File Sharing Difficulties Frustrate Tiger Admins

rmallico wrote in to mention a story currently running on Eweek about technical difficulties sites running Tiger are experiencing. From the article: "A number of sites running Apple's new 'Tiger' operating system are experiencing problems with SMB file sharing and authentication with Microsoft's Active Directory, Ziff Davis Internet News has learned. Although Apple Computer Inc.'s Tiger increases support for Server Message Block file sharing and Active Directory, several sources say that the Finder fails to log on to Windows and Linux Samba file servers."

Comments

Oh, right, error code -36!

[xiando]

The most interesting thing I noticed in the article was actually that the error message for the Connect to Server failure is "error code -36". A friend of mine who uses Mac OS X has always complained much about how the Mac never tells you anything about what is actually wrong, only gives you a number that is in no way useful for solving the problem. It is amazing this is still the case in Tiger, what in the world would be wrong with giving at least a tiny bit of information or just a hint of what is wrong? Even the good old Windows blue screen is more informative than "error code 4".

Re:Oh, right, error code -36!

[FidelCatsro]

Its actualy very usefull if you have a list of the error codes and what they mean.
http://www.appleerrorcodes.com/

Re:Oh, right, error code -36!

[lachlan76]

Yeah...0x0000005c is so much better!

Re:Oh, right, error code -36!

[KiloByte]

0xffffffdc, you meant?

Re:Oh, right, error code -36!

[lachlan76]

I thought 0x5c was for a segfault? I've seen it in a BSoD before...

Re:Oh, right, error code -36!

[gullevek]

actually you can be happy to get an error code. Because I don't get anything. Either he connects or he just spins and does nothing.

BRRR. Sometimes, this can be really annoying.

Re:Oh, right, error code -36!

[KiloByte]

Ah, we're just talking about different environments. 0xffffffdc = -36, while what you meant is:
0x0000005C: HAL_INITIALIZATION_FAILED

But, if you check it in the MSDN, you'll be really, really upset. The whole documentation is:

Bug Check 0x5C: HAL_INITIALIZATION_FAILED

The HAL_INITIALIZATION_FAILED bug check has a value of 0x0000005C.

This bug check appears very infrequently.

(the entire MSDN article lifted from Microsoft without authorization -- but AFAIK data of this length is ineligible for copyright)

Re:Oh, right, error code -36!

[moonbender]

Error -1, I will never forget you.

Re:Oh, right, error code -36!

[Aphrika]

It is useful until you find that error -36 is written up as:

-36 ioErr I/O error

It'll point you in the right direction I guess, but it's by no means a definitive description of the error.

I must admit that I'm a little baffled as to why Apple don't include better error reporting and descriptions in OSX. It is because they are still assuming these kind of errors will only be seen by techs that know what they mean, or are they still living in a world where they refuse to acknowledge that Macs do throw up the occasional message to the user?

Re:Oh, right, error code -36!

[Anonymous Coward]

IO error can not read or write to the directory . meaning it is not there , the reason for this is Apples implementation of samba on tiger requires the full path

Re:Oh, right, error code -36!

[Anonymous Coward]

A friend of mine who uses Mac OS X has always complained much about how the Mac never tells you anything about what is actually wrong, only gives you a number that is in no way useful for solving the problem.

I've seen this with SMB filesharing, Mail.app, and sometimes Safari. They've all given me frustratingly useless error messages. Anyone frustrated by this should open an Apple Developer Connection account and submit a bug report to Apple's bug tracker. Maybe if enough people do, they'll realize this is a problem. Until then, I noticed that one of the other replies at least mentioned this site that gives some information on these codes.

Here's what I'd like to happen: error messages like "Filesharing error. Please relay these technical details to your system administrator: I tried to log in to 192.168.0.1:139 by sending a SMB_FOO_BAR and it replied with the unexpected SMB_GO_AWAY. See this link for details". They could even have the link contain interoperability information like "you're trying to connect to a Windows ME server, which doesn't work. Sorry." (Hypothetical; I've never tried this. But there's probably some such situation, and knowing it up front would save a lot of hassle.) Or even "you're trying to connect to Windows XP x.y.z; we suggest updating to x.y.z+1 to fix KBxxxx. Should work then." This is the sort of information I can often get by googling, but it's hard when the error messages can have so many different underlying causes. Better error messages and having Apple concentrate on an appropriate page (with the "Did this help?" thing at the bottom) would go a long way.

Other parts of OS X have better error behavior. For example, the crash dialog is excellent. It gives you the options of report, relaunch, and cancel.

If you pick relaunch, it will do so. If it crashes again during startup (by a timer? or before entering the main event loop? I'm not sure), it will give you the option of temporarily starting with fresh preferences.

If you pick report, it will pop up a dialog box with a stack trace in the lower half. You can examine it yourself. If you fill in information in the upper half and hit "Submit", it will send it off to Apple. It also keeps core dumps in a standard place.

Re:Oh, right, error code -36!

[FidelCatsro]

More info can be obtained from console.app in the Utilities directory under Applications(/Applications/utilities , or just go through the system logs from the terminal , but console.app is a rather nice time saver), its just a colection of the systems logs but its rather usefull and searchable .
It does give a more detailed output. for example when i try to connect to my existant SMB share it gives me

May 7 11:32:53 Xcomp kernel[0]: netsmb_dev: loaded
May 7 11:32:53 xcomp[0]: netsmb_dev: loaded
May 7 11:35:39 xcomp[0]: smbfs_aclsflunksniff: user sid S-1-5-21-2466424394-2119469220-2469460652-2002 didnt map

I would have given an example of the error output from the specific problem , but i am doing some work on the linux comp that runs my nfs and samba shares right now .

Re:Oh, right, error code -36!

[JonXP]

Remember, Apple's mantra is 'Users are Idiots'. They simplify everything from the buttons on the mice to the error messages.

Really, it's probably part of their 'Keep the UI as SIMPLE as possible' ideals. If they don't think a standard user will be able to do anything with that information, don't even bother telling them.

Re:Oh, right, error code -36!

[Tim Browse]

If I remember correctly (and I probably don't) the Mac OS error numbers came about because Steve Jobs was fed up with how long the original Macs took to boot, and loading the table of error numbers -> error messages was one of the things that got taken out to streamline the boot process. I guess it's just stuck.

I seem to remember the slow booting thing was the cause of the infamous 'throwing the prototype Mac down the stairs' Steve incident, although it's even more likely that I'm wrong on that one.

Re:Oh, right, error code -36!

[Trillan]

... or are they still living in a world where they refuse to acknowledge that Macs do throw up the occasional message to the user?

I think this is the case. Ultimately, they'll be right -- there are only a few places where the Mac shows obscure error codes. Actually, file sharing is aobut it now. Prior to Tiger, you could also get obscure error messages for dropped connections, but Tiger introduces a pretty neat Network Diagnostic tool that it offers instead.

Considering that SMB file sharing has been a problem since 10.1, it seems to be time for a SMB troubleshooter as well.

Re:Oh, right, error code -36!

[Megane]

Those low negative number error codes date back to 1984 with the original release of the Macintosh, but usually only a few come up. When you see them with OS X, you know you've got something with roots in the old days, like the HFS file system. And then there are the larger negative numbers (usually 4 digits) from when blocks of error codes were assigned willy-nilly to stuff like the Appletalk network stack and AFP file sharing.

And -36 doesn't help even if you know what it means, because it's just a generic "I/O error". Originally it was for media problems (like an unreadable floppy), usually accompanied by strange sounds from your disk drive, but for a network file system it's kind of silly. So even the old-timers say "yeah, that sure tells me a lot".

Other -3x range errors include file not found (-34?), end of file (-39?), and file name too long. Another good one is -50, parameter error. Well, duuuuuuh, which parameter? What's wrong with it?

The worst one to see is -127. That one means your file system data structures are in deep doodoo.

But seriously, the days of 400K floppy disks are long gone. It's total laziness that nobody bothers to print a text error message along with the number. I've been doing that in my own code since the days of 800K floppies. Even printing out the ten most common error messages as text helps most of the time.

Re:Oh, right, error code -36!

[NeedleSurfer]

Well in all fairness, have anyone ever solved any Windows crash by reading a blue screen? No, the information presented there is totally useless to anyone but the person who develloped the software. And even then.

this goes hand in hand with the concept that most "programmers" do it to impress people with their "intelligence" and "skillz", if you see a BSOD and pretend its usefull, you get extra auntie-impress points.

There is no use telling something to the user if what you tell him has no use...

Re:Oh, right, error code -36!

[matric]

Quite the contrary .. I was amazed as I watched a tech friend of mine troubleshoot my computer back in the Win95 days. BSOD came up, he noticed that there was a fault in a certain *.exe and was able to fix the problem soon after.

Re:Oh, right, error code -36!

[jellomizer]

Well I don't agree with your idea of Apple's mantra 'Users are Idiots' I use to but it is more of mantra 'Don't make it needlessly difficult to do common tasks'

As a programmer I will often give me error numbers because when I need to fix it the error numbers help me find it in the code quicker. And when I give more detailed error messages. The users will try to analysis my message outside of the context of my code and try to fix it them self. So if I put an error message "Out of Allocated Memory" except for error 49112, the user will go out and buy some more ram hoping it will fix the problem except for going to me and saying hey I have an error 49112 where I will know that I will need to change my code to either be more memory efficient in an area or allocate more ram.
It is not a situation that the User is an idiot it is that they may not have the context of how things are running in the programming level. So when they see an IO error they will go trying to fix there network cards, reinstall their printers and other drivers before reporting the problem with the program.

Re:Oh, right, error code -36!

[m50d]

Yes, people can. I have myself repeatedly workarounded driver faults by seeing from the blue screen which vxd is the cause and disabling it. If I had the driver source I could also fix the error in the actual driver; I haven't done this myself but have seen it done.

Re:Oh, right, error code -36!

[Junior J. Junior III]

Yup, that and the "spinning beachball of death" mouse icon that tells you the system is busy doing... er... not doing?... uh... something.

Horrible interface decisions in an otherwise pretty good UI.

Re:Oh, right, error code -36!

[hey!]

They're just pandering to the Geek crowd.

"Oh, yeah, -36, that's an I/O Error. Check the logs, then sacrafice a pure white chicken under the full moon and pour its blood into the NT server."

They're just trying to rope in the Geeks along with their artsy-fartsy core fanbase, with the hope that some will mate, producing a new generation of geeksy-farts ultracustomers who will be irresistably drawn to Apple's unique blend of superior design and industrial strength Unix aracana.

Re:Oh, right, error code -36!

[As Seen On TV]

Actually what the spinning cursor icon means is that the program that has focus has events waiting to be processed by the run loop. That cursor appears automatically when an event waits for longer than a hard-coded threshold ... I think it's three seconds, but I doubt myself and I don't feel like looking it up right now. It would usually happen when the process was waiting for a kernel lock for some reason, usually disk or network I/O. The incidence in Tiger should drop dramatically thanks to finer-grained kernel locking.

Admittedly this is an esoteric implementation detail. It's not really meant to communicate anything to the user other than "I'm waiting."

Re:Oh, right, error code -36!

[Maserati]

Oh yes. Googling for the exact error message will sometimes lead to instructions on fixing the problem. Sadly, the advice is usually "run Windows Repair from the installation CD, if that doesn't work then be glad you already found your installation CD".

The useful error messages I've seen have been from BSODs during bootup. If memory serves, you want the hexadecimal codes on the first lines.

Re:Oh, right, error code -36!

[Anonymous Coward]

Either that or someone needs to take SMB out back and put it down for the count like Ol' Yeller. It's unreliable and pretty inefficient, in my experience.

Re:Oh, right, error code -36!

[westlake]

Even the good old Windows blue screen is more informative than "error code 4"

and, on recovery, Microsoft's Crash Analysis site will give you a plain english explanation of what went wrong and often a fix, such as a link to an updated hardware driver.

Re:Oh, right, error code -36!

[repvik]

Yes. I have. Several times. Last time was yesterday.
Often (But unfortunately not always), it lists the offending DLL (which in my case was NetPeeker). It was as simple as rebooting, uninstalling netpeeker, rebooting again, *then* try to enable my wireless card ;)

Re:Oh, right, error code -36!

