Mysterious Avid Issue Knocks Out Mac Pro Workstations Across Hollywood | Hacker News Reader

Mysterious Avid Issue Knocks Out Mac Pro Workstations Across Hollywood

variety.com

242 pointslnguyen200 comments

yikes. having written aax plugins for a while, i have to say, i’m kind of surprised this came from AVID. from what it sounds like (having scraped the surface), it’s not their editing software (Pro Tools, et al.) but their copy protection management tool/device driver - iLok.

i’m sure there are some very anxious engineers in the AVID offices today.

> their copy protection management tool/device driver - iLok.

They're still using that? These things were successfully cracked in the nineties, I was convinced nobody uses hardware dongles anymore.

yes, they are still an industry standard.

Buy your third-gen iLok today for only $49: https://www.avid.com/licenses/ilok

I believe the current generation is stronger.

Oh, it's not just Avid. Some audio production software such as ReFX Nexus still uses a USB dongle. There is no way to use the software without it plugged in.

It also has a policy where if you damage the dongle, you have to buy again and get a 10% discount off the FULL price of the VST.

In the age of people gigging on laptops, I'm not entirely sure how they think they are gonna sell any more of them. I fancy it myself as it has quite a few 'familiar' sounds on it, but not with all the downsides.

To add to that, it's not just media production.

In the medical field, almost all software for performing and analyzing EKGs, X-rays, MRIs etc. comes by default with some sort of dongle-based protection. Apparently, it's not enough that you need the biggest dongle in the world (i.e. the machine itself!) to use the software. Some vendors do offer alternatives if you fight them hard enough but the default still is a trip to dongle town.

Ah, but they need to control that CD recording device on the workstation, just in case you're handing out CDs to your patients WITHOUT A CD RECORDING LICENSE. They have the weirdest licenses.

So there is software on a medical device concerned with users copying a Billy Joel CD?

They sell every functionality as an aditional "license" or "plugin".

So, for example, it's not unusual that, by default, your half a million USD scanner, that includes a powerful workstation to control it (the one the technician uses behind the window) will only let you write CDs (for example, to give your patient a copy of his/her scan) if you pay an additional "CD Writing license"

Now, with PACS and centralization that have changed a bit, but still...

Thanks for the explanation. Still gross but not as bad as RIAA checks being performed somehow.

Apparently, it's not enough that you need the biggest dongle in the world (i.e. the machine itself!) to use the software.

Same for a lot of other industrial equipment. In my (admittedly limited) experience, they get cracked pretty quickly by the owner, either before the factory ever gets any downtime, or right after the first time.

This is fascinating, thank you for sharing it. Medical technology fascinates me, but I am nowhere near the industry. Same with things like CarPlay et al.

> In the medical field, almost all software for performing and analyzing EKGs, X-rays, MRIs etc. comes by default with some sort of dongle-based protection. Apparently, it's not enough that you need the biggest dongle in the world (i.e. the machine itself!) to use the software.

Actually, it's not.

Some idiot will get the idea that they can just "transfer the hard drive immediately" because they didn't pay enough money on their support contract to get 4 hour response rather than 48 hour response. The dongle prevents that.

Medical devices are certified for the entire device--that includes the specific PC and the specific version of operating system running on it.

I used to work with cat scan machines a lot, and I found out how to emulate their dongle - you just returned 0 when it called for the check...

> There is no way to use the software without it plugged in.

This is only true if the software hasn’t been crack?

Thing is, the majority of iLok protected software has never been cracked, or if it has been cracked, only very old versions that aren't exactly relevant anymore.

Ah, right, fair enough :)

The problem is, there are some industries where the piracy rate would be 100% if it weren't for dongles. Music production software, for one. And another surprising one is church software! (They feel they shouldn't have to pay because they're a "church" and serve a higher purpose.)

I go to the NAMM (musical instrument manufacturers) trade show every year and work with music, sound, and church software vendors.

Embroidery software is another ( https://www.wingsxp.com/ )

Steinberg Cubase uses these aswell

Cubase uses eLicenser, not iLok, and it hasn't been cracked since a version about 10+ years ago so it does a pretty good job - even if that does mean cubase has fallen in popularity.

The new Mac Pro tower has an internal USB-A port, explicitly for this reason [1].

1: https://appleinsider.com/articles/19/06/04/apples-new-mac-pr...

I look at what the industry has allowed itself become and cringe.

iLok doesn't require the hardware dongle anymore, licenses can be installed onto the hard drive now. Some people prefer the dongle because they can carry their licenses from machine to machine, studio to studio etc.

