|
crazysim posted:How about using a trial version of Windows and Virtualbox USB passthrough? Guess I should have quoted BigHandsVince for better context, since I got two people with that one
|
#
¿
Feb 2, 2014 08:14
|
|
|
|
| # ¿ May 9, 2024 04:20 |
|
|
door Door door posted:So I've got a question about a weird situation I just found myself in. My macbook pro was suffering from the "no goo boot image" four beep failure, so I flushed the NVRAM and it starts fine now. But I also have a guest account in addition to my account, and it didn't fail on startup, it failed when I tried to log into my account. I could log into the guest account no problem. Shouldn't the only thing in NVRAM be whatever image is necessary to get you to the login screen? Why would flushing NVRAM fix a problem I was having after getting to the login screen? Am I misunderstanding how OSX handles boot images? Also the system is encrypted with file vault 2 if that matters. Guest accounts are different with FileVault 2 enabled. As chimz explained, a real user's password must be entered into the EFI based login screen in order to decrypt and boot your main partition. The guest account has no password, so it can't do this. Apple's workaround is that with FV2 on, logging in as guest actually boots the recovery partition (which isn't encrypted) in a special guest account mode. Guest accounts can normally do a significant number of things, but because it's the recovery partition and nothing on your main partition is available, the FV2 guest account is restricted to just Safari. (And silently phoning home if you have iCloud and Find My Mac set up.)
|
|
#
¿
Feb 2, 2014 12:25
|
|
|
Xabi posted:My friend's got a 2011 MacBook Pro that he's putting a new SSD disk into. His plan is to use some sort of Icy Box to clone the old disk, before putting the new disk into the laptop. Will he also need the original CD to install MacOS or will the cloning sort all that out? The reason I'm asking is that he can't find the old CD. Also, is there anything else he needs to remember before he starts this whole project? Tell him to not worry about finding the original disk. If it's a 2011 model MBP (or even some/all 2010s), he can upgrade the firmware to get Internet Recovery (chances are that he has already done this since I believe it should have come up as an automatic update). He can also use the App Store to download Mavericks and make a USB stick from the Mavericks install app (Terminal-based instructions here, easy to use GUI tool here). What I'd do: download Mavericks, make a USB installer stick, boot from it, select Disk Utility from the menu instead of installing, use Disk Utility's restore function to clone the old drive to the new, shut the computer down, swap the drives, done. And yeah, he doesn't need to actually perform an OS install. OS X is easily cloneable so long as the destination disk has enough space. There are a couple reasons why I'm suggesting booting from a USB installer to perform the clone, though. One is that cloning the drive you're presently booted from is a bit iffy, since the running OS could change files after they've been copied. The other is that if anything goes wrong, a USB installer stick is a very useful tool to have.
|
|
#
¿
Feb 4, 2014 06:20
|
|
|
If he's got too much stuff to fit on the SSD he's going to have to prune some files no matter what he does, but otherwise that's a good plan.
|
|
#
¿
Feb 4, 2014 10:53
|
|
|
Xabi posted:Yeah, that's been taken care of. We thought that would be enough to clone the stuff, but apparently it's the size of the disk you want to clone that matters, not how much of the space you've used up. Oh, that. I forgot to tell you an important detail. Unfortunately Apple provides no hints about this in the user interface. If you use Disk Utility's restore function on whole HDDs, it does a blind copy of every block (partition table and all), in which case the destination must be at least as large as the source. But if you specify a partition as the source instead, it can detect the filesystem and resize it as it copies. This lets it copy to a smaller destination as long as there's enough space for the files. So, you'd want to partition the destination drive first, then do a partition clone. But now I realize there's a flaw with that too. Before Lion, this worked perfectly, but as of Lion Apple has begun installing a second hidden Recovery partition. If you clone just the main partition the computer will boot and work fine, but you won't be able to boot into the Recovery HD and you won't be able to turn FileVault 2 on (it requires a recovery partition). If your friend doesn't care about FV2 and doesn't mind dedicating a USB key to the installer (which does everything the Recovery partition can), a clone will work. Otherwise just go ahead and perform an OS install and use Migration Assistant. It'll get him to the same place in the end, just a bit slower (the Disk Utility clone method is faster because it's a block copy, even when resizing a partition).