[SA Stevens]

Yep. I once had my NT 4.0 system crash because I dared boot it up without a disk in the (parallel port) Zip drive. The bluescreen indicated exactly (well, through a little bit of investigation) that.

As to why the system *crashed* because there wasn't a zip disk in the drive... well....

Re:Oh, right, error code -36!

[Fallen_Knight]

Not all the time, and sometimes a wide range of errors can be covered with a single error code. So it really is NO help at all alot of the time.

Re:Oh, right, error code -36!

[not_anne]

Most every error message you may get while using Apple operating systems are listed in the Apple Knowledge Base (in the Support tab from apple.com).

A search for "-36" finds article# 301580:

Mac OS X 10.4: Error -36 alert displays when connecting to a Windows server
http://docs.info.apple.com/article.html?artnum=301 580

Re:Oh, right, error code -36!

[NeedleSurfer]

Reading all the answer here bring me to the next logical observation:

You guys consider that MacOsX saying error -36 is bad because you have to lookup what it means in some error number reference guide in pdf format but Windows BSOD is good because, sometimes when you google the entire error you can get some help somewhere online...

Re:Oh, right, error code -36!

[Anonymous Coward]

it's a little of both.

having worked tech support, I know what the standard response to "what was the error message"

the answer is "I don't know..I closed the window"

so most people don't read them..they just make the window go away.

the ones that DO read...then immediately want you to replace their I/O by this weekend. I mean..how log can it take to get a new I/O? and what went wrong with the I/O that's in there?

Re:Oh, right, error code -36!

[lastchance_000]

Actually, it's more a case of the MacOs error code doesn't provide enough information to pinpoint the problem, whereas a BSoD occasionally can.

Re:Oh, right, error code -36!

[Leghkster]

Reading that solution as a long-time Linux user/professional Linux & Windows network admin/2nd week Mac mini owner, the thing that struck me was the last line of the instructions - "10. Restart your computer". Bleh. Is there no equivalent to "10. Type "sudo /sbin/service smb restart" and press return"? Or SWAT? Ouch. Don't flame me too hard, Mac pros, I'm just trying to learn this your platform for learning's sake :)

Re:Oh, right, error code -36!

[JFitzsimmons]

And what exactly would you suggest is better? SMB is supported by every major operating system, has several different methods of authentication, and file transfer speeds are about the same as FTP - it pretty much fills the pipe no matter what. SMB is reliable and efficient when administered properly; be it the Active Directory/LDAP setup for windows/UNIX we have at work, or the host based configuration that I use for the substantially smaller network I have at home.

Re:Oh, right, error code -36!

[hxnwix]

How about this one: mounting of certain authenticated, plain jane, non-active-directory smb shares that worked just fine in panther fails in tiger 100% of the time with other obscure errors:

mount_smbfs: spnego blob2principal error 1
mount_smbfs: tree connect phase failed: syserr = Permission denied
mount_smbfs: error from NetrShareEnum call: exception = 382312522

Looking at the samba logs of the server providing the problematic share, it appears that the bug does indeed lie in panther:

mount_smbfs: spnego blob2principal error 1
mount_smbfs: tree connect phase failed: syserr = Permission denied
mount_smbfs: error from NetrShareEnum call: exception = 382312522

Re:Oh, right, error code -36!

[klubar]

It's the classic Apple attitude of "it just works". Why bother with documenting those pesky errors (which never occur) for tech users. Go back to using itunes, iphoto and other apple applications.... The Mac isn't really a cross-platform business ... as indicated by their low market share in the medium to large business market.

Re:Oh, right, error code -36!

[hxnwix]

DAMMIT. Here's the relevant snippet from the samba logs:

[2005/05/07 12:11:49, 2] auth/auth.c:check_ntlm_password(312)
check_ntlm_password: Authentication for user [nobody] -> [nobody] FAILED with error NT_STATUS_WRON
G_PASSWORD

Re:Oh, right, error code -36!

[commodoresloat]

...you can see lots of stuff on the blue screen. Sometimes, just googling one line in there will get you an answer.

How do you use google while you have a blue screen?

Re:Oh, right, error code -36!

[ColMustard]

Actually, you don't have to be a developer to submit bug reports/enhancement requests. Anyone can submit feedback.

Re:Oh, right, error code -36!

[wft_rtfa]

You must have a book or a reference to know the meaning of an error... reminds me of my days of maintaining Fortran code written in the 70s.

Re:Oh, right, error code -36!

[bahamat]

The thing that really frustrates me is that OS X's SMB support uses Samba. I don't have any problems accessing SMB shares using Nautilus, but the Finder can't seem to connect.

Actually, I've been running into this problem at work the past few days. For some reason browsing to the server doesn't work, but if I use the "Connect to Server" menu option and type in "smb://1.2.3.4/" it connects no problem in just a few seconds and shows me all the shares available. I'm beginning to think it's merely a problem in resolving NetBIOS names to IP addresses. If anybody knows a tweak to take care of this.

Re:Oh, right, error code -36!

[Maserati]

No, I was directly addressing the claim that the Windows error codes are useless. Sometimes you can actually fix the problem rather than reinstall, but you have to search for the info. Same with the Mac codes, a website was mentioned earlier that's dedicated to the subject, and any search engine will turn up a wealth of information. In my experience the Mac error codes are generally easier to find information about than the Windows codes. Or, there are more obscure Windows error codes (and many more requiring a reinstall). Neither vendor installs a troubleshooting reference with the OS, it'd be nice.

And keep your ERD disk handy on any Windows system, I just keep one in the drive (not inserted) and update it occasionally. Like just right now. Everyone, do that right away.

Re:Oh, right, error code -36!

[madmancarman]

"Oh, yeah, -36, that's an I/O Error. Check the logs, then sacrafice a pure white chicken under the full moon and pour its blood into the NT server."

If you've ever administrated an NT server before, you know that sometimes pouring blood into it is the only appropriate solution.

Re:Oh, right, error code -36!

[J. Random Luser]

You could always burrow deep into the bowels of (panther & watch for wrap ;-)

/System/Library/Frameworks/CoreServices.framewor k/Versions/A/Frameworks/CarbonCore.framework/Versi ons/A/Headers/MacErrors.h

There are some quite humanly readable comments on some numbers as to what might have happened. I'm surprised this document hasn't been put up somehwere in clean html. Oh, of course, Apple would sue...

Re:Oh, right, error code -36!

[MoneyT]

How is that anymore helpful though?

Network error what?

No connection?

No network inferface?

Bad authentication

Bad protocol?

Bad Packet?

Re:Oh, right, error code -36!

[MoneyT]

you probably could do the same thing, but for the sake of most of their users, and the sake of keeping with the never have to use the command line thought process, they just tell you to resart.

As it stands, service is a command in OS X so I assume it would work.

Re:Oh, right, error code -36!

[MoneyT]

They are documented for tech users, you just might actually have to, GASP, read the documentation.

Re:Oh, right, error code -36!

[WMD_88]

"Oh, yeah, -36, that's an I/O Error. Check the logs, then sacrafice a pure white chicken under the full moon and pour its blood into the NT server."

Sounds like Bastard Operator From Hell's "DUMMY MODE" to me. :)

Re:Oh, right, error code -36!

[qw(name)]

While there isn't a restart command available, there are the stop and start commands. Also, on the Mac the service is called smbd.

$ sudo /sbin/service smbd stop
$ sudo /sbin/service smbd start

if you type:

$ service --list

you'll get a list of the available services with which to work

Re:Oh, right, error code -36!

[Trillan]

Well, I don't suggest putting down SMB just yet, much as I'd like to. But isn't WebDAV client support nearly as ubiquitous as SMB now? Yeah, I know the personal server stuff isn't there, but I hope in a few years...

Re:Oh, right, error code -36!

[Leghkster]

Thanks, I discovered that after reading this discussion and poking around on the system. It was just a layer of curiosity I hadn't gotten around to yet. It still seems odd to me that they'd document how and what to change in the smb.conf, but not how to do this and reinforce their HI guideline's "always-on" point. For that matter, I'm surprised there isn't a nice GUI panel for services management, or options in the existing panels to start/stop related services.

I suppose that's just too much geekiness to expect. It's nice to know that it's available at the command line.

Re:Oh, right, error code -36!

[qw(name)]

Actually, there is another way to accomplish this. In your System Preferences, click on Sharing. In the Services tab you'll see Windows Sharing. You could stop and start that because the help page for Mac OS X Services show that one "using SMB/CIFS (Samba)". I haven't tried it but it should work.

Re:Oh, right, error code -36!

[JFitzsimmons]

Good thinking, I should look into that.

Re:Oh, right, error code -36!

[Guy Harris]

You might want to get network traces of the attempt to connect from the Tiger machine, and from the Panther machine, and tell Apple to open a bug and add those network traces to the bug report. (If you do the trace with tcpdump, use the flag "-s 0", as well as using "-w" with the name of the file to which to send it.)

Re:Oh, right, error code -36!

[drsmithy]

Consider yourself lucky to get an error message. All that happens when I try to connect to my FreeBSD Samba server is a perpetual "Connecting" dialog (no timeout, no errors, nothing). I even left it for a good 30 minutes the other day just to make sure it wasn't a really long timeout.

Same server worked fine with 10.3 :(.

Interestingly enough, my Linux samba server still works fine...

Re:Oh, right, error code -36!

[TClevenger]

They're also continuing the tradition of such useful messages as "Not all files could be copied because one or more items could not be read." Since you can't even view the files scrolling by anymore like you could in Panther, you can't even tell what DIRECTORY it's having the problem with.

Here's a bet:

[Capt'n Hector]

Whatever the issue is, my guess is Apple will have it fixed within the month. It's possible they will have a patch out by the end of next week. It's just a bug, and last time I heard, unless active measures need to be taken by network admins NOW to shore up potential security issues, bugs aren't news. Major new OS versions will always have wrinkles to iron out, stop the presses!

Re:Here's a bet:

[xiando]

Really? "Major new OS versions will always have wrinkles to iron out, stop the presses!"? The reason for doing beta testing would be what? Is it too much to ask that vendors use beta versions of their own software in-house for a month before they release it? Is it too much to ask that they ship the software to a small number of beta testers before the final release in order to find those wrinkles and iron them out? If I were to pay for commercial software, would I be paying THEM for doing the work of beta-testing for them? If you bought a car, would you really accept that it broke down after a few hours, even if the store told you that "it is a new car, you can expect some wrinkles to be ironed out, we will take it into service and give you it back in working condition in a few weeks?"

Re:Here's a bet:

[gullevek]

I agree with you. Its super annoying that an "Gold" has these kind of errors. If it is with a super rare hardware or a super special software, okay.

but THIS? Has nobody there ever tried to connect to a SMB sever? It's kinda strange. Annoying. Every OS has this, everyone.

But I can imagine how this is, I can imagine this very good. The coders will say, we need to the test, the managers say, we need to release, and of course the managers are right, they get their bonus, because the release in time, and the coders then get the blaim for the code errors.

Perhaps IT needs a revolution.

Re:Here's a bet:

[Graymalkin]

So, say it worked great in all beta builds until the gold master. It had been tested and came up green so in latter beta builds it wasn't tested anymore because it worked. Then say sometime between the last beta build and the GM (which are a few builds apart) a butterfly flapped its wings bug caused SMB mounting to break in Finder. Errors happen because systems are complex and there's dependancies that depend on more dependancies, a error in the chain can cause really weird errors in seemingly unrelated parts of the system.

Your car analogy is flawed. New cars do have bugs when they roll off the lot. You would be really surprised at the number of real issues every car or every batch of cars has off the factory floor. Many times however these flaws and bugs don't crop up and cause a noticeable problem for a long time if ever. There are some problems that do crop up quickly however. It would be one thing if the manufacturer ignored this and went on its merry way. It is entirely another if they repair your car for you. I just had the dome light fixed in my car because of a faulty latch, should I be screaming about the manufacturer not having any QA? No.

The car analogy also falls flat when compared to something as easily changed as computer software. A patch containing the repair can be very small and be distributed to millions of affected users very quickly. If your car is in the shop for a week you're out one car. If SMB shares don't show up in Finder's Browse window properly you're not out SMB shares as you can work around the problem if need be.