But as a user, I still much prefer the old-school serial number unlock without internet activation, though.

It depends - a lot of software still requires the hardware dongle and it's much more secure than the machine licensing option I'm told.

Actually the current generation of iLok hasn't been cracked yet (and from what I've heard it's unlikely that it ever will be).

There was a time a few years ago when the wrapper got cracked but as far as I know the security itself is still strong.

> but their copy protection management tool/device driver - iLok.

Maybe Hollywood will learn that DRM hurts legitimate users.

But they probably won't, I won't be shedding any tears either way.

Ironically film tools have some of the most stringent copy protection systems, and big studios break the licenses agreements without hesitation when they need to get something done.

I wonder if Hollywood is so obsessed with DRM because they themselves are such f'ing pirates. The entire culture of Hollywood is around stealing (e.g. the endless so-and-so stole my script idea lawsuits) and defending oneself from having one's stuff stolen. Maybe they're projecting and imagining that everyone is constantly trying to steal everything.

As a normal movie and TV viewer I will always look for an option to rent, buy, or stream stuff because I like to support the production of stuff I like and because the time it takes to jank around with pirated crap is generally worth more than the cost of just paying. The only time I've turned to piracy in recent (post-broke-college-kid) memory is when something is literally unavailable by any other means. I suspect I'm typical.

DRM has zero effect on my behavior. If anything DRM encourages me to consider pirating because it adds pain and hassle. In the case of actual purchases (vs streaming and virtual renting) I really despise DRM since I purchased a copy and should be able to play it on any device I have and store it for posterity. I generally will not outright purchase DRMed content.

Since we're talking Hollywood, what's your success rate in being able to buy movies that don't have DRM? I can't think of any way to do that, really. Or are you saying that you basically can't, and therefore don't buy movies and instead stream or rent?

Music yes, movies no... you are right. That's why I don't buy many movies.

Physical media has DRM but it's pretty much completely cracked. I can rip Blurays and even UHD Blurays now and end up with a non-drm-ed movie. It's not nearly as convenient as streaming (initially) but I can "own" the movie in a non-drm-ed form for as long as I can keep the bits alive.

How can you rip UHD (4K) movies? MakeMKV doesn’t support them last I used it (a month ago)

MakeMKV does support them, but you need a UHD drive with specific firmware. On the forums you can figure out which drive will work the best and which is the easiest to flash old firmware on. Either way requires a hacked firmware updater because the drive makers have to give it their best shot to avoid downgrades (to satisfy DRM licenses).

Also you can decide between old stock firmware or new firmware that has been modified by the MakeMKV team. I opted for the old stock firmware, I'm not really sure what the modified firmware buys you.

It seemed overwhelming at first, but after an hour of reading and searching on Amazon I figured out what the easiest route was. Once the drive arrived, the flashing was simple—I plugged the drive into my Windows box and ran the flashing code, then plugged it into my Linux Plex server and it worked right away.

A huge color company out in the valley is notorious for calling up 120 days past the invoice date on NET30 terms and only offering to pay on a credit card. I swear some of these companies would spend a dollar to screw you out of 50 cents.

It would be fun if those affected companies put a nice lawsuit for the downtime they're facing, but they might not want to rock that boat

I hope you don't think that the people dealing with issues like this are anywhere near the discussions on DRM of digital video. It isn't as if MPAA board members are also in a basement somewhere editing films.

> their copy protection management tool/device driver - iLok

MacOS Catalina makes running kernel extensions hard, they probably were testing new iLok version for Catalina.

Catalina major change is ANY dll should know pass through security. That's why notirisation also for software plug-ins is needed meaning also runtime hardening in future and proper entitlements by host and plug in

Your comment is a bit hard to parse. Would you mind clarifying it a bit?

I’ve started to write something but here is a good tl;dr - https://forums.developer.apple.com/thread/122566

Right, Library Validation is tightened when you opt-in to Runtime Hardening.

> MacOS Catalina makes running kernel extensions hard

So it has bad compatibility?

Is that a good thing which we’re excusing now?

OT, what plugins? I’m a keen Pro Tools user!

A few of the Twitter replies say they've seen it on non-AVID devices, which might lend more credence to the iLok theory.

https://twitter.com/FamousJerry/status/1176422298535940097

I'm kinda hoping iLok is the culprit, because I've been refusing to buy any plugins that require iLok, even though that means avoiding some effects I'd really love to have & developers that I'd love to support. Thankfully iLok is optional for the iZotope stuff...