|
|
#
¿
Feb 4, 2014 11:53
|
|
|
The potential problem there is RAM. 8GB isn't a lot to run two VMs plus the host OS X. Keep in mind that the new Retina machines have Iris/Iris Pro integrated graphics, which eats up to 1GB on its own. I think that for the 13" 16GB RAM is only available as configure-to-order, might not be easy to get that through Amazon. edit: I have never bought AppleCare and have never been burned by not doing so, but many swear by it. You can delay buying it till the original warranty is about to run out. BobHoward fucked around with this message at 20:47 on Feb 4, 2014 |
|
#
¿
Feb 4, 2014 20:40
|
|
|
tractor fanatic posted:is this lag due to differences in settings, or just like the latency from wifi? Almost certainly the latter, I don't think there's any settings to screw this up. If you don't have a way to connect wired ethernet to the Mac, you can try loopback ("ssh localhost") to see what SSH is like without any physical network latency. Unless there's something badly wrong, it should feel exactly like a normal terminal window.
|
|
#
¿
Feb 5, 2014 13:32
|
|
|
Ashex posted:Edit: Nevermind, running repair permissions seems to have solved it... Unless the log actually showed it fixing something relevant to VPN, repair permissions is unlikely to have been the real fix. If you rebooted, that was probably what did it. I've seen this many times before (starting in 10.6, which did it a lot), and a reboot always fixed it. The VPN service sometimes gets in a bad state where it won't connect, and rebooting "fixes" it by reloading it from scratch. Alternatively you can reload it from the command line, which may fix the issue without a full reboot: sudo launchctl stop com.apple.racoon sudo launchctl start com.apple.racoon
|
|
#
¿
Feb 8, 2014 21:17
|
|
|
wolffenstein posted:The white paper you linked indicates it may be possible to recreate the encryption key using a user account password and regenerating the keys on the affected hardware (meaning I'd mail my Mac mini and its drive). The paper mentions there may be static values used for generating encryption keys for device compatibility reasons, so if that's still true for 10.8.5 (what was installed at time of wiping), then it may even work on the first try. I read through the paper and I'm afraid that you're screwed. What I take away from it is that the important encryption keys are not predictable at all (if they were, FV2 security would be essentially useless). They're always generated by a pseudorandom number generator (PRNG), and the paper's analysis establishes that Apple's PRNG and seeding are strong enough as a whole to make brute force attacks futile. The static value the paper mentions is part of the cryptograhic hash algorithm Apple uses to transform a passphrase (a user account password or a recovery key) into an encryption key. (More on that algorithm below.) In FV2, these passphrase-derived keys are used to decrypt a key encrypting key (KEK), which is then used to decrypt the volume master key (VMK), which is (finally!) the crypto key that actually unlocks the volume. A small area at the start of the disk stores multiple passphrase-encrypted copies of the KEK (one per passphrase), and one encrypted copy of the VMK (encrypted with the KEK). The VMK, KEK, and salt for the passphrase hashes are all randomly generated and should never be the same across any two FV2 volumes, even if you make them on the same computer using the same OS. The remote wipe presumably erases the key storage area. If so, it doesn't matter whether you can re-derive the passphrase derived key. There's no encrypted copies of the KEK left to decrypt using that key, and more importantly no encrypted copy of the VMK to decrypt with the KEK. The static value the paper mentions won't help you. Apple is using a cryptographic one-way hash to transform a passphrase into a fixed length key, but instead of running this hash algorithm once they loop its output back through itself 41000 (the "static value") times. This is solely to make passphrase based brute force attacks on an intact FV2 volume harder -- an attacker needs 41,000x as much time to try each guessed password as they would if Apple had used 1 round of hashing. But in your situation an attacker could guess your exact passphrase and it wouldn't help at all. Basically the only hope I can see is if (a) the disk is a SSD and (b) Apple just overwrites the key storage area. In that scenario you have a small chance; if you can figure out how to do a raw dump of the SSD's flash media (not necessarily easy) you may be able to recover the key storage area since SSDs don't erase overwritten data immediately. But if the SSD has already recycled any of those blocks (and note, they often do so if left powered on long enough, without any initiating event on your part!), or if Apple uses Secure Erase, or if Apple even does a custom firmware extension to let them "secure erase" individual blocks (wouldn't put it past them to request that as a feature for Apple branded SSDs), you're out of luck. BobHoward fucked around with this message at 14:18 on Feb 10, 2014 |
|
#
¿
Feb 10, 2014 14:13
|
|
|
wolffenstein posted:I read it at 2 am tired as poo poo, and well that's what I get. So essentially option 1 is out. I don't suppose you have any idea on how to do option 2? Afraid not, I've never looked into brute force crackers for that kind of thing. The paper's analysis of Apple's PRNG suggests it would be futile to attempt anyways. They say Apple is seeding with at least 320 bits of entropy (potentially a lot more if it's not the first boot or the computer has been up a long time before creating the encrypted volume), and that ought to be enough to put a directed brute force search for possible keys way out of reach. I'm amused at all the self righteousness you've attracted here and in e/n. Depression can hit anyone. Sounds like you're doing better, and please continue irritating people who try to shame you for not treating it like a taboo subject. (Seriously guys, he mentioned it briefly and in passing in his initial post and it gets you that bent out of shape? And what's with this amateur "OMG formerly suicidal guy is too attached to his old life!!!" psychoanalysis?)
