invalid checksum

Question

Level 1

0 points

invalid checksum

After waiting for some of Apple's huge update files like the 180.8 Mb MacOSXUpdCombo10.4.11PPC.dmg to download, when I try to launch the disk image Warning message keeps appearing that reads -

Warning

The following disk image failed to mount

Image - MacOSXUpdCombo10.4.11PPC.dmg

Reason - invalid checksum

So, it has been Impossible for me to apply any updates that open with this warning, which in the case of this MacOSXUpdCombo10.4.11PPC.dmg is EVERY download of it, since it will Not Open

Sawtooth AGP capable of being upgraded with a Dual CPU upgrade, Mac OS X (10.4.10), Upgraded with a Powerlogix 800 MHZ single CPU, ATi 9800 Mac Pro 128Mb AGP video card & 1Gb Ram

Posted on Oct 1, 2008 2:45 PM

Reply

Answer 1

BDAqua

Level 10

239,045 points

Oct 1, 2008 3:33 PM in response to AMCJ

Hi, how are you downloading it? Through Software Update, or directly...

The combo update for PowerPC-based Macs...

http://www.apple.com/support/downloads/macosx10411comboupdateppc.html

If you haven't lately, "Try Disk Utility

1. Insert the Mac OS X Install disc that came with your computer, then restart the computer while holding the C key.
2. When your computer finishes starting up from the disc, choose Disk Utility from the Installer menu. (In Mac OS X 10.4 or later, you must select your language first.)
Important: Do not click Continue in the first screen of the Installer. If you do, you must restart from the disc again to access Disk Utility.
3. Click the First Aid tab.
4. Click the disclosure triangle to the left of the hard drive icon to display the names of your hard disk volumes and partitions.
5. Select your Mac OS X volume.
6. Click Repair. Disk Utility checks and repairs the disk."

http://docs.info.apple.com/article.html?artnum=106214

Then Safe Boot from the HD, (holding Shift key down at bootup), run Disk Utility in Applications>Utilities, then highlight your drive, click on Repair Permissions, reboot when it finishes.

Reply

Answer 2

AMCJ Author

Level 1

0 points

Oct 1, 2008 4:59 PM in response to BDAqua

How would I know that the file is 180.8 Mb in size if I had not waited almost a half hour to download it onto my home computer?
I've seen this answer you've posted also posted on other users threads, so I must assume you are just copy and pasting the same info.
Is my issue identicle to theirs? I am not so sure.

I am on Verizon DSL's 768k service, which is slower than their 1.5 Mb/s DSL, but faster than dialup. So, the downloads take longer than high speed services. I'd imagine and have read elsewhere that the components that the signal are going through can possibly cause corruption. So, I had Verizon send me their newest modem to see if it would stop the checksum errors. It did not.

I have used Disk Utility often lately, as I also have been seeing Firefox, Safari and Photoshop CS either get the spinning wheel of death before a warning message says that the apps are not Responding, which I've usually resolved by Force Quiting the affected app to get to the Finder.

Then I run Disk Doctor and defragment with Techtool Pro. But, after a week or so of intensive Photoshop manipulation of hundreds maybe thousands of .jpg files, so they will be the appropriate size for a coin database I am building, the problems come up again.

I'm wondering if the Force Quiting is causing corruption.

I'm also wondering if a my Ram, 1) 512 and 2) 256 chips might be causing the problem. Can bad Ram still start up the system but cause an app, that is trying to access the bad sector, to become unResponsive/QUIT?

I am assuming that all of the changes to the file sizes and their being renamed tends to screw up Spotlights ability to find them after a while, so I've been using the aforementioned maintenance routine to get myself up and working again.

I believe I recall having tried to do an update to OS 10.4.11 from the updater service automatically. At some point during that attempt, the service announced that the update had failed. Of course I could now not access that volume/partition at all after the failure. So, I booted from another partition which has another 10.4.10 OS on it and re-installed a good copy of the 10.4.10 update. I was then able to use the volume to boot up again.

should I assume you are going to tell me that this Failure and subsequent reinstall may have worked, but is now the source of my problem?

Reply

Answer 3

BDAqua

Level 10

239,045 points

Oct 1, 2008 5:41 PM in response to AMCJ

How would I know that the file is 180.8 Mb in size if I had not waited almost a half hour to download it onto my home computer?

You can get it through Software Update>Download Only, which would tell you the size, but be by a different, sometimes flakier method... that's how.

Is my issue identicle to theirs? I am not so sure.

Since I dare not ask questions, I don't think I'll ever know. 😟

I've seen this answer you've posted also posted on other users threads, so I must assume you are just copy and pasting the same info.