The iLok is just a repository of keys and can’t hurt anything by itself. It’s communicated with by code either built into a license protected application or in a “wrapper” added to an otherwise unprotected app. Pace AP (maker of the software and reseller of the iLok) is essentially in a never ending arms race with the warez crowd, and in the past has resorted to some gray-hat, rootkit-like techniques to prevent cracking. It wouldn’t surprise me if this was either some old Pace code going rotten or some sort of crack gone viral/rogue. I used to develop an iLok protected app and am still on their developer mailing list, but note that I haven’t seen any recent alerts from them so it’s entirety possible it’s something else.

Yep, after I posted I realized that if this is really widespread & iLok related, surely someone would have posted on Gearslutz complaining about it by now. So far I can't find any threads about it there.

In my iLok case years ago, it was a problem with the iLok KEXT kernel extension drivers causing my Mac to kernel panic at startup that put me off. That would've been around the Mac OS X 10.5 / 10.6 era.

I bought a plugin from UVI and had to use iLok for the first time recently and the experience was atrocious.

Not only that, if you want to sell and transfer your iLok license you have to pay $20.

I will never buy anything that requires iLok.

Ironically, the folks pirating the software didn't run into any interruption because there is no iLok

iLok has been extremely effective at stopping cracks or at least slowing them down enough to where the software is obsolete by the time a crack appears. The pirates don't have access to this software in the first place.

We can debate the merits of DRM all we want, but PACE's stuff actually works (at stopping pirates, and sometimes paying users)

It is truly a cancer and the worst part about using Avid tools. I wish they would get rid of it.

This is what Avid sends you after creating an account:

Your iLok account allows you to:

  - Use the free iLok License Manager application to manage your licenses
  - Take advantage of Zero Downtime and Theft & Loss Coverage premium services to help protect your investment

Take advantage of Zero Downtime I never understand what they mean with this, but now it definitely lost all its meaning.

Zero Downtime is a paying option provided by PACE (makers of the iLok and the copy protection software) for extended support if one loses their licenses. It covers a broken/lost/stolen iLok, not sure how it would fit in this case.

A few years back I was working for a company that used iLok protection and had a major outage like this, we nicknamed it the iLokalypse.

From what I recall zero downtime simply meant if a ilok went missing then PACE would assist you in replacing license(s) to a replacement ilok from their end. Trouble is often times you couldn’t quickly find a replacement ilok without a guitar center nearby and would still have to order it through avid store and wait for it to arrive if you didn’t plan for this.

ZDT also covers RMA situations where you iLok broke. Instead of them sending you a new iLok after a couple of weeks that has whatever licenses they were able to recover off the old one, they send you a new one with all of the licenses from the old one the same day you send them the old one, so you end up with like a 24-48 hour turnaround instead of losing a week or two.

Ah yes, the excellent service of paying additional money to be able to actually use the software you've paid for...

rule 1 of hollywood tech support:

Always take what the editor says with a massive pinch of salt.

I spent ten years in the movie industry, and the people that were the most forthright, and almost universally wrong about technical things were editors.

"The monitor is out of grade" No, you've set the background lights to red.

"The sound is out of sync" No, you've turned off the amp and somehow managed to get the crappy test tv to play sound

"the hardrive is corrupt" no, its clearly asking you to update OSX

my personal favorite:

"my station is slow" No, what you appear to be doing is copying 135 TBs of data from one share to another.

Any non tech savvy would provide or will be treated by grain of salt. But, I have several reports from colleagues.

oh indeed, its a pinch of salt, not ignore entirely :)

What you're describing sounds like industry-specific anecdotes from a very common, cross-industry reality about non technical users.

I'd revise that to the far more irritating segment of users who think they're more technical than they really are. Actual non technical users are fine, because they generally know when they don't understand a thing. The thinks-their-technical users will bravely throw in whatever phrases they've heard mentioned in relation to a completely unrelated problem.

Amen.

What, will you refuse to help me if I call you in deep distress and said that "xcode command line tools is failing to update because of the nginx 503 errors clearly visible in Safari"

Yes

well, you're not wrong

I wouldn't absolutely refuse to help, but if you are like some users I've dealt with and you keep repeating that instead of answering specific questions I've actually asked† then I'll drop the call and mark the support ticket as paused‡ until the information requested is provided.

[†] Q: can you check A, User: why if the problem is Z? Q: could you do B and tell me what the response is?, User: but I'm sure the problem is Z, Q: what version of C do you have?, User: that shouldn't' matter because the problem is Z, ..."