|
|
#
¿
Feb 11, 2014 00:44
|
|
|
It does seem like a memory leak. Exporting video / transcoding / etc. should be a streaming process which needs very little memory regardless of the size or resolution of the video.
|
|
#
¿
Mar 1, 2014 21:36
|
|
|
Pivo posted:Yeah there are obviously ways to do it I'm just saying Terminal.app doesn't. You know that whole thing Apple started doing in Lion where they're trying to make it so all apps can restore windows on relaunch? Terminal.app is no exception. It saves the scrollback buffer and the current directory, but not environment variables, so it's not good enough for everything. But the working directory is all tractor fanatic was asking for. Unfortunately, unlike Safari etc., they haven't plumbed this functionality to a menu item which reopens the last closed tab.
|
|
#
¿
Mar 2, 2014 21:26
|
|
|
kuskus posted:Thanks. I wish re-pairing worked. We've tried with 3 wireless Magic mice- no scrolling or gesture settings are ever present or even optional under the typical settings panes. Might have to just get down to bare Mavericks again and just transfer software and files but not user prefs. I'd love to discover the specific trouble files in question. Phone posting from memory so I might not have the paths exactly right, but under both /Library and /Users/username/Library there should be a dir named "Preference Panes". These are where third party prefpanes are normally installed. If the old SL pref pane somehow got copied to one of these locations, Migration would've copied it over, and maybe it's overriding the system's prefpane (which is under /System/Library/Preference Panes).
|
|
#
¿
Mar 4, 2014 00:35
|
|
|
Caged posted:I might be talking out of my rear end but isn't there something with sector alignment that happens when you clean install that might not if you image? Only on older versions of Windows, which used to default to partitions not aligned with 4K sectors. OS X partitioning was 4K aligned by default long before SSDs were a thing and should not suffer from that problem.
|
|
#
¿
Mar 4, 2014 03:00
|
|
|
lua posted:10.9.2 broke this on my 2010 Macbook. In 10.9.1, tapping power did send it to sleep, now it does nothing. It's me, I reported the new Mavericks power button behavior as a bug (because it's really easy to tap that key by mistake on many rMBP/Air keyboards when you're trying to hit delete). Someone at Apple listened. It was already filtering out extremely short taps, but in 10.9.2 they extended the duration a bit, so now you need to hold it down for maybe half a second to get the computer to sleep. I still don't really like it, to be honest. If I want to sleep the computer I just close the drat lid. Far as I'm concerned, immediate sleep on button push is worthless and doomed to never do anything but annoy me. I'm honestly curious if anyone loves this immediate-sleep feature, and whether the extra delay ruins it for you. Hope I didn't help screw you guys over, if you exist.
|
|
#
¿
Mar 5, 2014 03:49
|
|
|
Tapedump posted:Next question: Would having only 1 GB of RAM make a newer OS X version like Mavericks take hours and hours to install fresh? Put in some RAM first. Mavericks requires a minimum of 2GB and I believe the installer enforces it.