Yes it was pasted, are you surprised that many people have the same problem? Also, I'm nearly blind and type with one or two fingers, so copying when I think it is the same issue & possibly the same cure... you bet, I'm going to do it.

And, we never did get to a cure, just the preliminary ground work to try to get some answers, and do a few things to insure that any future suggestions will go off as expected.

I do not take anybody's problem lightly, and do not go pasting willy-nilly.

Reply

Answer 4

Oct 1, 2008 6:35 PM in response to AMCJ

Unique.Designs wrote:

A checksum error indicates that the file that left Apple is not the same as when it arrived at your machine. Usually that means the file has been damaged in transit. But it could also indicate a hardware problem with your computer or somewhere on your network.

I had Verizon send me their newest modem to see if it would stop the checksum errors. It did not.

Do you have a firewall/router or an ethernet hub on your local network? I had a bad firewall/router once that was corrupting downloads. I replaced it and the problems went away. However, I also had a lot of problems with web sites where it wouldn't load some of the graphics or even the HTML text would be corrupted.

I have used Disk Utility often lately, as I also have been seeing Firefox, Safari and Photoshop CS either get the spinning wheel of death before a warning message says that the apps are not Responding, which I've usually resolved by Force Quiting the affected app to get to the Finder.

When you get the spinning gumdrop of death, look in the Console application from your Utilities folder and see if there's anything in either the console log or the system log. Particularly, you'd be looking for disk I/O errors. That could indicate bad blocks or a drive controller that's going bad.

Then I run Disk Doctor and defragment with Techtool Pro. But, after a week or so of intensive Photoshop manipulation of hundreds maybe thousands of .jpg files, so they will be the appropriate size for a coin database I am building, the problems come up again.

Since you have TechTool Pro, try testing the RAM and run a surface scan on the hard drive. Or just run all the hardware tests and see if it comes up with anything. You might want to run the RAM tests in a loop if that's possible, since they won't always show a problem the first time through.

I'm wondering if the Force Quiting is causing corruption.

I'm not sure, but I'd guess it's only a problem if the app is writing to the disk at the time you quit it. And even then, the kernel might intervene and make sure the file system is clean.

I'm also wondering if a my Ram, 1) 512 and 2) 256 chips might be causing the problem. Can bad Ram still start up the system but cause an app, that is trying to access the bad sector, to become unResponsive/QUIT?

Absolutely it could be RAM. Especially if it's in a place where the application's data, but not its code is getting corrupted. Bad RAM will cause all kinds of problems: kernel panics, apps that quit unexpectedly, hard drive corruption when corrupted memory structures get written out to disk, etc.

You could moving around, removing, and/or replacing the RAM chips and see if there's any difference.

I'd also look at the CPU upgrade you have. Perhaps it's failing. If it's easy to remove, try that and see what happens.

Finally, how much free hard drive space do you have? You need to keep enough space free for temp files and, especially, the swap files. I try to keep at least 8GB free. I've gotten by with as little as 4GB, but it makes me nervous. Keeping a bit of free space also allows the OS to move files around as it needs to in order to keep fragmentation down.

charlie

Reply

Answer 5

AMCJ Author

Level 1

0 points

Oct 1, 2008 8:19 PM in response to BDAqua

BDaqua,

I really did not mean to offend you. Having seen the exact reply to other post made me wonder why you had not addressed the entire set of thoughts i was posing.

I'd appreciate any questions you might ask that could lead me to think of what might be causing these checksum errors and the other issues I've posted.

I see your helpful replies to many posts and really do appreciate your even making the effort to address mine.

Thank you.

Charles, TIA for your complete reply.

Yes, I have a separate Linksys wired router in the chain. It has some sort of built in firewall, I believe.

I've read elsewhere that interruptions during download can cause corruption to occur. Does this include any actions I might be doing with my other apps? I always thought that background downloading meant I could do some email retreiving, surfing other web sights, etc. But, as I think about it, could the fact that even going to another sight downloads cache files to my HD, and so these files Must Be Interrupting the flow of the Update file. So, should All other Mail apps or automatic web surfer apps be disabled while most if not all updater files are being downloaded? And, Even avoid working in other apps, like Photoshop, etc?

Reply

Answer 6

BDAqua

Level 10

239,045 points

Oct 1, 2008 9:01 PM in response to AMCJ

So, should All other Mail apps or automatic web surfer apps be disabled while most if not all updater files are being downloaded? And, Even avoid working in other apps, like Photoshop, etc?

Always a good idea if you're having problems with downloads, and though I still don't know the means you used to download, I;ll mention that the Download option in Software Update seems far more prone to this timing issue.

Having seen the exact reply to other post made me wonder why you had not addressed the entire set of thoughts i was posing.