[‡] meaning it is not running against any SLA clocks

The two matters could be separate symptoms of the same underlying problem for which more information is needed. Admittedly in this case you would probably be right and the tool is likely failing to update because it is getting the same 503:Overloaded responses from the http server, but that is not necessarily the case (perhaps the server is agent sniffing and provides a different, more human interpretable, response to requests from browsers, the response for humans is proxied by this server from another and that is currently overloaded, so the update tool isn't getting the same 503 responses that you are).

It's definitely not unique to Hollywood... Before I switched to infrastructure development I ran a Linux engineering group and among other things handled production support escalations. The job was often finding the actual problem while gently dissuading my customer the issue wasn't a pet theory and/or something they just Googled. In fact sometimes the escalations came specifically so I could weigh in on root cause after joining the bridge because the various parties couldn't agree.

A favorite was someone to run free and declare the system was "out of memory and must be swapping". It recurred so often I think that theory passed from group to group like a treasured heirloom until we wrote a FAQ on the subject. Fortunately in my opinion getting ops and development to work together built mutual respect for respective skill sets.

That's the super-user syndrome. They know enough to get themselves into a bind but not enough to get themselves out. It's difficult just trying to get an accurate description of the problem because they lace the description with theories about what the cause is.

"Please just describe the problem and, if possible, how you reproduce it. No, just describe the problem. What's happening when you press the button? No, just the problem! Not what you think is causing it."

It's a conundrum because good super-users can help less technical people in their groups. But, their issues tend to take more time to unravel.

I imagine this is especially prevalent for editors because they are expected to be the technical ones in a room of high-ego professionals. Every second of their day is them being judged on how super-user they are.

I tend to associate "super-user" with an admin who knows what they are doing. Back in the 90's we called these folks "power users".

Yup. Power user was the name we used back then. If you had a good one, (s)he could save you support time.

I don't understand this.

> The thinks-their-technical users

English was not my first language and sometimes I get tripped up. What does this mean?

It means users who believe themselves to be highly technical, but are not.

It means someone who believes they have a high technical ability but in reality has a low to mid ability and know some relevant words. Typically they apply them incorrectly when describing a problem or solution.

I believe it’s misspelled, but meant to be “think-they’re-technical” as in, describing users who think they are technical.

I think parent meant "Thinks-they're-technical users", as in the users in question believe themselves to have more knowledge about the domain than they actually do.

That looks like a misspelling that should have been "thinks-they're-technical users" i.e. users who think they're technical but aren't as technical as they think they are.

It was supposed to be "thinks they're technical". In other words, people who think they are more technical than they really are.

Should have been “thinks-they’re-technical”

They think they're technical users, but really they aren't. Generally these folks are people who have skills in other areas and have a partial or basic ability to use some complicated software. However, they don't have well-developed computer skills but are convinced that they do.

They meant "thinks-they're-technical", which might clear things up.

Putting dashes between a bunch of words isn't really a correct way to do this, but it's a way to smash a bunch of words together and pretend it's one word. In this case an adjective describing the users.

It's perfectly cromulent to hyphenate words this way. Doing so causes them to become adjectives. For instance, you would say "fast-paced agenda," not "fast paced agenda," or "fast, paced agenda." The latter two have a different—and in this example, weird—meaning.

In this case, I think it's a poor style choice, because there are two verbs and an adjective being combined, but there's nothing 'incorrect' about doing so. It might be clearer to say "users who think they are technical," but fixing the mixed pluralization and incorrect use of "their" would go a long way on their own.

What you did there, it has been seen. :)

What? Hyphenated compound adjectives have been a standard feature of English for centuries:

https://www.grammarbook.com/punctuation/hyphens.asp

I suspect the trip-up was caused by mixing up their and they're.

http://3e.org/xy.png

On two occasions I have been asked, — "Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?" In one case a member of the Upper, and in the other a member of the Lower, House put this question. I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.

- Charles Babbage - Passages from the Life of a Philosopher (1864), ch. 5 "Difference Engine No. 1"

If you want to use incorrect data to get correct answers, you need a human. That's one of the fundamental differences between humans and computers.

Floating point numbers are trying hard to rectify this ;-)

> If you want to use incorrect data to get correct answers, you need a human.

More often, that just gets answers that are incorrect in ways other than (sometimes, in addition to, othertimes, instead of) those explicable by the incorrect data.

"I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question" is my all-time favorite way to say "What is wrong with this clueless dipshit."

The question makes sense rhetorically, to establish that the machine is still only as good as its operator, just faster.