Re:Here's a bet:

[Daengbo]

Yeah, I'm wondering where "As Seen on TV" is and why he hasn't posted 20 times for this story. Can the Apple troll defend "his" company for this one?

Re:Here's a bet:

[peragrin]

Exactly. i bought a 2002 Jeep liberty brand new.

I took a known risk, new car model, in the first run of production.

I have had to general recalls. First one was a wire harness hasn't really there and the controls lines for airbag deployment could get severed preventing the airbag from going off.

The second was a poor boot cover design for the tie rods.

I owned the car for two years and they replaced the tie-rods and put in new covers.

Minor problems should be expected. If you buy say a Ford Taurus you won't have many of them as they have been worked out for the most part.

Re:Here's a bet:

[BasilBrush]

Yeah, where can he possibly be between 4:29am and 6:29am? Clearly it's a cover up!

Re:Here's a bet:

[Trillan]

A new car built by my company leaves somewhere traveling at 60 mph. The rear differential locks up. The car crashes and burns with everyone trapped inside. Now, should we initiate a recall? Take the number of vehicles in the field, A, multiply by the probable rate of failure, B, multiply by the average out-of-court settlement, C. A times B times C equals X. If X is less than the cost of a recall, we don't do one.

Sorry, couldn't resist. :)

Re:Here's a bet:

[Megane]

Many times however these flaws and bugs don't crop up and cause a noticeable problem for a long time if ever.

And many of them don't cause the vehicle to go out of control or even fail to operate. Like the wonderful paint GM used in the '90s which had a bad habit of flaking off after a few years. My old Chevy Blazer had that paint, but I didn't really care because I bought it used.

Re:Here's a bet:

[Megane]

I mean, geez folks, it's an x.0.0 release. When has one of those ever been perfect? By anybody? I've only upgraded one of my four Macs because that one was the one I use the least, and I still had to tweak it to get what little I used it for working right, because the install overwrote a config file in /etc without even saving a backup copy.

Re:Here's a bet:

[eraserewind]

It had been tested and came up green so in latter beta builds it wasn't tested anymore because it worked.

No offense, but what the hell sort of software engineering practice do you call that?

Re:Here's a bet:

[NeedleSurfer]

I guess your post and the parent one will be switched when a new Windows or Linux release gets out.

fanboys... pfffff

Re:Here's a bet:

[newrisejohn]

If you bought a car, would you really accept that it broke down after a few hours, even if the store told you that "it is a new car, you can expect some wrinkles to be ironed out, we will take it into service and give you it back in working condition in a few weeks?"

That's why there's lemon laws.

I agree that the problems Tiger is having should have been taken care of, but let's remember it's not as bad as it could be. (Win 98 BSOD on startup, anyone?)

Re:Here's a bet:

[Atryn]

Take the number of vehicles in the field, A, multiply by the probable rate of failure, B, multiply by the average out-of-court settlement, C. A times B times C equals X. If X is less than the cost of a recall, we don't do one.

The problem is when you underestimate either B or C. So X begins to approach the total cost of a recall and management thinks "hmmm... maybe we should go ahead with the recall". Only now your total cost is approaching 2x. If you had just done the recall immeadiately (the honorable thing) you could have limited cost to approximately X.

I own two Toyota vehicles, both have had recalls for minor issues where X was probably far too low to justify the recall. But they did it anyway.

Re:Here's a bet:

[jonadab]

> The reason for doing beta testing would be what?

As a general rule, alpha testing (what the programmers themselves do in house) catches bugs that happen pretty much every time, on pretty much every setup; beta testing catches bugs that happen more often than not. The rest don't show up until you ship the thing.

Of course, beta testing can be more or less extensive, and the more extensive it is, the more it is likely to catch. For really heavy-duty beta testing, you need to release the beta publicly, on the internet, for free. I'll leave you to employ your imagination as to what why Apple might not desire that approach.

Re:Here's a bet:

[Trillan]

It's a quote from Fight Club. :)

Re:Here's a bet:

[gullevek]

there is already a 10.4.1 build in the dev tree. I didn't see the samba problem in the chanelog, but as it is a kind big one, I hope it will come in the next build.

Re:Here's a bet:

[Mr. Underbridge]

A new car built by my company leaves somewhere traveling at 60 mph. The rear differential locks up. The car crashes and burns with everyone trapped inside. Now, should we initiate a recall? Take the number of vehicles in the field, A, multiply by the probable rate of failure, B, multiply by the average out-of-court settlement, C. A times B times C equals X. If X is less than the cost of a recall, we don't do one.

What about D, the cost of the public relations nightmare when your car is called the "New Pinto?"

Re:Here's a bet:

[jskiff]

It had been tested and came up green so in latter beta builds it wasn't tested anymore because it worked.

No offense, but what the hell sort of software engineering practice do you call that?


So with a codebase as large as Tiger, you're implying that they should regression test +every+ build? Surely they're doing automated validation testing, and perhaps these components were continuing to show green?

I can understand a heavy regression test on the release candidate, but if you tried to test every component with every build, you'd never release anything!

With this logic, Duke Nukem Forever should be the highest quality release ever!!!

Re:Here's a bet:

[ioslipstream]

Works fine here, it is hardly OMG widespread.

Re:Here's a bet:

[J. Random Luser]

Whatever the issue is, my guess is Apple will have it fixed within the month. It's possible they will have a patch out by the end of next week.

Tried this one? dated 05 May 2005

Re:Here's a bet:

[Samrobb]

Its rather easy to check the bloody charts and make sure that you test all the other functions that are dependent on any code that you changed. Thats the responsible thing to do.

Sure. But at this point, you're assuming that your regression tests do indeed cover this case, and that they're written in such a way that they will always catch this particular failure mode.

Note that I'm not saying that there was a test-case failure in this particular situation. I don't work for Apple, I have no idea what their tests are like or what tests were run for this release. What I have seen are regression tests that passed, even when a feature was broken, because the regression tests were written in a particular way that failed to tickle a bug that (in hindsight) should have been caught by the tests. Tests are code, too: your regression tests are only as good insofar as you've tested them.

I've also seen properly written regression tests, that should have caught a particular bug, fail to fail (hah!) becuase of a seemingly trivial aspect of the test environment or the test hardware. This kind of test case failure often crops up specifically when you're doing integration or interoperability testing.

In short: regression tests are great, they're essential, but they can be far from perfect. It's easy to complain and ask questions like "What, didn't they bother testing this?" in the face of a seemingly obvious bug. If you don't know the details about how something was tested, though, you really can't say if a problem like this was a symptom of generally bad testing, or the result of a set of test cases that somehow just managed to miss the problem.

Re:Here's a bet:

[Anubis350]

it works for me interestingly enough, samba 3.0.1 on debian testing and osx tiger. I should prolly go read the article to see what the the prob is.

Active Directory

[October_30th]

Interesting. I didn't know you could authenticate non-Windows computers with Microsoft Active Directory servers. It's rather surprising that Microsoft supports such interoperability.

Re:Active Directory

[Tony Hoyle]

Linux has had it for a while (via samba & winbind).

Panther had it, but it was a bit busted (didn't give you a proper kerberos key for example, and didn't really integrate with the samba filesystem). Panther also had the issue that you had to practically destroy your network security to get it to work as it didn't support encrypted connections or NTLMv2.

Tiger fixes most of this (connected to my Win2003 ADC without having to tweak anything, and smbclient -k works now) - it's just that you can't actually browse then network with finder as the links it creates don't work... it can find the server names but not any of the network shares.

One thing they haven't fixed is that AD users can't share folders on the OSX machine - you have to create a physical user and give them a global shared folder (I just created one called 'public' and paste all the files to that). It's more a workaround than a fix & I hope they address that someday.

There's also the new 'mobile' account settings. Not really sure what they are - I tried enabling it and it just gave errors about home directories when I tried to log in... didn't appear to actually change anything. I'll come back to that sometime when someone explains what it's supposed to actually do.

Work-around

[Noksagt]

Easy workaround:
Command-K to bring up the connect menu and type in the full address INCLUDING THE SHARE NAME:

smb://SERVER/folder

Re:Work-around

[Noksagt]

Also note that you can also do it on the CLI. Open up term and do a:
$ mount_smbfs -W workgroup //user@SERVER/folder ./mntpoint

Re:Work-around

[teh kurisu]

Doing it from the terminal using mount_smbfs seems to create an unmountable reference to the share in the Finder. Of course, it could be my own damn fault... but the Command-K method seems more reliable to me.

Re:Work-around

[wimvds]

If this is true, then this issue/bug is definately not restricted to Tiger. I'm having the same problem in Panther since one of the upgrades (I don't recall which one :p), and I know quite a few people on Mac forums I frequently visit that are having the same problem.

Re:Work-around

[squiggleslash]

This is one of the things I like about OS X. You have essentially two ways to do everything, Apple's way, and *IX's way. Despite the assertion of many Mac fans that everything "just works" with OS X, frequently it doesn't and there are often things missing that would really, really, help with whatever it is you want to do.

Want to create an arbitrary share like you can under Windows? Right clicking on the directory will not help. Pretty soon you realise there's actually no easy way to do it. Apple presumably wants you to buy OS X Server for that.

But it doesn't matter, because OS X has Darwin underneath, and Darwin's just another made-over Unix-clone. It's running Samba. So quickly sudo vi /etc/smb.conf, add the entry, kill -HUP {smbd's PID}, and you're up and running.

The Finder is just a shell over all of this, and if it doesn't do something, either because the designers didn't think about it, or because it doesn't work properly, you can generally workaround the missing functionality. As you've just described.

It's an OS semi-guaranteed to be undamaged by marketing hype and marketing decisions. What more could you want? (Well, othe than the entire thing to be FOSS? *sigh*)

Re:Work-around

[Tony Hoyle]

Now *why* isn't that somewhere more obvious.. I spend nearly an hour looking for something like that before giving up.

There's a lot to be said for having a location bar.

Re:Work-around

[As Seen On TV]

This has been changed in Tiger. In versions of Mac OS X prior to version 10.4, command-line mounts wouldn't show up in the Finder unless the disk arbitration service were manually refresh by typing "disktool -r" at the command line.

We've changed the way filesystem events are propagated through the system in Tiger, so this is no longer necessary. Command-line mounts work just like Finder mounts now.

Re:Work-around

[mytec]

That's great until you notice that the domain name has been truncated from seven characters down t o one and constantly asks you to re-authenticate because some part of the process continues to truncate the domain.

I'm one of those who went from having a laptop that functioned just fine on the Windows domain to accessing maybe 50-60% of the shares and not consistently. Some of us do have some real bugs on fresh installs.

Re:Work-around

[Noksagt]

However, at times you will also have to include the workgroup/domain for that user, I think it's -W.

Yup.

Re:Work-around

[eltoozero]

Yes, this is confirmed; putting the share name works right.

I've been doing this properly since MacOSX could connect to SMB shares, and I've never had a problem.

Thanks Noksagt, some people just can't RTFM.

No bugs, just switchers.

Re:Work-around

[phillymjs]

Want to create an arbitrary share like you can under Windows? Right clicking on the directory will not help. Pretty soon you realise there's actually no easy way to do it. Apple presumably wants you to buy OS X Server for that.

More like Apple wants all data on an OS X client machine to be somewhere in a user's folder rather than placed arbitrarily elsewhere on the drive. I have to agree with this stance-- in the pre-OS X days people would put their files wherever they wanted them (and frequently, accidentally and unknowingly where they didn't want them). If their machine became problematic and needed to be rebuilt I'd have to look in every directory for errant data files that might be important and retrieve them before wiping the drive. With OS X, stuff *must* go in their user folder. If the machine needs to be rebuilt I just have to back up the Users folder to know that I got everything of importance.

If you're too lazy to use your Public and Drop Box folders for sharing your local data with peers on the LAN, (or if you legitimately want to use a spare OS X Client machine as a cheapie file server with a 10-simultaneous-connection limit) you can always download and use SharePoints-- just not on any network that I admin. :-)

~Philly

Re:Work-around

[Baloo Ursidae]

It's an OS semi-guaranteed to be undamaged by marketing hype and marketing decisions. What more could you want? (Well, othe than the entire thing to be FOSS? *sigh*)

What, Linux, FreeBSD, OpenBSD, and the bajillion other free OSs; KDE, Gnome and the 8 bajillion window managers and desktop environments weren't enough options for you?