The Disk & then Permission Repairs suggestion is three fold in purpose, it's like looking both ways before you cross the street... for safety, secondly, it may point to other problems/solutions if many repairs are needed, thirdly, it may fix the problem. I hesitated to go further until I had more info, and was sure that we weren't going to cause more damage by proceeding.

Charlie is an excellent helper, far more knowledgable than I by far, so his suggestion should be taken. 🙂

Reply

Answer 7

AMCJ Author

Level 1

0 points

Oct 1, 2008 9:24 PM in response to Charles Minow

Charles,

I included you in my reply above to BDAqua.

Now to the rest of your replies.

I just checked out the Console for the 1st time ever and see what you mean, I think. I've copied some of the different messages in the console and am pasting them to see if the make any sense to anyone.

Adobe Photoshop 7.0(417,0xa000ed88) malloc: * Deallocation of a pointer not malloced: 0x907126a0; This could be a double free(), or free() called with the middle of an allocated block; Try setting environment variable MallocHelp to see tools to help debug
Sep 29 22:35:04 AMCJ-2 /Volumes/MacTiger/Applications/Adobe Photoshop 7/Adobe Photoshop 7.0/Contents/MacOS/Adobe Photoshop 7.0: The "IgnoreForEvents" window property is obsolete. Use CGSSetWindowTags with kCGSIgnoreForEventsTagBit flag instead. Set a break-point on CGSLogMessage to find out where this property is set.\n
THE ABOVE OCCURS 30+ TIMES IN SUCCESSION(NO TIME STAMP)

StuffIt Expander(422,0x1841e00) malloc: * error for object 0x582dd0: incorrect checksum for freed object - object was probably modified after being freed, break at szone_error to debug
StuffIt Expander(422,0x1841e00) malloc: * set a breakpoint in szone_error to debug
Sep 29 22:37:05 AMCJ-2 crashdump[423]: StuffIt Expander crashed
Sep 29 22:37:05 AMCJ-2 crashdump[423]: crash report written to: /Users/AMCJ/Library/Logs/CrashReporter/StuffIt Expander.crash.log
Sep 29 22:37:46 AMCJ-2 crashdump[427]: sh crashed
Sep 29 22:37:46 AMCJ-2 crashdump[427]: crash report written to: /Users/AMCJ/Library/Logs/CrashReporter/sh.crash.log
Sep 29 22:45:01 AMCJ-2 crashdump[435]: Adobe Photoshop CS crashed
Sep 29 22:45:03 AMCJ-2 crashdump[435]: crash report written to: /Users/AMCJ/Library/Logs/CrashReporter/Adobe Photoshop CS.crash.log
Sep 30 00:21:11 AMCJ-2 crashdump[461]: firefox-bin crashed
Sep 30 00:21:52 AMCJ-2 crashdump[461]: crash report written to: /Users/AMCJ/Library/Logs/CrashReporter/firefox-bin.crash.log
2008-09-30 00:41:14.133 SyndicationAgent[473] WARNING: BestCalendarDateFromString - can't interpret: 'Mon 29 Sep 2008 19:06:48 -800'
THE ABOVE OCCURS 50+ TIMES IN SUCCESSION A FEW THOUSANDSTH OF A SECOND APART

Sep 30 03:14:59 AMCJ-2 crashdump[501]: sh crashed
Sep 30 03:14:59 AMCJ-2 crashdump[501]: crash report written to: /Library/Logs/CrashReporter/sh.crash.log
2008-09-30 10:55:54.826 SystemUIServer[219] Could not find application for bundle ID: com.apple.iPhoto
Oct 1 00:15:59 AMCJ-2 crashdump[685]: firefox-bin crashed
Oct 1 00:16:34 AMCJ-2 crashdump[685]: crash report written to: /Users/AMCJ/Library/Logs/CrashReporter/firefox-bin.crash.log
Oct 1 16:13:16 AMCJ-2 /System/Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/Metadat a.framework/Versions/A/Support/mdimportserver: Corrupt JPEG data: 2 extraneous bytes before marker 0xe0\n

********
Disk Utility started.

2008-10-01 16:21:13.509 Disk Utility[952] * _NSAutoreleaseNoPool(): Object 0x399a30 of class NSPathStore2 autoreleased with no pool in place - just leaking
THE ABOVE OCCURS HUNDREDS OF TIMES IN SUCCESSION A FEW THOUSANDSTH OF A SECOND APART

Oct 1 16:34:25 AMCJ-2 /System/Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/Metadat a.framework/Versions/A/Support/mdimportserver: Corrupt JPEG data: 2 extraneous bytes before marker 0xe0\n

I did not see any I/O or Bad Block errors in this mess.