On slashdot, someone proposed the theory that this is a good question to ask if you have rational worldview, and know this kind of thing should be possible, but aren't sure if you're dealing with a scam artist who will tell you what you want to hear.

Yes, I've used that principle before; ask a question you know should be "no" and see what happens. Can that car tow this (way too heavy) thing? Will my dishwasher clean my dishes even though my water is full of sulpher? Will this (integrated graphics) computer play $LATEST_GAME at full settings? Will this plastic thing last ten years of solid use outdoors?

In my personal experience, I actually don't get the full-throated lie "Yes!" all that often, if ever, but I get a lot of uncomfortable waffling.

Indeed. In buying a new car some years back, I found a reliable way to translate answers. When the dealer answered, "Yes" it was generally truthful. But when they answered, "I think so", the actual answer was definitively "No".

The 'light' variant of this question makes a lot of sence, though:

The askers were non-technical but very jaded in human nature, so they know the human-provided input might be slightly imperfect but not completely wrong.

Then the question is: Can a machine based on perfection cope? Will it output reasonable but slightly-wrong conclusions? Will it go off the rails and produce completely insane errors?

Answering this question requires a lot of interesting subtility, which has been completely rejected by babbage.

Working with support issues for at lest 20 years...

My take on this is that there are two types of non-technical users:

- stupid ones

- intelligent one

The intelligent ones know what they don't know. Or at least they understand what is their area of expertise. They focus on explanation of the problem and they do not make random conclusions.

The stupid ones do not know what they do not know and, of course, they think they are smart and know everything. It is so hard to work with them.

It is interesting that education level, nationality, race, etc. does not give any reasonable indication whether a non-technical users is intelligent or not.

That's generally true, but there's a more specific affliction you sometimes run into that the parent comment is referring to.

There are some single software programs so big and complex that you can build entire careers out of them. People who do that become wizards within that software suite, and then incorrectly generalize that to believing they know everything about computers. When really, they are Photoshop jockeys, Avid editors, Maya artists, etc. and they are mostly lost outside of those applications.

If you haven't worked in creative fields like video you may not have run into them, but they are definitely a breed.

(Developers fall prey to this sometimes too, but our job tends to require us to use a variety of tools and the underlying operating system itself, so we do end up with a greater breadth of experience.)

With the exception of the OSX update, most of these sound like technical issues they're just not bothered to fix themselves. My understanding is that editors don't usually set up their own workstations so why would they troubleshoot them?

mostly its down to them fiddling.

editors _fiddle_ but then don't take responsibility for what they've done.

The monitor issue its a two-for-one. The editor was doing a technical edit, that is matching some shots up to someone else's spec. They were proxies(as in small previews to make the machine run faster) and were not graded.

What they had done is fiddle with the room lights, because they wanted to make the room feel better. Obviously this made the monitor look more blue.

My college (who is a junior, in a junior department) gets an angry phone call because the workstation isn't up to par.

Its a lot of dick waving for no real gain.

Another fun one was in about 2009 an editor decided for a quick preview edit that it would be quicker and easier to laser out a digital shot to film, cut and splice by hand on a stienbeck, and then re-scan it in.

We had all the shots loaded on the workstation.

Not all editors are like this. but there are enough to give them a bad name.

Yeah but when they are all having the same problem that means it's a problem. I've been on both sides of that situation, generic dismissals like this are not useful or helpful.

You are moving the goalposts. The topic here is that people misjudge a problem but believe they are correct.

Kinda hard to be moving the goalposts when there was no prior comment of mine to move them from. I'm just disagreeing. When everyone is suddenly having the same problem the probability that they are all coincidentally falling victim to the same misjudgement asymptotically approaches zero.

What was claimed in your parent comment was that people do all sorts of idiosyncratic things with their machines but were convinced the problem was elsewhere. No?

You seem to have mistaken me for a man who likes to repeat himself.

Addendum:

I used to support a large range of artists and technical staff. Every class of worker would have their own proclivities.

Compositors, at the slightest hint of any technical issue would immediately start thinking up workarounds. By the time we'd get there to see whats gone wrong, they'd be telling us there's a problem with the workaround, and not tell us the original problem.

Modellers, normally maya has crashed because they are really pushing it, they would have forgotten to publish for a while.

producers, love busting out to spreadsheets, sometimes it doesn't marry up to shotgun/ftrack, or the macro is broken because someone has changed the naming convention.

pipeline, Always trying to sneak the next new thing into production by the back door.

ha, I'm in the same boat. Spent 8+ years doing engineering for live/post-production television. And the younger they got, the less technical they were.