|
|
#
¿
Mar 5, 2014 03:55
|
|
|
slogula posted:Does anybody have any opinions on the current state of/ options for zfs under OSX? I understand ZEVO is no longer under development, and I'm concerned about MacZFS's stability. I've been relying on journaled hfs+ atop two-member raid1's for data intake, and I'd like to make use of the data integrity and raid handling features that hfs lacks. You are right to be concerned. IMO, relying on developers outside Apple to deliver a high quality ZFS implementation is a bad idea. That's not commentary on their competence, it's just that you probably need to be a member of Apple's kernel team to do the job right. The former Apple guy behind ZEVO (Don Brady I think?) was the best bet to do it as an outsider, but it's been three major Darwin releases and six or seven years since his ZFS port was cut from OS X. Lots of stuff inside the kernel has probably changed, Mac ZFS isn't his day job any more, he can't yell at other kernel engineers sitting in nearby cubes, and he doesn't have access to Radar bug reports these days. (Bug database access is really important for something like this...) This is all worse for ZFS, in that it's a huge and complex code base that's notoriously difficult to port to operating systems which aren't Solaris. I'm sure there are people who have had good luck with one or more of the Mac ZFS implementations, but I wouldn't trust any of them. I've never heard of one which wasn't considered a kernel panic risk, not even the official one while Brady was at Apple -- it was very much work in progress when Apple pulled the plug.
|
|
#
¿
Mar 5, 2014 04:37
|
|
|
Fats posted:I don't get any sleep anymore, it just pops up the Restart/Sleep/Shut Down dialog after a couple seconds. That said, I never used it, anyway. Sorry, I should clarify. I get the dialog after about 2 seconds too. In all versions of Mavericks, the power button can do four different things: 1. Very short press and release -- ignored 2. Slightly longer press and release -- sleeps when you release the button. 3. Hold for ~2 seconds -- shows the restart/sleep/shutdown dialog 4. Hold 5 seconds -- Force shutdown In 10.9.0 and 10.9.1 it was difficult to press and release fast enough to actually notice that case #1 existed. In 10.9.2 they changed the timer and now it takes a fairly long and deliberate press to sleep the machine.
|
|
#
¿
Mar 5, 2014 07:45
|
|
|
smr posted:So, my first Macbook Pro (Retina) and I've been loving it but this happened today: Let me guess, it's the 15 inch model with a discrete GPU? I've been seeing that on mine and it seems to be graphics driver bugs related to GPU switching. I use gfxCardStatus to keep track of which GPU is active. For me it seems to happen when the system decides to stop using the Nvidia GPU and switch back to the Intel one. System logs seem to show a poo poo ton of toggling back and forth between the two GPUs whenever this graphical corruption happens, instead of a single clean transition, so there's clearly something buggy going on in the handoff. I think I've only seen Safari tabs get scrambled, and if that's so for you too I have a workaround - reloading the tab always resolves it for me. P.s. use imgur for images, it's super easy and you can upload without an account BobHoward fucked around with this message at 23:27 on Mar 13, 2014 |
|
#
¿
Mar 13, 2014 23:25
|
|
|
Zenostein posted:Incidentally, why does OSX crash when the GPU fails, when Windows 7 just restarts the driver and then chugs along? Is that just a result of how drivers are used in OSX? At least some of the OS X GPU drivers actually do try to do the restart and chug along thing. It's not as well realized as the Windows 7 feature, though.
|
|
#
¿
Mar 14, 2014 06:41
|
|
|
Fortuitous Bumble posted:I don't know how to tell exactly what it's doing on OS X. I am running a Windows VM with 3GB allocated but usually Activity Monitor and htop say OSX is using like 2-3GB (not counting the VM I guess), but I get random beachballs switching to programs I haven't used in a while so I kind of assumed that at some point the OS goes berserk trying to swap memory for whatever reason. With VMWare, and probably Parallels and VirtualBox too, all memory used by VMs is "wired". Wired memory cannot be swapped or demand paged by OS X. When you run a 3GB VM on an 8GB Mac, effectively you have a 3GB Windows machine and a 5GB Mac. This may be leading to more memory pressure and swapping on the Mac side of things than you'd otherwise expect. Even without VMs being involved, for both 10.8 and 10.9 4GB is the practical minimum for no-swap light use (browsing with a reasonable number of tabs, plus a few lightweight productivity apps). With what amounts to 5GB and (I presume) heavier use patterns, it's not unexpected for 10.8 to swap a bit. The good news is that 10.9 may help. Its memory compression feature noticeably reduced swapping on a 4GB Mac for me.