I had done a full disk surface scan of both of my drives with TTP4.6.1 yesterday. Both came up clean. So did every other test in the Suite. No File or Finder Info Errors either.

I'm starting to think that a bad RAM chip is causing these corruptions, EVEN though TTP reports that the chips are OK. TTP must not do the whole chip. If it misses a bad sector/block on the RA and reports it as good to go, then Photoshop, the HOG that it is might be accessing the bad spots that are deep into the RAM's memory blocks. I guess even bad memory can still be good enough to not cause a boot up to make the Broken Glass sound.

These Sawtooth boxes take PC100 and 13. I live 10 minutes from Data Memory Services DMS in Salem, NH. I can get a new 512 chip for $40.

As for the FREE HD space, the HD I'm booting from which is the Master is a Western Digital that came with this box. It's 19 Gb and has 7.97 Gb free. I bought is used from CA and it did not come with the bootable System Installer Disks. I'll have to buy a used Tiger Installer from ebay soon. The last OS I bought was OS9.

The slave drive I added is a 7200 RPM Cuda that I partitioned to Five 25+ Gb partitions. I like to isolate Work docs from System and Apps. Two partitions are Scratch disks.

Alan

Reply

Answer 8

AMCJ Author

Level 1

0 points

Oct 2, 2008 12:51 AM in response to Charles Minow

Charles,

I ran the TTP RAM test again and apparently I missed noticing that this one was coming up as failed.

I started pulling Ram and found that one of the 256 chips was readong as bad.

I just ordered a new 512 chip from DMS at $36 w/FREE S&H if you choose the USPS First Class option which doesn't come with Tracking or Postal Insurance.

I intend on picking it up anyway.

I'll post my results after a few days of running the box at my regular pace.

I also wiped out the more than 12,500 .DS_Store files for FREE with the download of EasyFind.

Thanks for the suggestions.

Alan

Reply

Answer 9

Oct 2, 2008 5:55 AM in response to AMCJ

Alan--

Unique.Designs wrote:
I ran the TTP RAM test again and apparently I missed noticing that this one was coming up as failed.

I started pulling Ram and found that one of the 256 chips was readong as bad.

Aha. That could be it. One more thing to try is to put the "bad" 256 chip in another slot and see if it tests bad there. If it test bad in different slots, then it's likely the chip. If it doesn't, then put the other 256 or the 512 in the same slot where the "bad" 256 was. If the "good" chips test bad in the different slot, then it's a problem with the slot itself. But $36 for a 512 isn't bad at all, so even if it is the slot, you can just try leaving it empty, I think.

charlie

Reply

Answer 10

Oct 2, 2008 6:12 AM in response to AMCJ

Alan--

With a few exceptions I'll note below, a look at your logs do suggest RAM problems.

Unique.Designs wrote:
Adobe Photoshop 7.0(417,0xa000ed88) malloc: * Deallocation of a pointer not malloced: 0x907126a0; This could be a double free(), or free() called with the middle of an allocated block; Try setting environment variable MallocHelp to see tools to help debug

This could either be a programming error on the part of Adobe, or it could be a bad section of memory. The problem being that most programming errors are memory errors, so they would kind of look alike. But the sheer number of crashes and the apparent "randomness" of them supports bad RAM.

Sep 29 22:35:04 AMCJ-2 /Volumes/MacTiger/Applications/Adobe Photoshop 7/Adobe Photoshop 7.0/Contents/MacOS/Adobe Photoshop 7.0: The "IgnoreForEvents" window property is obsolete. Use CGSSetWindowTags with kCGSIgnoreForEventsTagBit flag instead. Set a break-point on CGSLogMessage to find out where this property is set.\n
THE ABOVE OCCURS 30+ TIMES IN SUCCESSION(NO TIME STAMP)

Probably nothing to worry about. This is just a notice to the programmers that a certain API is obsolete and they should change it in their next update.

2008-09-30 00:41:14.133 SyndicationAgent[473] WARNING: BestCalendarDateFromString - can't interpret: 'Mon 29 Sep 2008 19:06:48 -800'
THE ABOVE OCCURS 50+ TIMES IN SUCCESSION A FEW THOUSANDSTH OF A SECOND APART

This is actually harmless. It's from the iTunes RSS feeds that are in the default Safari bookmarks. They're formatted improperly (missing a comma) and so this error gets dumped into the log. If you leave Safari running long enough, they'll recur about every half hour.

I'll have to buy a used Tiger Installer from ebay soon. The last OS I bought was OS9.

Yeah, you may need to re-install the OS (and your apps), depending on the level of corruption the bad RAM has caused. Just make sure it's the retail version, not the model-specific version (those are usually gray, and in plain sleeves).

charlie

Reply