Someone on Reddit posted some more details.

Apparently it's iLok removing the /var directory.

https://www.reddit.com/r/apple/comments/d8nxhk/mysterious_av...

It now looks as though it might be Google Chrome. Whoooops.

https://mrmacintosh.com/google-chrome-keystone-is-modifying-...

e: hn discussion here: https://news.ycombinator.com/item?id=21064663

I can't wait for the post mortem, I wasn't expecting anything with Avid/PACE but if it's Google's problem then maybe we'll get info.

I work in this space (and am a casual user in the space) and I really want to know why /private/var was getting unlinked through a normal update and why Avid got pinged for it.

> why Avid got pinged for it

I think that was just because it was Avid users discovered it on the FB group[1].

I thought it was cool of their CEO to talk to the community, despite no evidence of it being their fault.

[1] https://www.facebook.com/groups/mediacomposer/permalink/1015...

Sounds unlikely to be a virus. Certificate expiry seems probable if it’s happened suddenly to everyone using this software and dongle.

Whatever it is, it's changing operating system settings, so not just certificate expiry:

https://twitter.com/MarcusPun/status/1176369954255331328/pho...

huh .. to the point of changing what groups a user is in (or corrupting the local user info).

This site says Chrome Updater is the cause of the problem: https://mrmacintosh.com/google-chrome-keystone-is-modifying-...

It only happens on Macs with SIP disabled, which includes a lot of Mac Pros with Avid (for third-party video card support)

Sound like someone messed up a removal routine, equivalent to "rm $base_dir/$temp_dir -rf".

Yeah, that’s exactly what the behaviour sounded like to me too, having inadvertently done this to myself a few times over the years, and given that many autoupdaters run with root permission, it’s perfectly plausible.

Yeah, I think we had that twice: 1. "I can't work if all the NFS mounts empty" 2. "You're aware that closing the GUI empties /home/$(whoami)? Related: Did you see our sysadmin?". Backups are awesome.

Not possible with SIP. The OS files are read only. Not even root can change them.

Someone else in thread mentions an kext for the iLok dongle AVID uses, so are kexts covered by SIP? Or are they counted as part of the kernel, and thus all powerful?

Looks like some situations AVID requires SIP disabled for external video cards?

So dumb. Why do companies pull this crap instead of fixing their issues with the vendor instead of leaving their users vulnerable

Sounds like the case when EvE Online removed some Windows boot files: https://www.eveonline.com/article/about-the-boot.ini-issue

Ah bless, I recall that (played at the time), I was fortunate to of had separate drives for programs and temp/swap to the OS and didn't have eve installed upon th C: drive, but knew many who didn't. That said, it was somebody who did the same kind of thing that tested the update and equally had no issue and it rolled out. Though fair play, they put their hands up fast and if needed, paid for peoples to get a tech to fix the issue for users who just gamed and unable to handle the minutiae of fixing themselves.

Sounds like another 'DRM so strong you can't even boot anymore!'

DRM so strong they can’t even produce the film.

Oh man I’d hate to be on the receiving end of this shit storm. There’s a film called swimming with sharks [1] that perfectly illustrates the wrath of producers

[1] https://en.wikipedia.org/wiki/Swimming_with_Sharks

Did anyone else get a drive-by download from synchroscript.deliveryengine.adswizz.com?

Yes, what was that?

I have no idea. It was a 0 byte afp.php file. You only really get those when a webserver isn't setup to process the PHP files and ends up sending it out like a download.

Is this really an issue localised to LA? I've worked for companies equipped with Avid suites that were (at least in geographic terms) as far away as you can get from LA.

Well, looks like Google has some splaining to do in this incident.

One Twitter thread I followed said something was trashing /var.

Its unlinking /var from /private/var and laying down a new incomplete /var

Isn’t this just the Google Updater issue that removes the /var link on Macs that have SIP disabled? Perhaps that is required for Avid?

Hm, I'm not using any AVID software, but I had to re-install from Time Machine after trying the 10.14.6 update on my Mac Mini 2018 last week.

I switched to Adobe Premiere under Windows 10 with NVIDA 2080 for accelerated rendering and have been very satisfied.

It's really warming to me that Avid and Premiere, two applications I learned video editing on at school when I was 10, are still big things today.

Why people still use Avid?

Because it's the industry standard when you need to work with more than one person. Most other offerings don't scale with team size/production pipeline like Pro Tools/Media Composer.