|
|
#
¿
Mar 14, 2014 23:47
|
|
|
Bummey posted:Yes. It's always happened since I got this, uhhh, two? years ago. What's it do if you select a word and use the contextual menu to define it? One of the menu items should do exactly the same thing as the three finger tap, maybe having the word selected ahead of time will break its obsession with 'function' and 'var'.
|
|
#
¿
Mar 18, 2014 23:40
|
|
|
Bummey posted:I think it is just a browser thing. I rarely use this laptop for anything else so I didn't think to check if it was happening system-wide or just in firefox. So it's just Firefox? The surprising thing is that it even works in a broken way for you. According to the Firefox bug database, three finger tap isn't even implemented in Mac Firefox. And I've just tried it on Firefox here and I get absolutely nothing at all on a triple finger tap, as expected. Other linked bugs also complain about partial and buggy implementation of the control-cmd-D keyboard shortcut to do a dictionary lookup of the selected word. Sounds like Mozilla people have been aware of this for years and years but nobody's ever fixed it.. Unfortunately this is all too typical with Mac Firefox. It suffers a lot as a native Mac app because the whole thing's built on top of Mozilla's XUL crossplatform system, and apparently XUL is super difficult to fully integrate with MacOS, or any other operating system for that matter. EDIT: waitasec, I figured out what must be happening. If there's selected text in nav bar or search boxes, three finger click (on anywhere) gets you a dictionary lookup of that text. You're probably keeping one of those two words in a search box all the time, am I right? This must be some of that Firefox integration awkwardness: if recent versions of FF implement these controls as native Cocoa widgets instead of XUL, they're getting the systemwide three finger tap functionality automatically, but it's not active in the main view because that poo poo is all XUL all the time. BobHoward fucked around with this message at 04:35 on Mar 19, 2014 |
|
#
¿
Mar 19, 2014 04:24
|
|
Aside from the hellish local keychain incident after upgrading to Mavericks, this laptop has never really given me any problems, but it sure has its quirks.
|
Malcolm XML posted:diff them on the console Who looks at diffs without -u??? Also, for reviewing diffs even "diff -u" isn't as good as tkdiff, which you can install through MacPorts or Homebrew.
|
|
#
¿
Mar 24, 2014 21:34
|
|
|
Toe Rag posted:I have an iMac12,1. Lately, it has been crashing. I want to run the Apple Hardware Test. I bought the iMac refurbished, and I believe it shipped with 10.6.6. I am currently running 10.9.2. I am not sure if the refurbed machine did not come with the system disks, but I cannot find them. I have a system disk for MacBookPro5,5, but the iMac12,1 keeps ejecting it (disk is not damaged). Do I need a system disk for an iMac12,1 in order to run the AHT, or can I run any 10.6 disk (the MBP disk is 10.5.7)? Everymac.com shows the MBP5,5 as being 2 model years older than the iMac12,1, so you might as well forget about its system disk. I wouldn't be surprised if the disk-spitting is thanks to an Apple firmware check to prevent even trying to boot off a system disk so much older than the computer. According to Apple's directions, it sounds like any 10.7 or later installation with a recovery partition should be able to run AHT just by pressing and holding 'D' when powering up or restarting. Also, as that iMac is a mid-2011 model, if its firmware has been updated to current it should be able to run AHT via internet connection even without a recovery partition.
|
|
#
¿
Mar 31, 2014 01:36
|
|
|
If you're starting from a fresh ML install, just do it. Mavericks is, more or less, a better ML. And one of the specific ways it is better is on 4GB computers, thanks to the new memory compression feature.
|
|
#
¿
Apr 1, 2014 02:01
|
|
|
Toe Rag posted:Yeah, I had tried holding D while booting about 10 times, but nothing ever happens Make sure to hold down Option-D when trying to run internet AHT quote:I had always assumed this was just bugs related to Mavericks. At first it was just graphical glitches (first when using Maps, then other things), then the glitches + Finder crashing and/or I get logged out, but now the computer is outright crashing and rebooting. The console showed problems with the GPU (makes sense) when it crashed most recently, which makes me think it might be a hardware problem instead and just happened to start when I upgraded. Yeah, you've got a bad GPU. Glitches like those (8x8 blocks of corrupted pixels) are a common symptom, usually related to the video memory. You can sometimes get stuff like that from a driver bug or whatever, but the progression from mere glitches to crashing/rebooting says it's a hardware issue. You've got a component which hasn't failed outright, but isn't performing up to specifications and keeps getting worse as it works its way towards complete failure.