Re:Work-around

[squiggleslash]

We're talking about Samba not AppleShare here. PC users can only log in as real users on a Mac (if you use the default configuration) and then access that user's complete home directory.

Needless to say, this isn't really a sane option.

Re:Work-around

[warkda rrior]

OS X is finally catching up to the level of UI elegance of Linux!

A typical slashdot response.

[King_of_Prussia]

If this were a windows article there would have been an almost unanimous uproar about microsoft's ability to release a stable piece of software without major bugs. Look at the nature of the bug too -- how long until somebody blames this on windows being too "monopolistic" and deliberately making it hard for tiger to share or authenticate?

I've said it before, and I guess I'll have to say it once again -- zealotry should have no place on slashdot. If Microsoft turned around and released a perfect, bug free operating system that interfaced perfectly with all the competitions' offerings, there would be a 1000 comment shitstorm of complaint as the flock of rabid posters decried them for not releasing the source, or for charging for the software. Compare that to this, where a major operating system has been released with a large and quite frankly obvious bug present, and along come the apple fanboys. GET OVER IT. Base your opinion on the product, not the company, or the shiny form factor, or the how overpriced it is.

Don't get me wrong, as I sit here I am listening to a 40 gig iPOS, and I use a powerbook when I need mobility, so I don't have any bias against apple themselves, just their little army of braindead followers who would buy and defend a box of Steve Jobs' shit if it had a pretty shape and the apple logo.

Hah, and it seems after previewing the parent comment is already rated insightful. Funny how that works, isn't it?

Re:A typical slashdot response.

[FidelCatsro]

um .. what has not being able to properly connect to a SAMBA share got to do with stability.Tiger is very very stable i have not had one OS crash (a few programs have but they were built for 10.3 and the updated version run fine)
The fault here is in interoperability with a Microsoft SMB share (no such problem with NFS) and there is an easy work around (you just point to the share directly).

Had this been about microsoft products not connecting to a SAMBA share properly .Then quite rightly, there would have been a hell of an uproar.

Re:A typical slashdot response.

[popeyethesailor]

If Microsoft turned around and released a perfect, bug free operating system that interfaced perfectly with all the competitions' offerings, there would be a 1000 comment shitstorm of complaint as the flock of rabid posters decried them for not releasing the source, or for charging for the software.

But the problem has been that Windows has never had a perfect release whereas Apple has had a wonderful track-record in most of their releases. [insert long tirade about security here[ [insert monopolistic practices rant] [insert disdain for windows 98] [insert smug sense of superiority]

SAVE IT.

Re:A typical slashdot response.

[elecngnr]

I will just say at the beginning of this post that I am a fan of Apple products. I try not to jump in on every Apple story on this site because I think there is enough preaching to the choir on this site. Having said that, I will continue on this thread. I have used Windows machines for many years in addition to using Apple. The reason why there is no huge uproar, in my opinion, is because I know it wll be fixed soon. I also know that the fix will make the product better (i.e. it will NOT be SP2). It is not so much that we are brain dead followers....I would not just drink some kool aid if Steve asks me to....I think many of us have just had good experiences with their products. I upgraded to Tiger on Monday of this week. I expected some hiccups and there have been a few. However, they are not major hiccups and I do not expect to be dealing with them for long.

Re:A typical slashdot response.

[rokzy]

exactly correct. I've noticed a few little things with Tiger, but have complete confidence that it'll be fixed. I also submit a few bug reports because I believe they will actually make a difference.

conversely, does anyone really think MS could give a shit about the end-users' experience? does anyone bother with the "send report" feature? when has anyone at MS made something better just for the hell of it? for example why, after probably almost a decade, is it still necessary to run disk cleanup for every partition separately. is it really too much work to add an "all drives" option, or am I the only person in the world that thinks this is and appalling UI?

people get pissed off when MS does something wrong because if MS doesn't get it working first time then it usually stays that way or goes downhill.

Re:A typical slashdot response.

[Tony Hoyle]

How do you point to the share directly... genuine question... I don't know OSX well.

There's no location bar in Finder so you have to rely on the links in 'Network', but they don't work (actually it's wierd.. they work for the ADC and one other machine but none of the others).

There's also a wierd file called 'Library' sat in my networks folder that's undeletable. Doesn't appear to do anything (another bad link, which OSX offers to 'fix' but the 'fix' brings up a file selector and I'm kinda lost at that point).

Re:A typical slashdot response.

[Atryn]

zealotry should have no place on slashdot

Maybe you should change your reading preferences. I surf at 1+ and so far your post is the most extreme I've seen. The parent of yours is at 3, but I would hardly call his post "zealotry". Perhaps unsubstantiated opinion, but not "zealotry"... You post, OTOH, looks 10 times as long to go on and on about zealotry... And I'm assuming your use of 'iPOS' is a typo and not an insult or else your post would be entirely self-contradictory.

Re:A typical slashdot response.

[FidelCatsro]

you can do so from the top bar.
on finder select Go from the top bar then "connect to server" to then you can just adress it like smb://*IPaddy-or-server name*/*Share-directory*/ (or you can press command+k to open up the screen then do the same) .
You can also use samba on the command line if your more comfertable there.

Re:A typical slashdot response.

[Tony Hoyle]

Yeah the samba command line works (smbclient -k) but it's not as nice as browsing.

I guess I'll have to do it manually until they fix it.

Re:A typical slashdot response.

[FidelCatsro]

If you notice the grandparent said he was not fammiliar with OS X so i assume he is also not fammiliar with the Apple OS in general .
Fair enough i could have said menu bar , but if he is more fammilar with windows or even a DE like KDE ( i know you can have KDE put the menu bar at the top aswell) he may have been looking at a finder window trying to see where it was.
A little system literacy is always good , but it never hurts to try and simplfy things for people.

Re:A typical slashdot response.

[yasth]

Actually every OSX release has had a major bug fix pack shortly after release. It is expected, and lots of people wait till that to install. Same with MS. I mean, it is just like sending out any other thing you always have a few D'oh momments.

Opposite Experience

[Anonymous Coward]

Weird, I've found with Tiger that Windows file sharing has been easier, although I don't use Active Directory. With Panther my password was never remembered by Keychain, despite clicking the option to enable it. With Tiger my password is remembered. It also finds my Windows shares automatically, whereas with Panther I had to manually connect by entering IP addresses.

Re:Opposite Experience

[Chanc_Gorkon]

Hmm....weird. Never had a problem with Tiger or Panther?

Re:Opposite Experience

[solios]

The panther-not-remembering-samba-passwords thing has been annoying the crap out of me since 10.3.0 - more for the fact that (a) I have to interface with samba and (b) my freenix friends who don't run windows AT ALL for some damned reason think Samba Is Awesome and get all >:| whenever NFS is mentioned.

Good to know at least ONE thing about the work network is going to suck less...

Anecdotal...

[Shag]

One friend indicated that things refused to work in plaintext-password mode, but once he turned on encrypted passwords, they worked fine.

I'm not sure whether he had to turn on the encrypted passwords at the Mac end or the PC end, but I seem to recall thinking "gosh, imagine that, doing something the secure way."

Re:Anecdotal...

[mattyrobinson69]

newer versions of windows require attacking the registry to enable plain text passwords.

Re:Anecdotal...

[Slur]

I believe this is the definitive solution to the problem.

Re:Anecdotal...

[doon]

We did this at the office on a FreeBSD box and it fixed the problem. Enabled encrypted passwords on our Samba Server, cleared our keychains and smbpasswd our FreeBSD accounts, and now it works fine again.

Re:Anecdotal...

[MyDixieWrecked]

I believe the encrypted session has to be disabled on the server end in windows2k3. In win2k, it's off by default.

IIRC, the encrypted session is a proprietary M$ extension to CIFS/SMB that the samba guys haven't (hadn't) implemented (reverse-engineered) yet.

Re:Anecdotal...

[Guy Harris]

I believe the encrypted session has to be disabled on the server end in windows2k3.

Encrypted session, or packet signing?

IIRC, the encrypted session is a proprietary M$ extension to CIFS/SMB that the samba guys haven't (hadn't) implemented (reverse-engineered) yet.

Recent versions of the FreeBSD smbfs (client VFS) do packet signing, so that might have been reverse-engineered, at least on the client side.

If you have to turn packet signing off on the server to connect from Tiger, presumably Tiger hasn't picked that up.

Re:Anecdotal...

[rodac]

Nothing really proprietary here...

Encrypted sessions is merely just applying GSS-KRB to DCE/RPC.
This is not a CIFS/SMB extension since it technically has nothing to do with CIFS/SMB. (eventhough CIFS/SMB is often a transport for carryiung DCE/RPC on older windows boxens, moderns ones transport DCE/RPC over raw TCP)

The only extension here that is "proprietary" would be that windows prefer to use ARCFOUR encryption due to the way it hashes passwords into secrets.
While ARCFOUR is technically "propriatary" since it has no RFC, it is a well known algorithm and many many people has implemented it. Even MIT and Heimdal Kerberos has supported ARCFOUR since ages, though they do not support GSS-KRB for DCE/RPC (yet).

Encrypted sessions is not that difficult, it is just that since is mainly used/enforced today between DCs so no one has really bothered to implement it yet in other implementations.
For someone that is bothered enough to implement "encrypted sessions" in a CIFS client/server it should not take very many days/weeks to implement/test, it is pretty simple (at elast if you only aim for ARCFOUR which is 99% of all encrypted DCE/RPC anyways).

It only took me a couple of days spare time to implement decryption of them in ethereal...

Re:Anecdotal...

[Guy Harris]

eventhough CIFS/SMB is often a transport for carryiung DCE/RPC on older windows boxens, moderns ones transport DCE/RPC over raw TCP

Actually, I suspect Windows NT has always done DCE RPC over raw TCP (and probably raw UDP), and I think for SMB-related services they might still do DCE RPC over SMB by default (or perhaps do DCE RPC only over SMB - do they make SRVSVC available over raw TCP, and, if so, do newer versions of the Windows SMB client code use that to enumerate shares or do they still do that via DCE RPC over the SMB connection to the server?).

Re:Anecdotal...

[rodac]

Microsoft used to use DCE/RPC over SMB or in dce-speak over nca-cn-pipe over well known endpoints down in \PIPE on the C$ or the same special files down in IPC$.
This had the benefit it piggybacked on the SMB authentication since it was accessed through authenticated named pipes.

All these endpoints were also available through raw TCP ports, but that wqould require explicit EPM calls and explicit authentication in the DCE/RPC layer, so it is just easier to map the IPC$ share (and automatically authenticate to the server) and then open the named pipe with the well-known magic name.

SMB related interfaces have always been available also through TCP portsd in addition to the well known pipes, try some of Todd's tools over at bindview they can enumerate all endpoints where a interface is registered. SMB related services always used to have a well known pipe, these are services such as NETLOGON, SAMR, LSA, WKSSVC, SRVSVC, SVCCTL, etc.

Non cifs/smb related services do not use named pipes but always use the EPM to find the port to talk to (or talk through the service through the EPM endpoint). This includes the two dce protocols used by Exchange.

Starting with w2k3 Microsoft has stopped using named pipes also for normal cifs services and have started talking dce over tcp directly even for most of the classic interfaces such as samr and lsa. At least when w2k3 boxens talk to eachother they seem to now prefer nca_cn_tcp instead of nca_cn_pipe, even for the interfaces with well known pipe endpoints.
i have no idea why and have never been curious enough to try to find out

Re:Anecdotal...

[rodac]

"do they make SRVSVC available over raw TCP, and, if so, do newer versions of the Windows SMB client code use that to enumerate shares"

Yes, this service has always been available through TCP but has not been used by common clients).

Yes, when w2k3 boxens talk to eachothers nowadays, they do use nca_cn_tcp to enumerate shares.

Not sure if it's this...

[mferrare]

I had a problem with 10.3 authenticating to a W2k3 AD server and mounting shares. Turned out I had to modify the Domain Controller Security Policy on the server and set Microsoft Network Server: Digitally Sign Communications (always) to Disabled. I am now running 10.4 and I have no problems connecting to this w2k3 server.

