What's New - Dec 19, 2000

By: Mike Kimball - Revised: 2006-07-03 devin


What's new for the Mac in the previous month:
  • General issues, workarounds and fixes
  • New Software and hardware
  • Software and hardware updates/patches


Disk First Aid 8.6 wrapper error reports and repairs: a follow-up

Following up on yesterday's item on Disk First Aid 8.6 and "wrapper errors":
  • Wrapper error depends on drive formatting utility? John Oswald found that, on his Macs, Disk First Aid only reported the wrapper error on drives that had been formatted with Drive Setup. It did not occur when checking drives formatted with SoftRaid. [Note: The Ask Al column noted below suggests that drive formatting software should be irrelevant to this problem.]
  • FireWire CD-RW drive "lost" after DFA repair Isaac Rivera claims that, after Disk First Aid repaired reported wrapper errors, Toast and Apple System Profiler no longer recognize that his APS FireWire CD-RW drive is connected to his dual-processor G4. It seemed to us unlikely that Disk First Aid's repairs are the cause, but we have since received an additional confirming report from Kevin Williams: "I had the exact same problem on a 400 MHz G4 running OS 9.0.4. My Que! FireWire CDRW vanished from the Apple System Profiler and Toast after correcting a wrapper error with DFA 8.6." Isaac has since informed us that his Epson Expression 1600 FireWire scanner is also no longer recognized.
Ask Al answers Today's Ask Al column tackles the subject of these wrapper errors. It notes (as we have also covered here before) that the wrapper is a basic component of all HFS Plus formatted drives. However, "a bug in one version of the wrapper System file startup code" can prevent the Mac from starting up from that disk - if the disk is sufficiently fragmented. The bug only exists in systems running Mac OS 9.0.x. Apple's fix is to replace the OS 9 buggy wrapper code with the code from Mac OS 8.6. Thus, Disk First Aid essentially downgrades your wrapper code. It does this as a preventative, whether or not you actually are in imminent danger of having the startup problem.

TIL article For more background, see TIL article #31056: Disk First Aid: "HFS Wrapper Partition Damaged" Error.

[At the moment, I have chosen the "if it ain't broke, don't fix it" option. I suppose I have little to fear, having no external devices myself, but.... -mk]

Monitor won't wake from sleep: the power cord connection is the cause?

In Bob LeVitus' most recent column at MacCentral, he writes:

      "Something strange has begun happening to my Mac (a G4/500) and monitor (17" Apple Studio Display). All of a sudden, the monitor won't wake up from sleep. My Mac is fine. If I turn the monitor off and back on again, everything is hunky dory. But after sleep, the monitor remains black unless I turn its power off/on. I've done everything I can think of -- trashed the Energy Saver preferences, trashed my Finder preferences, switched to Sleeper, zapped the PRAM -- but nothing seems to work."

Michael Johnson offered this potential solution:

      '"I was able to trace the problem to the power cord's connection to the G3. After fiddling with it, unplugging, replugging, etc., I was able to finally seat it in there just right and the monitor would respond properly again. Over the past year, this has happened 2-3 times and every time, this was the only thing that would fix it. The machine doesn't get moved a lot, but it seems that a little jiggling over the course of several months causes the plug to move just enough to display this symptom." [Note: it is possible that simply unplugging the monitor for awhile, letting all remaining charge dissipate, fixes things. This was a common fix for AppleVision 1710 problems, as covered on MacFixIt years ago.]

Bob also discovered that: "if I plug the monitor into any power source except the G4 itself, I don't have the sleep issue anymore."

Coincidentally, Jose Nieves just reported a similar wake-from-sleep problem with his ADC Apple Studio Display 17 connected to a G4 500 DP with a Radeon OEM board. He adds: "I have a dual monitor setup with a PCI Voodoo 5 driving a VGA NEC MultiSync 70. The problem appears sporadically and seems to happen more when the machine is in sleep mode for extended periods. I haven't been able to relate it to anything else. A restart is required to wake the monitor." Perhaps the power cord "solution" will work here as well.

[I have noted a related problem with both power cords - not that they have flakey behavior associated with sleep/wake, since we have that disabled on all the machines we support; but that the plugs and ports do not seat themselves very snugly on the backs of the monitors or the G3/G4, and have a tendency to be bumped/jostled loose, cutting the power. -mk]

Apple CPU Plugin: fixes G4 mouse freeze bug?

Apple CPU Plugin 2.7 is now posted on Apple's web site. Apple states: "This update is required for Power Mac G4 systems using Mac OS 9.0.4 or later. This update is not needed by and will not be installed on systems using Mac OS X Public Beta. This software updates the Apple CPU Plugin extension and provides more reliable performance from your Power Mac G4." As we mentioned previously, the update appears only applicable to dual-processor Power Mac G4s.
  • George Warner adds: "It sounds like CPU Plugin is supposed to fix the '1-more-click-and-you-die' bug in dual G4 Macs (see previous coverage). The bug occurs after the Mac has been sitting idle for a while - like about an hour. When you attempt to use the Mac, the mouse moves but the mouse button does little or nothing. Click the mouse a few times and the computer is hung."
  • Update: The Apple Support Discussions thread titled "USB/Pro Keyboard Problem" contains replies from Apple that confirm that CPU PlugIn 2.7 should fix this mouse freeze problem. (Thanks, Dana Keil.)
  • Update: Reader comments VersionTracker now indicate that this file will install on at least some single-processor G4 Macs. One suggestion is that it only installs if you do not have Mac OS X installed on any partition.
  • However, when Hib Halverson tried to install the Plugin on his single processor G4, he got a message that said: "The Apple CPU Plugin requires a Power Macintosh Dual Processor and cannot be installed on this computer." We found this as well - even though we currently have version 2.4 of the Plug-in installed and even though the Multiprocessing folder (where the extension is stored) is needed to avoid wake-from-sleep crashes even in single processor Macs (see previous coverage).
[I tried this update but it would not recognize my startup disk as available (the "switch disk" button was greyed out), only the blank disk partition I have reserved for a more final release of Mac OS X ;) But I have version 2.5 of the Apple CPU Plugins extension, and don't seem to have this problem. Most users have complained about version 2.4 (and before?). -mk]
MRJ security hole MRJ "security hole" item updated We added an update to yesterday's item on the MRJ security hole.

      Update: Partially contrary to the update we posted yesterday, you remain vulnerable to the security leak even if your hard drive has a name other than "Macintosh HD." It is more accurate to say that if your hard drive name can be guessed, you are vulnerable. And with "/" as an entry field in the applet, virtually any name may be vulnerable. (Thanks, Brent Neal and Ken Hinckley.)

[This bug has to do with "Inconsistent use of CODEBASE and ARCHIVE attributes" (though I'm no Java guru so don't ask me what that means). Yet another reason to avoid MSIE, which primarily uses MRJ. Sure, you could just turn off Java, but... Symantec's Java Accelerator for PowerPC plugin (Netscape's default) may be old and slow, but at least it is secure ;) Netscape's MRJ plugin is still way too buggy and unstable to use yet. -mk]

NOTE: source of all above [except parenthetical notes by mk] is www.macfixit.com.