|
|
#
¿
Apr 1, 2014 03:58
|
|
Isn't the recovery partition different from the AHT? I just booted into the recovery disk, and it seems the only things I can do are reinstall, run disk util, or restore from a time machine backup.
|
El Miguel posted:I've got a MBP 3,1 (Core 2 Duo, 2.2 GHz, 4 GB Ram) and I just got a weird pop-up from Apple suggesting I upgrade to 10.9.2. Can I even run 10.9.2 on this thing? And if so, does it run better or worse than 10.6.8, which is what I'm using now? Yes, you can run it. 3,1 is the oldest MBP supported by Mavericks. Better or worse? Personally, I'd do it. There's a few regressions, but overall the OS has moved forwards since 10.6.8. Also, 10.6.8 is probably not getting any further security updates from Apple, so if you're worried about that machine getting hacked it's a good thing to move forwards.
|
|
#
¿
Apr 2, 2014 04:25
|
|
|
JAY ZERO SUM GAME posted:It's a brand new MBP. Might be? Hard to say. RAID enclosures are sometimes a bit funny. You can open Terminal and do "diskutil list" to see the partition structure of all attached disks. What's it show for the RAID, and the other NTFS disk which works? Kingnothing might be on to something.
|
|
#
¿
Apr 5, 2014 01:52
|
|
|
Axiem posted:I recently stumbled upon a tens-of-gigabytes cache of old data that I had sitting around. However, I'm pretty sure most of it is stuff that I actually already have, and this just has a bunch of duplicates because of backups or whatever. Comfortable with running a script from Terminal? Then have this perl thingy I adapted from someone else's dupe-finding script a while back. Paste into a plain text file, "chmod +x" the file, and play around: http://pastebin.com/uvYhus45 You point it at one or more folders. It scans the first folder for duplicates of files found under itself or any of the other folders, and prints what it finds. It can also delete. Once you're sure you've got it finding exactly what you want to delete, add the "-d" switch and it will murderize the dupes for you. It will not ask for confirmation and it does not put things in the trash, so be sure.
|
|
#
¿
Apr 6, 2014 07:54
|
|
|
The White Dragon posted:I just upgraded to 10.9 (from, uh, 10.6.8 Blank what selection? Literally have no idea what feature you're talking about, need a little more context
|
|
#
¿
Apr 13, 2014 12:45
|
|
) and I'm really missing the "blank selection with esc." Is there a substitute for that shortcut in this OS or am I gonna find myself clicking on blank space a whoooooole lot more?
|
Star War Sex Parrot posted:Pixelmator's half off (now priced at $14.99) for the next week to coincide with their 3.2 update. And to compete with Acorn's $14.99 sale price, no doubt. Good time to get your pick of indie photoshop-lite apps.
|
|
#
¿
Apr 17, 2014 23:10
|
|
|
jackpot posted:There's something I'm doing when I'm using just my laptop (no external mouse), where if I'm in a browser and I swipe with two fingers, it's the same as hitting the back button. I'm struggling to think of any loving scenario where I'd actually want this to happen, especially since it always seems to happen when I'm 2/3 of the way through filling out a form. System Preferences -> Trackpad -> More Gestures, uncheck "Swipe between pages" or change it to something other than 2-finger swipe. p.s. this feature is great, don't hate
|
|
#
¿
Apr 29, 2014 01:33
|
|
|
There's a freeware menu item in the App Store called Display Menu which will set 2880x1800. It's probably a decent utility on non-Retina machines, but on Retinas it is poo poo. Doesn't detect all the HiDPI modes correctly, so you need to use the normal Displays preferences to set the retina 1440 or scaled 1680/1920 modes. However, it will detect and set 2880, and you can't beat free if you just want to find out whether you can live with 2880.