I got this solution from here by the way. Thanks to Drew McLelland.

Re:Not sure if it's this...

[joel48]

Do you mean than 10.4 has no issues connecting to the Win2K3 server *with* encryption enabled, or it still requires it to be disabled?

Re:Not sure if it's this...

[mferrare]

All I know is that it works with it disabled. Haven't re-enabled it since getting 10.4

Re:Samba supports it

[October_30th]

"Try"? By support I meant, of course, that MS is not suing the pants of the Samba team and is not obfuscating the protocol beyond all hope for reverse engineering (which they probably could do).

I find this interesting, because at the university where I work, the security policy requires centralized AD authentication from all computers in the network. After that I've hardly seen any Linux PCs or Macs around anymore. When I asked about it from one of our IT guys, he said that you can't authenticate non-Windows computers with MS Active Directory.

I fixed my problems

[mr_zorg_mobile]

I had this problem too after upgrading. I found that deleting my SMB keychain entries solved it allowed me to login again (after getting my admin to unlock my account from all those failed attempts).

Re:Samba supports it

[spauldo]

The admin's wrong. Samba can do it now, although in all fairness it took a while after active directory was released for it to be able to work with it well. He's probably just basing that on old information.

As far as the protocol, SMB is (IIRC, I could be wrong) an IBM-designed protocol. It's been around for ages - hell, NT domains were just hopped up lan manager networks. The authentication in active directory uses a slightly modified form of kerberos - also an open protocol. They have tried to put a few legal barriers in the way, but those have been mostly ineffective.

Now, there is another possibility - it might be against policy at your university for non-windows machines to authenticate. If it's set up so that all machines have to be added to the tree by an admin, it's certainly enforcable, and thus your admin would be right in that particular case. He's just not right in the general case.

Re:Samba supports it

[CowbertPrime]

hi. AD is just LDAP with some extra cruft/bloat/stuff added; which is mostly documented anyway. Your IT department is clueless. You can also fall back to kerberos (which despite the FUD, interoperates with the majority of MIT Kerberos V implementations), if you did not have a functional (Open)LDAP infrastructure.

Connecting for SMB works better for me

[prodangle]

That's strange, in my case connecting to SMB servers has improved. In Panther connections always failed 1st time, and after reentering my login & password it always connected 2nd time. This issue seems to be fixed now in Tiger, as I can always connect 1sttime.

I do have some sympathy for apple regarding this. Anyone who uses Windows shares frequently will know that even different versions of Windows can have difficulty operating together.

Re:Connecting for SMB works better for me

[KiloByte]

Anyone who uses Windows shares frequently will know that even different versions of Windows can have difficulty operating together.

"different versions"? Hey, to get "difficulties" you need to stay at least within the same major version. Otherwise, you'll need to resort to tricks like copying files via a Samba box.

Re:? troll

[Mr_Escher]

It was troll because you largely missed the point of the parent.

Your post implied the parent said Tiger was unstable however I can't find such a point in there, just that there is a "a large and quite frankly obvious bug present". Which there is.

This flaw has little to do with Microsoft, it cannot connect to my Debian linux server in the office either.

Your post bitches and moans about the observation that there is a lot of pro-Apple bias on Slashdot and draws an irrelevant comparison to Windows SMB integration.

You are the very zealot the parent eluded to.

Oh, you also have no idea what a question mark is.

Finder and Linux Sambda shares

[reddish]

On a related note: I'm seeing really bad performance when copying a file from a Linux Samba share to my OSX machine (roughly 100 kb/sec, if that). Oddly enough, file uploads are ok (megabytes per second). Odder still, if I open a terminal and copy directly to my machine from the Samba share mount point, incoming copies are fast too. This has been going on from at least 10.2, and much to my dismay it is still an issue in 10.4. This really seems like the Finder is trying to talk Sambalese by itself (and does so differently than the SMB filesystem driver). Has anyone else noticed this behavior (and, perhaps, solved it)?

Re:Finder and Linux Sambda shares

[sammy baby]

red, I had a very similar problem under 10.3, except that I was often completely prohibited from writing to the directory through the Finder. Nobody on any forum was able to help me with this problem, although now I wonder if it's related to the resource fork issue mentioned elswhere in this discussion.

Re:Finder and Linux Sambda shares

[Arkan]

I had this problem too. I commented out the sockets options, and the performance problem disappeared. I did't took the time to fiddle around to determine what was the exact option that was causing the grief, but HTH.

Cheers,

--
Arkan

Re:Finder and Linux Sambda shares

[nicuramar]

Under Panther (at least), Finder doesn't like samba options such as force user or similar, which I use to reach my root mount-point on my local server.
Finder will not be able to write files into places it thinks it can't - apparently without checking if it really is the case.
Conversely, Finder will attempt to write into places it thinks it can, but it can't, only to fail with a somewhat weird error message.
I don't know if this has been fixed under Tiger.

Re:Finder and Linux Sambda shares

[whocares11]

I found using netatalk shares works better on a linux box then using samba shares. Netatalk allows OS X to connect to the linux server using AFP 3, which in my testing was much faster than SMB. Netatalk was not that hard to setup, but I did have a problem with setting up domain authentication.

Re:Finder and Linux Sambda shares

[KoolyM]

YMMV, but what worked for me was to replace the (default, IIRC)

socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192

with:

socket options = TCP_NODELAY IPTOS_THROUGHPUT

in my smb.conf It's not blazingly fast, but before I did that, I got 100 byte / sec connection speeds and worse.

Re:Finder and Linux Sambda shares

[rsax]

I was asking this earlier. Do you have any speed comparisons between Samba & netatalk while serving only Mac OS X clients? And what problem did you have while setting up domain authentication? I want to have authentication with an LDAP backend. Maybe it's me being naive but I didn't think it would be much of an issue if I use PAM + ldap to authenticate.

Re:Finder and Linux Sambda shares

[rodac]

That kind of massive performance drop can really only be caused by any of three things things:
1, congestion in the switch and packetloss ==> performance degradation due to the sender sitting idle most of the time to wait for a TCP retransmission to trigger.
2, duplex mismatch causing packetloss ==> performance degradation due to the sender sitting idle most of the time to wait for a TCP retransmission to trigger.
3, if you get poor results only when uploading files to the mac but good performance when downloading, a recent buglet in some versions of the mac os where it would stop sending out a tcp ack for every full sized segment thus causing the tcp layer to cap throughput.
i would recommend running a sniffer on the network and find out why it is so slow.

Re:Finder and Linux Sambda shares

[whocares11]

I did some testing with final cut pro 4. I do not have hard numbers with me, but with final cut pro if the capture drops a frame the capture aborts. On an isolated network we were able to do 3 simulations captures using netatalk and we were unable to do any captures using samba authentication. I know that does not directly correlate to performance, but we also found when copying large files (2-7 GB's) netatalk out preformed samba as well.

I tried configuring the PAM to use AD authentication, but I could not get it to work. It probably is not that hard, but we decided to go with a different solution before I could get it working.

All of our test were done on G5's running 10.3.

Re:Finder and Linux Samba shares

[Alioth]

It's another FTFF (Fix The Fscking Finder) issue I think. Finder behaves the same way with NFS - really poor transfer rates under Finder, but normal speed using 'cp' on the command line.

Oh, you mean, another problem?

[sammy baby]

I just installed Tiger this week and have not yet had the opportunity to test again, but in my Panther (10.3) install I had intermittent problems trying to connect to my Debian/testing Samba shares in my office. Or rather, I could connect to them with no problem, but copying files to it via the Finder was a no-go.

And, to be specific, this was definitely a Finder issue: I could use "cp ~/Documents/somefile.txt /Volumes/sambashare/" with no problem. But it made it very frustrating for me, and to the few other Mac users in my building, whom I was in charge of supporting.

(My eventual "solution" was to install netatalk to do an end-run around the problem. I still don't like that answer.)

Re:Oh, you mean, another problem?

[Bombcar]

See if you have hide dot files enabled on the Samba server, as that may make the Finder confused, as it stores resource forks and whatnot it dot files.

cp doesn't move the resource forks, in 10.3 at least. (I think).

I don't use samba anymore

[Sarin]

I used to work with samba, having a linux fileserver and a mac osx powerbook, but recently I started working with nfs. It seems a bit faster and more stable. When I change some file on the server, it's directly visible in finder - without having to refresh it.

I also was annoyed the fact when I turned my powerbook on after it went to sleep it would give me a lot of errors about unmounting a network drive. This also was the case with tiger. With nfs, those problems are gone an nfs mount will stay active after the powerbook comes back from sleep.

Re:I don't use samba anymore

[Anonymous Coward]

You're not alone, I use nfs too... but we're few and far between... the windows "is easier and "just works"" morons and lunatics are far more numerous... sad to say. I've been advocating NFS mounts for as long as I can recall... yet nobody listens, everyone wants SMB... to connect to those "higher performing fully responsible windows boxes that microsoft says they take responsability for (hmmm nobody's read the EULA but me I take it) :)

-Daedalus

Re:I don't use samba anymore

[m50d]

That's because, even on linux, samba is far easier to set up. Set up samba: add one script to the default runlevel, point browser to localhost:919 (port may be wrong) and go through the wizard. Set up NFS: two scrips need to run, and they have to go after the rpc portmapper, and configuring it involves editing several files in /etc.

Re:I don't use samba anymore

[failedlogic]

Is there a good 'quick' start guide for setting up an NFS and print server in Linux? I'll probably be using Slackware. Some of the guides get over-my-head too quickly. I just want to setup a read/write file share and a print share and make sure noone but me (maybe password authentication and tcp_wrappers... I really don't know!) can access the share.

It always gets complicated when security is added to the equation. Its not a simple matter, but its assumed you are a sys admin with a degree. I am not.

Re:I don't use samba anymore

[SA Stevens]

I've been running a BSD for a long time. When did NFS on Linux cease to be the disaster it was?

(disclaimer- the first time I installed NetBSD was on a 486 laptop (with no CD drive) with a PCMCIA ethernet card, over an NFS share from a Slackware box)

Re:I don't use samba anymore

[m50d]

A samba client is just as easy, mount -t smbfs //server/share /mountpoint. But to use it over your home network you need to set up the server too. *I* can set up a NFS server fine, but for joe public it's too hard, wheras the friendly samba wizard is doable.

Re:I don't use samba anymore

[captaineo]

Unfortunately OSX still has some problems with NFS...

- mounts disappear occasionally for no apparent reason, and the automounter won't remount them, forcing me to reboot.

- NFS client performance is significantly worse than Linux (~20MB/sec vs ~100MB/sec reading from the same server over the same gigabit network)

- Some (very important to us) OSX apps have significant problems dealing with NFS paths. Final Cut Pro doesn't use symlinks properly, instead it hard-codes the target of the symlink into your project files, making it impossible to change where the link points without breaking your project. FCP also doesn't record projects on NFS shares in its "open recent" menu. (though DVD Studio Pro does).

And while I'm ranting about OSX filesystems:

- their FAT implementation has performance problems when dealing with very large directories. Copying thousands of film frames into a single directory starts quickly but then gets MUCH slower as the directory fills up. Linux's FAT driver does not exhibit this slowdown.

Re:I don't use samba anymore

[m50d]

I'm running slackware (well, sort of) and have no trouble with the number of steps it requires, but samba is far easier. Even compared to the steps you list, setting up samba on slack is easier - and I'd imagine the friendlier distros make samba much easier too.

Re:? troll

[curbion]

Let us analyse his post shall we.

um .. what has not being able to properly connect to a SAMBA share got to do with stability

Correct the first post does imply this is a stability issue

Tiger is very very stable i have not had one OS crash (a few programs have but they were built for 10.3 and the updated version run fine)

This is could be true , I do not know .Though it reads fine to me .

The fault here is in interoperability with a Microsoft SMB share (no such problem with NFS) and there is an easy work around (you just point to the share directly).

Awnsering the question of the issue , I see no problem.

Had this been about microsoft products not connecting to a SAMBA share properly .Then quite rightly, there would have been a hell of an uproar.