And there's a certain class of professional who resists change/upgrades at every turn, specifically because of stuff like this where updates would typically break everything.

I don't know about the video production side, but Pro Tools is fantastic software.

How is this possible with SIP in modern MacOS?

Did they turn SIP off?

SIP is a whitelist, not a blacklist. It's possible they (Apple) missed something.

The use of 'trashcan' to refer to the MacPro by multiple users in the link is pretty amusing.

> Has anyone had their trashcans go down today out of nowhere?

We have a bunch of Mac build servers here. They are also called trashcans by everyone that works with them.

Literally everyone in television and film production calls them trashcans. I've seen them called that in contracts even.

They threw the kitchen sink at the problem but just ended up buying cheese graters!

It's a level up, now it looks like a tool for processing ingredients rather then collecting the junk :)

Apple Cheese Grater CNC and performance testing on Pecorino:

https://m.youtube.com/watch?v=s29YZqe9Cso

This is exactly what I was looking for in this thread. Of course it exists.

That must really be annoying someone over at apple. Spend all that time design something beautiful. And... Everyone calls it garbage

Surely the implication is that what's inside is garbage. :P

(which, these days is mostly true, a single socket CPU on a workstation of that grade is a poor design choice)

Personally, I actually like the design of the current Mac Pro, but I can understand why many wouldn't.

Surely the implication is that what's inside is garbage. :P

Not at all. Even people that love them refer to them as trashcans. It's just a cute nickname, since Apple doesn't give their machines any other distinguishable names.

We also call them "trashcans" in conversation. It's an identifiable and memorable term, rather than "mac pro (original/2006)", "mac pro (cylinder/2013)" or "mac pro (cheese grater/2019)"

Cheese grater will definitely stick just like trashcan. Both highly utilitarian tools, mind you.

Am I old? Everyone's talking about whether this new 'cheese grater' name will take off. But that's what we called the old aluminum case since the G5.

Didn't the original Mac Pro look like a cheese grater, too?

It had a mesh in the pre trash can version but the new one looks far more like a grater with spaced holes and an indented piece of metal that looks like the blade.

https://icdn2.digitaltrends.com/image/digitaltrends/mac-pro-...

https://blog.macsales.com/wp-content/uploads/2012/05/prod_ap...

Sure, but for more than a decade the original aluminum tower was called the cheese grater.

Beautiful? There's a reason why it got the name, it looks like a trashcan. And the old mac pro is filled with garbage(how fitting, right?). If it was any good then apple would have stuck to it, but it was garbage, so they didn't.

Now we have the new mac pro, and it's a big step up from the trashcan.

Yes, and lots of very smart designers have put lots of time over many decades into making trashcans look good.

The apple trashcan certainly looks vastly better than other commercially available workstations.

> The apple trashcan certainly looks vastly better than other commercially available workstations.

Not really, when given the tendency of them to roll around when being transported in production environments, they tend to end up bolted within wooden boxes with little wheels on the bottom.

There's a reason Apple listened and put wheels on the new one.

To be fair, the machine's tendency to roll around during transport is irrelevant to the claim that it looks better than its competitors.

I meant more that it doesn't look better if it's inside a plywood box.

> The apple trashcan certainly looks vastly better than other commercially available *trashcans.

I don’t think anyone who’s ever shopped for designer trashcans would agree with this :P

It's a shame really, I feel like we've lost something from the years of SGI workstations that looked like they came straight from the future.

You mean bright LED's and transformer-like mods?

I don't think so, seems like ALL gaming PC's are like that.

SGI workstations had hardly any lights. Their essential distinguishing feature where that they sometimes had slightly rounded corners and came in 'fun' colors. Kind of like this

https://en.wikipedia.org/wiki/SGI_Tezro#/media/File:Silicone...

https://en.wikipedia.org/wiki/SGI_Octane#/media/File:SgiOcta...

https://en.wikipedia.org/wiki/SGI_O2#/media/File:Silicon_Gra...

The Tezro and Octane both have prominent power-on lights, but I think they're tasteful and cool.

I mean... they definitely look like they're from the 90's future.

Those unfortunately now just remind me of the Hot Wheels PC https://www.reddit.com/r/pcmasterrace/comments/azdp7i/hot_wh...

I'd rather see more like the SGI Indy workstation: https://en.wikipedia.org/wiki/SGI_Indy#/media/File:Silicon_G...

I think Sony's got your back there. Looks like a source of inspiration for the PS4

No, he's talking about classic SGI machines like the Tezro:

https://d2w9rnfcy7mm78.cloudfront.net/746965/original_3c751a...