|
|
#
¿
May 5, 2014 01:34
|
|
|
BobHoward posted:There's a freeware menu item in the App Store called Display Menu which will set 2880x1800. It's probably a decent utility on non-Retina machines, but on Retinas it is poo poo. Doesn't detect all the HiDPI modes correctly, so you need to use the normal Displays preferences to set the retina 1440 or scaled 1680/1920 modes. However, it will detect and set 2880, and you can't beat free if you just want to find out whether you can live with 2880. Heh. So the developer just released an update which supposedly fixes the retina issues. I say "supposedly" because I can't test it -- he's made full retina support (and scriptability) a $1.99 in-app purchase, and I haven't bought. No reason why it wouldn't work, though, and at $1.99 it's a much cheaper alternative to SwitchResX if you just need something in your menubar to flip between resolutions.
|
|
#
¿
May 6, 2014 09:20
|
|
|
I.W.W. ATTITUDE posted:Yesterday a reboot failed to resolve the issue and I had to verify & repair permissions. They were definitely hosed up and Disk Utility took a while to repair them. My boss has had similar problems with the recycle bin on her iMac in the past. My computer is barely over two weeks old. I've heard that Macs sometimes have issues managing file permissions before. Is this the Achilles' Heel of Mac OS, is my workflow just bad, or one or more of my disks just P's of S? It's more like the Repair Permissions tool in Disk Utility is terrible bullshit which is virtually guaranteed to produce huge reams of false positives. Because it makes so much noise, it convinces people there must be real problems that it's fixing, and they feel a need to do it a lot since it almost always seems to have something to do, and when things resolve themselves for other reasons they'll swear fixing permissions was what fixed it, and they spread the religion on to folks like you. Phone posting so I don't want to go into the gory details of permissions repair and why it doesn't mean as much as you might think right now, but one thing to understand is that it literally doesn't check any of your data files such as these photos you had problems with. It has no idea what their permissions should be, so it assumes you know what you're doing.
|
|
#
¿
May 22, 2014 00:31
|
|
|
I.W.W. ATTITUDE posted:Wow this is all very helpful, thanks. It seems like Ubuntu has the same general approach to permissions as osx and I've never had to mess with them while using it, and I'd expect a commercial os to be even less prone to having such problems. I just see a lot of workaround/issue resolution instructions on various Mac forums that mention fixing permissions. Still phoneposting but Linux familiarity makes it easy to explain. Permissions repair is equivalent to checking that the permissions on all files installed from dpkgs provided by Ubuntu are correct. Except implemented poorly, so often it generates tons of meaningless noise when there's really nothing wrong, and in the Apple world the equivalent of a Linux package manager is used for a much narrower range of installed software, so it covers much less territory than an equivalent Debian tool would.
|
|
#
¿
May 22, 2014 03:39
|
|
|
Doesn't sound normal at all, and searching for "UFR II BackGrounder" confirms it's part of Canon's printer drivers, and that other people seem to have had problems. Is it spamming the main system log at all? Not sure whether system log writes get accounted to the originating process or the log daemon.
|
|
#
¿
May 23, 2014 06:13
|
|
|
Dixie Cretin Seaman posted:Aside from a conspicuously huge log file somewhere (that I don't see), is there something I should be looking for? Also, in Activity Monitor, "launchd" and "kernel_task" processes have even larger amounts of data writes; Is that possibly related? I assumed that was legitimate system stuff.. Don't know anything specific you should look for, I'd just fire up Console, select "all messages" on the left, and try its filter function on likely phrases from things you've found already -- "Canon", "backgrounder", "ufr", and so on. For what it's worth, on one week of uptime, Activity Monitor shows my top 3 writers as kernel_task (3.10GB), launchd (2.53GB), and mds_stores (651.1MB). I assume it's legit, but launchd and especially kernel_task (the Darwin kernel itself) shouldn't just do a lot of I/O on their own. I believe most of the writes attributed to those two processes are really work done on behalf of normal user processes. It's unfortunate that this can obscure things. As for my #3 writer, mds_stores is part of Spotlight, and it's pretty believable that it's written 650MB of updates to the metadata databases since last boot.
|
|
#
¿
May 24, 2014 06:12
|
|
|
|
| # ¿ May 9, 2024 04:20 |
|
|
ambushsabre posted:Well FileVault is proprietary so it's impossible to know the full extent of what's really going on. Nonsense, Apple disclosed quite a lot about how FV2 works in this white paper: http://training.apple.com/pdf/WP_FileVault2.pdf and these guys did some independent digging: http://eprint.iacr.org/2012/374.pdf
|
|
#
¿
May 29, 2014 11:55
|
|