Too right , if MS had a problem with its own server software it would have been an issue . However since this is apple it is a diffrent issue. Apple fans get called zelouts all the time.Is it any wonder they get a bit annoyed when it hapens for no good reason. PS: I am not an apple user ,I use Win2k. I get rather annoyed at all the MS bashing but i see just as much Apple bashing by MS fanboys . Fair enough his final response was a bit angry , Understandable though . This hapens all the time on Apple Threads and it also happens on MS threads for the same reasons. The last response Mod that over-rated , i don't think he was trying to troll . He just lost his temper . It happens ! Now lets all just chill out (In certain languages the ? comes at the start of a sentence)

Does ANYBODY proofread anymore?

[SynapseLapse]

"Some installations are also finding that Windows workstations can't to log on to the Mac after upgrading to Mac OS X 10.4." ergh....

Re:? troll

[FidelCatsro]

Better apoligise i got a bit irate there , It happens and a big man can admit he was wrong .Yes i was in the wrong.
Though i was honestly not trying to troll in the grandparent , i do apoligise if it seems that way.
As an apple user (partly ,mostly i use Debian) I get sick fed up of people bringing out the "zelouts" card and it frankly turns me into a bit of a zelout , well i get defensive .
Its natural to defend something you spend a great deal of time on .
So perhaps im not a zelout , perhaps i was just annoyed at constantly hearing the line that all apple users are zelouts.

On the issue , yes apple has a problem here. It will be fixed soon hopefully for now there are work-arounds.

(btw yes it was a typ-o not a click-o i was posting as myself . :P this will teach me to proff read i guess)

(/Big-man-says-sorry mode)

Re:? troll

[curbion]

Hell you will probably get alot of crap for the first one .Not from me though .Nice to see someone on here admiting mistakes , though I do agree you had some right to be annoyed. Being an MS user and supporter is no easier here on slashdot. Sometimes people forget that the blade stabs both ways. If I were hearing constantly that just because I use windows (and enjoy using it.Oh and I do hear it) that i was a zelout i would be a bit angry . Kudos for the apology :)

History of SMB problems with OS X

[tyagiUK]

I first started using OS X in the early days of 10.2 (yes, a relative latecomer). This was when my wife bought an iBook (after some *ahem* guidance... read encouragement) for studies she was undertaking. When she wasn't working on it, I got to play and set to work integrating it with our home network.

The pain I had getting SMB to perform acceptably under 10.2 nearly put me off OS X. Basically, the way that 10.2 handled mounting network filesystems really sucked. It was unreliable and often left the system hanging with a spinning beachball (the Mac equivalent of an egg timer). Often, powering off was the only solution.

This was fortunately fixed later on in the 10.2 lifecycle with some networking updates. Things got much better from then on.

When I got my own iBook several months later, it arrived with 10.3. This release seemed to have a reasonably good SMB implementation, but the performance was truly sucky. File transfer speeds between the iBooks and my Linux-based Samba server were low, but at least mounting was reliable.

As 10.3 progressed, this problem went away and performance/reliability are currently both very good. It means I can use SMB between my Linux server and both iBook and Windows XP clients. All works just fine.

I am, however, considering a move to WebDAV for file sharing on the network. WebDAV is a nicely lightweight protocol and has the benefit of being an open standard. Most good implementations are open source too. There are also client libraries for most decent scripting/programming languages. The added benefit is that you can integrate the WebDAV server in to OS X to perform iSync backups of your system and do calendar sharing etc. All nice, geeky, stuff.

The only major problem I can see at the moment is that the way the WebDAV server interacts with the underlying filesystem is a bit complex, given that my server runs under Apache. The model it appears to assume is that the server will have a dedicated directory or area for WebDAV files, and not simply share out a user's home directory or a backup drive.

I do need to go and RTFM, however.

Re:History of SMB problems with OS X

[allgood2]

I haven't tested this thoroughly, but for one of our clients we activated WebDAV, mostly for future use--they were discussing adding a calendaring system to the internal use area of their website. What I noticed, is that after WebDAV was activated, each users' Sites folder had a folder called "WebDAV". Based on that, I'd say yes, WebDAV requires a dedicated directory, but that directory can and would be set-up for each users, and would be accessible through the Site's folder in the Users home Directory. Of course access to this folder could be placed in the Finder SideBar next to Documents, Music, Movies, so one wouldn't have to dig into the site's folder to get to it.

Re:History of SMB problems with OS X

[eluusive]

The pain I had getting SMB to perform acceptably under 10.2 nearly put me off OS X. Basically, the way that 10.2 handled mounting network filesystems really sucked. It was unreliable and often left the system hanging with a spinning beachball (the Mac equivalent of an egg timer). Often, powering off was the only solution.

This is really common, even in 10.3, and it REALLY pisses me off.

Re:? troll

[Daengbo]

Note to readers. First, FidelCatsro attempts to defend himself either anonymously or with a different account in the third person, but then fails, and then posts again as this user to further try to defend his post (you will know that this is FidelCatsro because the misspelling of zealot is identical to this post). I would mod the whole set down, but I've already posted once.

This troll needs some art added to his artis.

Re:Some predictions

[Daengbo]

5. As Seen on TV will chime in with ten highly-rated comments about "his" company.

Re:? troll

[FidelCatsro]

no , Curby is a freind of mine from another site .No way i could prove that to you , But since i already admited i screwed-up and posted an apoligy its all water under the bridge . I will say sorry to you aswell as obviously i have made a negative impresion on you in a moment of weakness.

Re:? troll

[curbion]

I am not Fidel ,I am however a friend . The typo of zealot is easy to explain, it's the last time i copy his spelling ,I should know better by now .

Re:Samba supports it

[October_30th]

might be against policy at your university for non-windows machines to authenticate.

I believe it's fundamentally something like that. I had to jump through hoops and sign waivers to get a special permission for my work Mac (it still doesn't authenticate with AD).

This is normal

[Bones3D_mac]

These kinds of things are the normal evolution of Mac OS X after each major release. Get new features and added speed as an initial tradeoff for lower stability and reliability. Anyone who has used the Mac OS since the early days of OS X should know this.

I'm sorry, but if you are installing Tiger onto a mission-critical system, you deserve the problems you get. Give the software time to mature before rushing to employ it in your networks.

Should all new software have bugs?

[guet]

Perhaps they need to do more automatic regression testing (daily) on each build then?

I think the car analogy is (for once : ) a good one. We have come to expect failure from Software, and that shouldn't be the case - it should be very rare, not inevitable with each new release. They did rush the release of Tiger, and certain things suffered for it. Yes they will probably fix it quickly, but it'd be nice if they had a more extensive testing program, with sufficient time alllotted to do the QA work, for catching regressions like this.

Re:Should all new software have bugs?

[Zero Sum]

It would be even nicer if there was a way to uninstall a package once installed. I've ben looking for 6 days and can't find one.

I think the occasional bug is forgivable, but relesing it without an uninstall facility? If they truly did this it is a bit think....

Same problems on FreeBSD/amd64

[koinu]

I've got problems with getting the file listing (with smbclient) on FreeBSD on the amd64-platform. I can tell you that it's a problem with samba and not smbfs. I've verified this.

This happens when I try to connect with some unpatched MS-Windows XP machines.

yeah man why can't macs say something useful

[commodoresloat]

Like "Fatal Exception Error 0028:C156DAD1"

Can't connect Windows - Tiger

[nicuramar]

Yeah, I have that problem. I can connect from Tiger to Windows (although only sometimes from the Finder; otherwise I have to use smbclient), and from/to my Linux server.
I can connect from/to the Linux server from Windows as well, but not from Windows to the Mac. It just keeps on failing no matter what.

Re:Samba supports it

[Dolda2000]

You can certainly authenticate non-Windows clients against a Windows server. You can, however, not authenticate Windows clients against a non-Windows server.

AD is, amazingly enough, (almost) all standard LDAP and Kerberos. The only "extension" of weight that they have made is the use the PAC field of a Kerberos ticket to include authorization data (IIRC, this includes the group SIDs of the principal).

The thing is that while the PAC field is standardized by the Krb5 RFC, so that all clients (Microsoft or not) can handle it and forward it correctly and everything, only a Microsoft Kerberos Key Distribution Center can actually produce one (the process of which includes a lot of proprietary Microsoft protocols). That's why you need a Microsoft server in an AD setup.

My take on this is that Microsoft is already sure of their desktop install base, and can practice some interoperability on the surface, while locking the server market into using the Windows Server System, since the server market is where they are feeling a bit more insecure.

Aha! Bonjour!

[Anonymous Coward]

They KNEW this was going to happen, and that they couldn't fix it... so THAT's why they released Bonjour for Windows! Ha!

Re:Aha! Bonjour!

[Chucker23N]

Except Bonjour has nothing to do with file sharing, nor is it on the same layer. You could implement file sharing on top of Bonjour, and I personally wish someone would, but Bonjour as released by Apple doesn't do anything to that regard at all.

An addendum to this

[BitterAndDrunk]

Additionally, sometimes beta testing doesn't reveal the bugs.

I did application development for a while in my past life, and I can't tell you how many times I fully tested a system on 3 different computers without a hitch. Then when I rolled it out to the remainder of the company, other computers simply couldn't run it as intended without me sitting down at their machines and troubleshooting.

Part of that is I'm simply not a computer scientist and most of my programming work is. . . infantile compared to those with a 4 year Comp Sci degree.

But a significant piece is still that the configuration variability (even within a company with relatively strict procedures with computers) wreaked havoc with something as high level as VB applications.

When you're addressing lower level functions (e.g. networking protocols) the problem can be more pronounced.

The myth of perfection

[tgibbs]

Is it too much to ask that vendors use beta versions of their own software in-house for a month before they release it? Is it too much to ask that they ship the software to a small number of beta testers before the final release in order to find those wrinkles and iron them out?

This is a common complaint heard about all kinds of products from cars to drugs. What it reflects is ignorance of the statistics of testing. By necessity, testing must be done on a pool of people that is orders of magnitude smaller than the final pool of users (a test on everybody is not a test, it is a product roll-out ). So let us say that you beta test on 1,000 people and roll the product out to a million. Then you will have about a 35% chance of missing a problem that affects 1 person in 1,000. On roll-out, each such problem translates into 1,000 people with problems.

No Admin worth their salt installs a new OS

[Beebos]

If an "admin" installs a brand spanking new OS immediately after release, that admin should have his pocket protector taken away from him. Particularly if one is working in a business or other mission critical environment, installing new OS without giving time for new bugs to be discovered and addressed is a sure sign incompetance.

Re:No Admin worth their salt installs a new OS

[subtropolis]

unless they're having these problems while testing on their own networks.

Re:No Admin worth their salt installs a new OS

[caseih]

This is surely true, although Apple apparently does not think so. Recently I received two e-mails from Apple's development team regarding bug reports that I and others have filed for Panther Server regarding critical OpenLDAP bugs. In a nutshell the e-mails said, "we think the bugs don't exist in Tiger Server. Please upgrade to Tiger Server and tell us if this is the case." I was stunned. I sent them a strongly worded response to tell them that this was not acceptable. Apple just doesn't yet understand what it takes to produce Enterprise software. We need very long support lifetimes (3-5 years minimum) and upgrading major OS versions outside of normal hardware replacement cycles (with proper testing) is *never* done except in extraordinary circumstances. Right now I am very unhappy with Apple. Does anyone even know what the life expentancy of Panther Server is? What about Tiger? I can't find this information anywhere and Apple has not yet responded to my queries. Judging by the terrible LDAP problems I had with OS 10.3 (not fixed until 10.3.9!) I am in no hurry to put Tiger Server into production. I learned my lesson the hard way.

Now that AFP support under linux is much better, I'm almost certainly going to go back to Linux for my main file servers. At least it is a known quantity.

Re:No Admin worth their salt installs a new OS

[Absentminded-Artist]

This is no different than Microsoft's responses, or any software vendor for that matter, PC or otherwise. So don't think Apple has cornered the market on this sort of mentality. This is standard business procedure for most of the software companies I've dealt with. Palm doesn't even offer the ability to upgrade the OS on your PDA and expects you to buy an entirely new PDA to get bug fixes. If I were you I'd try to escalate that issue and get a manager involved. That worker bee you dealt with was a bit clueless. Still, I think you are right in that Apple the corporation may not fully understand yet what it takes to produce Enterprise software. They've only been targeting Enterprise for a few years now...