The old SGI joke I liked was any system that required less than 4 people to carry was considered a desktop. The future must have very large desks for their users.

Well I can remember when fitting a new dual floppy drive (8 inch) to a pdp11 require two people to install it and a third to help run the cables through the chassis.

From trashcan to cheese grater.

You mean back to cheese grater. The cheese grater name was applied to the previous generations before the trash can.

I don't think Apple has said so publicly, but it seems pretty obvious that they knew about that nickname and decided to lean into it with the new model. The new one really looks like a cheese grater, even more than the towers before the trash can.

Eh, it was designed for a different system configuration. Apple was betting on multi-GPU architectures when they designed that thing in 2012/2013, but then tech didn't really follow that path.

The CPUs and multiple GPUs put out (relatively) equal heat, so they could make the single, big cooler work. But when the world went back to a single, big, hot GPU, it pushed things _just_ out of whack enough that it wasn't worthwhile to continue down that path.

I think that is a little too generous.

It was a poor design because it didn't take into account the thermals of the hardware it was sold containing. It throttled its graphics cards from the start but also got worse as dust/age set in. It couldn't be upgraded or fixed because they chose appearance over practicality.

It was, is, and will always be the "garbage can." It has nothing to do with the industry's direction, it was poor as-sold.

As far as I can tell, everyone calls it the trashcan, even inside the company.

Smaller Sun boxes where / are referred to a pizza boxs

Have never called them anything else even in Russian. It's truly an iconic design.

>Film and TV editors across Los Angeles were sweating Monday evening as their workstations were refusing to reboot

The irony of something refusing to reboot in Hollywood. Hopefully the production of Star Wars 12, Rambo 8 and Terminator 9 has not been too disrupted.

The rebellion has got to start from somewhere. Humans are still consuming mindless reboots, but the machines have decided it's too much and have given up.

Whats with this kneejerk advice?

Honestly who turns off their macs? Are they talking about sleep mode?

I had an issue recently with Virtualbox which caused my iMac Pro to crash and power cycle. That, as well as certain Xcode updates (required by other software), meant I lost my 200+ days of workstation uptime.

> meant I lost my 200+ days of workstation uptime.

So nothing was lost except a world record?

Do you have any idea how exactly did VirtualBox crash your iMac Pro? I’m pondering installing it on mine and your comment is scary.

i, as an app developer, have to reboot my mac atleast two times the week to get xcode back into a working state. not to sound like a jerk, but if you dont have to powercycle you prob not pushing your machine far enough

Or running stable software that locks up less? Maybe Xcode is just badly written?

any sufficiently complicated software can reach a state where problems arise. you just need to push it far enough.

e.g. i have the swift compiler giving up every few weeks because i feed it code that crashes its internal constrain-solver. good thing is, submitting such code samples i can contribute to its improvement. i believe everyone has a tale of some software exhibiting strange behaviour, ive seen it all: adobe products crashing when you exit them, broken text rendering in firefox until a restart, hell, i even had the windows' LogonUI.exe hang on me.

the point is: consider a software to be a state machine. since we stopped formally verifying correctness of our programs, it is in the realm of possible for each program to reach a state where its behaviour is undesired and there is no easy path to move it to a more stable state.

Don't you have to still reboot Macs to apply certain updates?

You do. OP sounds like a typical end user. Waits a year to reboot and then gets mad that updates take an hour to download and install.

It used to be you almost always had to reboot to update the Safari web browser, but interestingly this week's version 13, at least on macOS Mojave, did not require a restart.

I remember back in the day when every "creative type" doing anything on a Mac would reboot before starting a new big Photoshop or Illustrator project.

This was before Macs had memory protection and one process could stomp on another processes' memory if there was a wild pointer. After running any large application for a while, before starting a different one, it was standard practice to reboot.

I do after I notice it starting to get really slow, but I always check uptime in terminal before I do. Just happened last night, after 62 days of running!

I reboot once every week or two - there's always something that fucks up when you leave it running for a month or more. Or if I ever go through and quit everything (for whatever reason), I usually reboot then. It won't hurt, and it doesn't take long.

It's worth doing fairly regularly to at least check the process still works as you expect. At some point you'll actually have to restart your PC, and the last thing you want is to be surprised by some problem or other. (OK, so I've never had any serious rebooting problem with OS X, but I have had a reboot show up stuff missing from launchctl and the login items, the sort of thing that's an outsize pain to fix when you were in the middle of something.)

Reply on news.ycombinator.com