some way to find them. Dwight ________________________________ From: cctalk <cctalk-bounces at classiccmp.org> on behalf of Chuck Guzis via cctalk <cctalk at classiccmp.org> Sent: Thursday, January 4, 2018 3:17:48 PM To: Sophie Haskins via cctalk Subject: Re: Spectre & Meltdown On 01/04/2018 01:08 PM, Sophie Haskins via cctalk wrote: Funny, I've been saying since the 1980s that it you have something that's critical to your survival, keep it offline. Until any of my PCs develop the ability to go to my storage cabinet and fetch a DVD and load it into itself, I'm not sorried. --Chuck

On Fri, 5 Jan 2018 11:18:53 -0800 Rick Bensene via cctalk <cctalk at classiccmp.org> wrote: Of course, update your OS as soon as updates are available, You want to test those updates before you apply them to remote production VM's. The latest CentOS 6.9 kernel update (2.6.32-696.18.7.el6 64-bit, which addresses meltdown) is broken on at least some Xen PV platforms and fails to boot. See https://bugs.centos.org/view.php?id=14336. You can't even get into grub "default x", in /etc/grub.conf, where x = working kernel, before you update. Hindsight being 20/20. jbdigriz