Something did occur to me when I read your comments, though. I wonder how much of IT complaints are centered around not being used to the Apple way of doing things? I can't speak for you, but a lot of IT guys I know are trained/conditioned to do things the PC way and get frustrated with Apple not because Apple's solutions don't work but because Apple's solutions aren't familiar to them. Maybe there is a cultural difference involved with their frustration? I hear them complain about Apple as if the Wintel solution has never been without it's bugs, glitches, or pains.

Apple's SMB implementation hasn't been the smoothest process. Lot's of weird glitches and performance differences depending on which flavor of Windows you are using. I can see why you'd be frustrated to have to pay for upgrades just to get a fix. I just wish that software companies as a whole didn't do that to their customers.

Why SMB?

[rsax]

Has anyone else here compared netatalk with Samba? Which one is faster and offers the least amount of hassles?

Re:Why SMB?

[gullevek]

netatalk can use all the fancy mac characters like / and : it might be a bit faster, but it was for me a hassle to setup because the main 2 line has no proper japanese for old mac os 9 clients ...

Re:Finder and Linux Samba shares

[Anonymous Coward]

well, it has the overhead of the progress bar that
people are addicted to. In terminal , you just do a 'ditto' and then wait unless you chose verbose.

Re:So, what!

[nurmr]

Reports are that this layout bug has already been fixed in MS IE7.

Re:So, what!

[LurkerXXX]

Just to be fair, Windows servers have supported Appletalk sharing for years (since the NT4 days at least, I can't remember if if was in 3.x or not, it's been too long). They haven't had a client, but they have had a server. But yes, it would be nice if they'd add a client rather than make you use 3rd party clients.

I'm very surprised...

[h2d2]

...at how no one has come forward to blame MS and Windows for this problem.

hmmm... must be Mother's Day business!

Re:I'm very surprised...

[corrosive_nf]

if you had read the article you would see that it happens with linux shares as well.

Wait a minute...

[catdevnull]

Hey, I like Macs. I think Apple rules the roost in the OS world, etc. But hey, reality check:
SysAdmin Rule #1: If you depend on it, and it works fine the way it is, don't mess with it. [If it ain't broke...]
SysAdmin Rule #2: If you want to mess with it, test it before deploying it.

Why the hell did people install a .0 release and expect that it would not be without bugs? I say if any sysadmins out there were silly enough to make a hasty upgrade before testing (ignoring the above caveats) they deserve the problems they're experiencing.

We waited to deploy WinXP until the first service pack was released--and that saved our ass. I think it's ignorant to ignore that principle on the Mac side as well--esp. with a major update.

Early adopters are unpaid beta testers. Congratulations--you found the bugs!

Apple or not...

[phillymjs]

...you're a fool and deserve everything you get if you put a week-old OS on production hardware without doing non-production testing or having a fall-back.

If you insist, however, do it right. Prep a build of the new OS and put it on its own hard drive in the machine of your one or two most clueful end users. Let them beat on it for a while and document their problems/questions as they try to do their work. Once in a while go through the list and address their fixable issues. If they happen upon a show-stopper, they simply boot from the drive with the old build on it and use that until the next service release appears. Then you apply it, and test again. Repeat as necessary until the number of issues is low enough that you can confidently deploy the new OS build to all end users.

I have used this technique to great effect at several of my Mac clients, though I don't even consider giving them the newest OS until the .2 or .3 service releases have been out for a few weeks. A couple of my clients used to question this conservative method until some renegade users bought and installed Panther right after its release (without authorization from anyone) and ended up being basically unable to work until I reverted them to the standard OS/applications build.

As for OS X Server, that gets tested in my company's lab and on my bench at home from the day we get it, but it doesn't get rolled out anywhere until .4, and even then we clone the old drive to a FireWire drive before upgrading, just to be safe.

~Philly

Exception

[XanC]

What if it's Sarge? I think I'll be pretty OK rolling that out when it's released...

Re:Exception

[Bishop]

There are two problems:

1) you have no idea how much testing Sarge has actually undergone. For all we know only 5% of users are using Debian/Sarge on a regular basis. While in theory any package in Sarge should have gone through two weeks of Sid testing first, there have been bugs in Sarge packages.

2) Sarge may be the best release ever, but have you tested it in your environment? Is the new version of an application going to be able to import your existing data?

Regardless of the quality of the software a new release must always be tested first.

Does it matter?

[jleq]

99.9% of admins who know how to do their jobs correctly didn't go out and buy Tiger the first day, but chose to wait until a few bugs were worked out and the OS was generally seen as in good condition for mass-use. 99.9% of admins are casually going about their job instead of frantically trying to fix a problem that didn't need to be created in the first place.

Re:Does it matter?

[krray]

I guess I'm the 00.1% of the admins then. I went out and bought Tiger -- well, the SECOND day. :) Beyond that -- after doing normal backup routines I installed Tiger to external Firewire drive for testing. On various systems. Fun as heck to play with, and looking forward to rolling it in production.

But quickly retreated upon finding that my current backup software (Retrospect) doesn't work. I've used BRU previously -- and will take another look at it (it is supported already). The -36 Finder error tying into various Linux Samba boxes didn't help (though I haven't messed with the config file yet -- I've so far assumed it was a potential problem on MY end :). On a couple of boxes I've seen the Spotlight hang-up with the SystemUIServer process pegging the CPU at 100%.

Buying it now or in 3 months or 6 months won't change the price (most likely). It's still $129 -- and well worth it NOW to enjoy playing with it (still installed and used on a couple of temp portable drives). When all the issues go away -- I'll be ready to roll it into production overnight. Will you? :)

Re:I don't think they've ever fixed this.

[eviltypeguy]

Saynig it's "broken" is hogwash. Saying it's "broken in some circumstances" would be correct. The university I attend has had macs connecting to SMB shares and the like for at least the past year or so now without any issues. So, thank you very much, I don't want it disabled.

Works better than before for me

[__aaaaxm1522]

That's odd.

I'm running into the exact opposite scenario:

Under Tiger, SMB filesharing *screams* as compared to how it ran under Panther and earlier incarnations of OS X. I'm able to connect to my samba fileshare on my Linux box, and my Win XP box, without any trouble whatsoever.

In the past, I was always able to connect, but file transfers were dog-slow. They seem normal now.

Go figure.

Re:Works better than before for me

[tf23]

Same here - at work (Win2k3svr ADR domain, misc linux -redhat 8/9/FC2/3 smb/nfs servers) and home (fedora core3 - samba-3.0.10-1.fc3).

Re:I don't think they've ever fixed this.

[MacDaffy]

DING-DING-DING! SMB is a moving target. Microsoft has placed a number of potholes in what SHOULD be an open standard (delineated in RFC 1001 and 1002). There is a number of reserved, undocumented and falsely documented fields in the CIFS protocol that are used as stumbling blocks to seamless SMB connectivity.

I wouldn't be surprised to find a recent Windows update somewhere at the root of Tiger's SMB troubles. There may be a bug in Apple's code, but I doubt it.

Lesson learned.

[vmfedor]

It doesn't matter who makes the software, you never, ever run a production server/workstation with a brand new operating system for at least a little while past the release date. Even if it's Apple. :)

seen this before...

[rmallico]

Back in the mid/late 90's Apple had the same type of issue... The MAC could not enumerate AppleShare volumes on NT servers... don't remember the exact problem but it was a bug on the Apple site... just remember the pro-MAC folks were slamming MS (on boards and with the company i supported our 8 MAC users (as opposed to 230 MS users)... Microsoft actually issued the fix and surprised even me... (at the time i was going from a novell background to winnt)

Re:seen this before...

[cliffy2000]

What does the Media Access Control address have to do with this?

(Macintosh is abbreviated Mac, not MAC.)

Re:seen this before...

[rmallico]

uh... it does not... read the flow of the sentence... who the hell made you the resident spell/grammar checker?

Re:Some predictions

[Anonymous Coward]

I have a theory regarding ASOTV. Basically, we can all agree that (s)he is either an Apple employee or a very skilled troll. The problem with the Apple employee conclusion, and the reason he draws so much flak as of late, is that (1) Apple is secretive and tends to sack leakers as a general proposition; (2) (s)he makes no secret as to his/her "employer" with gratuituous use of the royal we; (3) while (s)he is usually correct with Apple insider-type information, (s)he does slip up sometimes.

So here's my take. I think (s)he does work for Apple. Now, any ordinary Apple employee would not have held onto his/her job for so long after so much Slashdot discourse, due to (1). But what we have here is no ordinary employee.

Who does Apple want to sell Macs to right now? Everybody, sure. But some folks are more amenable to switching than others. Windows users in general hate their computers, but not because of Windows insamuch as they hate the Microsoft 2-year upgrade cycle; they loathe having to "relearn their computer" at those fixed intervals. Luna and Outlook 2003 (yuck) notwithstanding, Windows/Office transitions are a hell of a lot easier to swallow for your grandma than switching to OS X. There isn't much Apple can do about that; we, their customers, want them to make NeXTSTEP++, not Longhorn or Blackcomb. Heaven forbid they ever start trying to "bend" OS X to be more Windows-like to get the switchers.

So your mom won't necessarily buy a Mac, and Apple cannot change this. But who will buy a Mac? Linux users, BSD users, Unix users in general. Apple does, in fact, occupy the space in the market that once belonged to SGI. Like SGI, they provide a system with (proprietary but damn good) value-adds for niche markets. But unlike SGI, their machines and software are affordable, often more so than the Wintel platform, and those add-ons (Aqua, Quartz, Spotlight, Finder when it behaves) make the system amenable to users of all skill levels, not just people with root. All the better to make root switch. And if root switches, grandma will eventually follow, because root makes software which grandma will use. Or so goes the theory. (Yes, even Linus uses a Mac now, but not for OS X.)

So here's my theory: Since root's /pub is Slashdot, ASOTV has likely been given a second job by Apple management. Her/His task is to try and convince /. users, by destruction of FUD and misconceptions regarding OS X, to give Apple a fair shake. So his activities are sanctioned, perhaps secretly, by Apple; this is why he hasn't been beheaded. If you think that's a crazy proposition, try to put yourself in the shoes of upper-level management in Cupertino. You need to get Unix folks to switch to Mac, because the Windows people keep saying "Windows has more software than Mac" (never mind that 60+% is malware). Meanwhile, people on Slashdot keep posting Apple FUD from the late '90s, which works quite effectively against your goal. Solution? Deploy a secret agent.

What about (1)? If ASOTV really worked for Apple, and needs not to conceal that fact, as in (2), why the secrecy? Because while Apple may want ASOTV to talk, they don't want their other employees to. This is the essence of the secret agent! And this secret agent probably is somewhere between PR and engineering, which is why his/her information is sometimes stale, which would explain (3).

So, I say, let him/her do her job. Since most Slashdot users bash Microsoft on a daily basis, then either resume using or dual-boot into Windows XP, I think there is some nobility in the goal of getting those folks to give the Mac a spin. So, there you go... take it or leave it.

"Of all those in the army close to the commander none is more in

The Mac OS X 'wait cursor'

[mpaque]

Actually, it's not just the foreground app. The wait cursor indicates that whatever app that owns the window currently under the mouse cursor has had pending, unprocessed events for over three seconds.

You can still switch to another application. Swinging the cursor over a window of a background app that was unresponsive will give you quick feedback in the form of the wait cursor if that app is still unresponsive.

Re:The Mac OS X 'wait cursor'

[mpaque]

This turns out not to be the case.

The window server event system inspects the state of the connection owning the topmost non-transperent window content under the mouse cursor hot spot. If that connection indicates the app it belongs to has not processed events within the allowed delay time (3 seconds), then the wait cursor is presented, as a UI cue that the app under the cursor is not responsive.

This overrides the cursor setting for the app that is foreground or 'has focus'.

Re:The Mac OS X 'wait cursor'

[mpaque]

Mac OS X is a 'click to focus' system, not a 'point to focus' system.

So far I'm having the opposite experience

[NtroP]

With 10.2 and Panther, getting client to successfully bind and work with Active Directory to something akin to VooDoo and several other flavors of black magic. That being said, when we did a thorough audit and clean up of Active Directory (Sites and Services, DNS, etc.) most of the problems disappeared, but there were often little things we did to increase our odds of things working smoothly.

The other day a colleague of mine installed Tiger on his laptop (he never had it bound before, just connected to whatever shares with Cmd-K, etc.). He asked about using his AD credentials to log on. I told him "Sure, we just need to bind it to AD, do a few tweaks and anyone with an AD account could log in, just like Windows." Meanwhile, I was mentally crossing my fingers that there wouldn't be any new tweaks that needed to be learned.

So I pointed him to Utilities/Directory Access and had him click the Active Directory option, put in his domain (this is where I would usually start my VooDoo dances with the "advanced" options -- but I thought, "what the hell, lets give it a shot") click on Bind. It asked for a domain admin account, which I entered, and it bound without a hitch (I about fainted). I had him reboot (just to make sure) and then had him log in with his AD account. I worked beautifully, including mounting his home directory off our Win2K server. This had NEVER worked without tweaking for us under panther (although with a little tweaking under 10.2.8+ it worked fine). We transfered files, which went smoothly and quickly, and we looked around the network a bit.

Although I haven't thoroughly tested it yet, I'd say my initial experience with Tiger and SMB/AD has been great. That being said, MOST of our problems with Macs using our AD domain has been Windows-related (missing DNS entries, Sites-and-Services borked, or WINS not working/configured right, etc). Hearing about problems like this after a major change doesn't exactly surprise me, and I'm willing to cut Apple a bit of slack here. They are dealing with a reverse-engeneered protocol on networks where it is very likely that AD isn't in pristine or "best-practices" condition.

We have 35 sites using AD right now in our domain, and the migration from NT4 to Win2K/AD was a learning experience, to say the least. We've learned a lot in the process and, we've found that if you mess up something in AD in the beginning, it's damn near impossible to cleanly remove or fix it. I suspect that there are a lot of installations out there that still have AD ghosts hanging around that make 3rd-party integration a crap-shoot at best. What apple needs to work on is improving their tolerance for broken AD implementations, like windows does.

Of course, if MS would publish the full SMB/AD protocol it would be easier.

Re:I don't think they've ever fixed this.

[fluffy99]

Actually one of the recent MS updates did break SMB connectivity to many appliances. I now have a Snap Server 4000 that I can't use unless I remove that critical MS patch. Snap knows about the issue and thinks its a great opportunity to blackmail owners into buying a software update. Bloody thieves.

I'm not sure this is the whole explanation as people have reported problems accessing files on Linux servers. In this case, its samba talking to samba.

Re:Misleading subject.

[jmelloy]

I can attest that Azureus is working perfectly.

VLC has a couple minor problems, though.

RE: Apple in the Enterprise

[King_TJ]

Well, you're DEFINITELY correct about this. It was only a scant few months ago that Apple finally put up a page on their web site geared towards the Enterprise customer. This is one area they're really only starting to dabble in right now.

Traditionally, Apple's server products were really only purchased by *departments* running all Macs.... so basically, "workgroup servers". In these scenarios, it's really not unreasonable to tell your group of 5-15 artists/designers that "Hey, as part of our troubleshooting - we'd like you to try swapping out your server with the latest release." Might be a bit frustrating and time-consuming, but hopefully less so than the error you're trying so hard to fix.

But yeah, this is *nothing* like the way things need to be handled in large, corporate settings. I'm not even sure it's fair to be "very unhappy with Apple" for any of this, though? Anyone making the decision to purchase Apple server products for use in an enterprise setting should have taken all of this into consideration first, IMHO. This is exactly why the company I used to work for skipped over the XServe products, despite thinking the ease of administration might be a huge plus. Apple is not yet really ready for the "enterprise", except in the ways another free Unix like BSD or Linux is. Stable, yes? Reliable, yes? Guaranteed fast problem resolution and long-term support for existing software versions, uh.... who knows, really?

Solution

[aristotle-dude]

Delete your SMB keychain entries and/or enable password encryption on your SMB server.

Re:Solution

[Guy Harris]

and/or enable password encryption on your SMB server

Note that Microsoft changed the "Windows OT" (95,98,Me) line in W98 and changed the "Windows NT" (NT 3.x, NT 4.0, W2K, WXP, WServer2K3) line in NT 4.0 SP3 so that, by default, they refuse to connect to servers that don't do encrypted passwords. There's a Registry entry you have to change to enable that - but they warn you that this means you're sending your password over the wire in a form that anybody capturing network traffic can see:

WARNING: Enabling this will allow unencrypted (plain text) passwords to be sent across the network when authenticating to an SMB server that requests this option. This can lessen the overall security of an environment and should only be done after careful consideration of the consequences of plain text passwords in your specific environment.

OSX and Windows

[orionware]

In our office, there are 4 macs and we could never get any of them to attach to the windows shares with any predictability. The resident mac gurus couldn't figure out how to get the macs to mount the windows shares and stay mounted.

If the machines were on a different network but routed internally, the macs wouldn't connect ever. The 5 people running linux as a desktop simply used the samba GUI and voila, no problem.

Since only 2 mac people were using it for creative purposes, the other three folks eventually asked for a PC so they could use the network shares reliabily.

Shame really that Apple can't get it together. Seems simple enough since they are using SAMBA for Christ's sake.

Re:OSX and Windows

[orionware]

Well. While there might be a grain of truth to that as I can not tell how much they really know or not. We've been able to get the linux boxes, which also use samba to connect to windows shares, to work just fine.

After reading the article about problems with OSX, I'm somewhat inclined to believe that it's a software issue with X... No?

Re:samba

[Guy Harris]

doesnt OSX use samba for SMB?

SMB server, yes.

SMB client, no. The SMB client was originally based on the FreeBSD smbfs, but has had a lot of additional work done.

Re:Samba supports it

[rodac]

Wrong, or our information is outdated.
There are multiple AD implementations that allow Microsoft clients to join to and authenticate to as if they were w2k3 AD.
See www.padl.com for one of the first implementations of one such AD. Samba4 also supports this but is a different implementation.

Fix that worked for me

[cavehamster]

I upgraded to Tiger, and attempted to connect to a Debian file server where I store most of my data. Could not connect, no matter what I did.

As it turns out, Apple now encrypts the password presented to the file server. I don't know how to turn that off, but I do know how to fix it so you can get into your server, provided you have root on it.

First off, set:

encrypt passwords = yes

In your smb.conf file.

This alone may fix you. I had to go a step further and do a 'smpasswd username' in order for Samba to actually authenticate me.

At this point, connecting to either smb://server/username or smb://server/ worked just fine to get to the username share.

Also note that Tiger has encrypt passwords turned on itself, so you'll need your windows clients to encrypt passwords to get to it.

Personally, the only reason I turned off encrypted passwords was so my mac could get to my file shares, so this is an OK workaround for me.

Re:So, what!

[cancer4xmas]

Windows does support AFP (2k server, you can certainly make a AFP share) - but it's a 15 year old protocol that was supposed to be done away with 10 years ago, and still it lingers. Most _networks_ don't support AFP, because you don't want to route the protocol because it's so chatty. Get your facts right.

Weirdness of the SMB Break

[DVant]

It's a truly strange break. It had me thinking just like the bowl of petunias, "Oh no not again." But unlike the SMB breaks I have seen in various 10.3 updates this one is far stranger.

In our office we have four machines: A Debian Sarge Box, two Win XP boxes and a current Al Powerbook (running tiger). The tiger running machine can connect to both the WinXP boxes fine, but the Debian box, which under 10.3 was used every day, no longer talks to the Powerbook.

Yey, wait for a month and I'm sure that we'll see an update that fixes half the problem and screws up something else.

Apart from my SMB woes, Tiger is the most finished .0 release of Mac OS X to date. Reasonably impressed. (Mutters "Bloody Samba")

Re:Weirdness of the SMB Break

[Guy Harris]

In our office we have four machines: A Debian Sarge Box, two Win XP boxes and a current Al Powerbook (running tiger). The tiger running machine can connect to both the WinXP boxes fine, but the Debian box, which under 10.3 was used every day, no longer talks to the Powerbook.

No longer talks to the PowerBook with the Debian box being the SMB server, or no longer talks to the PowerBook with the PowerBook as the SMB server (e.g., using Linux smbfs or cifsfs)?

If the Debian box is the server, do you have encrypted passwords enabled? If not, try enabling them (which, apparently, also makes it easier to connect to the server from W98 and later, and NT4SP3 and later, according to some Microsoft knowledge base items, as they made those OSes not, by default, connect to servers that don't support encrypted passwords, requiring you to tweak a flag in the registry to support it).

Re:Weirdness of the SMB Break

[DVant]

The Debian box is the SMB server. Will try encryped passwords tomorrow. Thanks Guy.

SMB has been flaky in OS X for ages

[onlyjoking]

STILL the same problems with SMB in OS X? How can Apple tout their "seamless" integration with Windows when this has been going on for so long? How long before they finally sort out their appalling SMB support?

Fix that worked for all of my users

[tinpan]

This week, some Macs at my workplace were unable to connect to smb shares on Windows boxes. All were showing error -36. These were mainly 10.4 machines, but also 10.3 and 10.2 machines that could connect before. And all of the users in question could connect from other Macs via browsing the Windows domain. I sniffed the successful connections and found that they all had addresses like "cifs://windows.server.address" instead of the expected "smb://windows.server.address". When I had the users connect from their "broken" Macs using Cmd+K and "cifs://", it worked for all of them. Don't know why, but it does.

Re:So, what!

[tinpan]

Isn't it AppleTalk that's chatty? From what I understand and seen, AFP over TCP/IP is not chatty.

Indeed...

[ArbitraryConstant]

I'm expected to do better than that with command line tools, and it's not that hard.

The error codes used might work in strerror(3), but it can't be that hard to add a similar function to handle Apple codes.

Re:So, what!

[jimmyharris]

You seem to be confusing AFP with AppleTalk. AFP can use AppleTalk or TCP/IP for its network transport - certainly AppleTalk is disappearing from network routers due to its verbosity but AFP over TCP/IP works perfectly on all the networks I've seen.

Microsoft has limited support for AFP with its Windows services for Macintosh but they don't support AFP 3.1 (long file name, files > 2GB) at all. You have to buy an expensive commercial product to get these features (or use Free/free Netatalk on a *NIX OS).

Maybe next time you should be a little slower to tell other to get their facts right.

Pissing contest anyone?

[rmallico]

http://support.microsoft.com/default.aspx?scid=kb; en-us;178364 CAUSE... (read very carefully here) RESOLUTION... (FIXED) Anytime you want to drop trousers and have a good ole' career achievement pissing contest you know where I can be found...

Re:What's the big deal?

[Guy Harris]

Plus Apple is passing a -d 10 to swat when it's called?!? -d is not documented in the swat man page.

It's documented in the man page on my Panther box:

-d|--debug=debuglevel
debuglevel is an integer from 0 to 10. The default value if this parameter is not specified is zero.

The higher this value, the more detail will be logged to the log files about the activities of the server. ...

Re:So, what!

[MemoryDragon]

which still is vaporware, nevertheless it just took 7-8 years to fix this, which is hilarious.

Also, it doesn't fail 100%

[sunbane]

Mounting samba directories from Linux w/ Finder under Tiger worked just fine for me. Used my stored password from my keychain and let me copy my pictures to my family website just like I'd never upgraded my os. Since it doesn't fail 100% for everyone, that makes it easier to miss... they likely have some certain configuration that causes the problem.

Wow...

[FredFnord]

You don't require much in the way of 'proof', do you?

I've made bigger mistakes than that, misstatements and such, on Slashdot, and I've been programming and babysitting (IT) Macs since, quite possibly, before you were born. In any case, since before THEY were (officially) born. I could certainly see myself making this little semantic error, although I know all too well how the spinning beachball works, and all the stupid little tricks I had to play to get it to never* happen in my last piece of software.

-fred

*Okay, well, never that I've observed yet.

Fix for the problem

[Crow Song]

I had believed that this problem was on my end. I had upgraded to Tiger and lost connectivity with my Windows 2003 shares. Exactly the problem everyone here has been talking about. When the 10.4.1 upgrade did not fix this, the sys admin and I took a harder look. It turned out to be the firewall on the Windows 2003 Server. Turned it off - and bang, I was back to normal.