Amiga 500 Office not in TOSEC

[prowler]

Originally Posted by HymnsomniaK

Great news!

This evening, I have restored all six virus-damaged data blocks on the Office disk, which has repaired the LUIS 2, Graph.info, Calc.info, Exemplos.info and File.info files.

The window which opens when you double-click on the disk's icon is much more impressive with all the icons showing up.

Unfortunately, I am unable yet to repair the Calc and Graph programs. Calc still has a corrupted data block (Disk Block 116), and Graph has two consecutive corrupted data blocks (Disk Blocks 918 and 919).

As I mentioned before, the Office.DMS file you dumped from your external floppy drive has captured different data for Disk Block 116, but whichever version is used, the Calc program gurus while loading.

There are no better versions of the corrupted data blocks in the RawRead images either.

I had been mapping the blocks on the Office disk when I discovered the filesystem errors affecting the Calc file and, having fixed these, hadn't quite resumed when I discovered how to repair the virus-damaged blocks.

Now that these have been restored, I shall return to mapping the blocks. I have a suspicion that there are some duplicate data blocks on this disk which may provide the key to fixing the Graph program at least.

Originally Posted by HymnsomniaK

If you ever need more dumps with a specific dumping software let me know.

Thanks, but I think it is probably better to try corrupted disks in different drives than with different dumping software. If you do have any more disk drives, please try them with DMS and DiskRipper. (The RawRead images have been the least useful so far.)

If I can't fix these disks using the images you are providing, then you are quite welcome to send them to me for salvaging. I would, of course send them back to you afterwards. I recently had some disks sent to me from New Zealand which arrived none the worse for having travelled half way round the world.

It might not come to that, but if it does, I have an enviable collection of floppy drives suitable for disk recovery with Amigas, Macontoshes and PCs, plus Catweasel and KryoFlux controllers with their dedicated disk imaging software and Disk2FDI Registered.

Originally Posted by HymnsomniaK

Disk saviour indeed!

Well, owning that little lot helps.

Edit: I forgot to mention that the Disk-Validator file on that disk was, in fact, the SADDAM virus, so the disk was infected. I disinfected it by overwriting the file with a replacement from a Workbench 1.3.3 disk.

[prowler]

Originally Posted by prowler

I shall return to mapping the blocks. I have a suspicion that there are some duplicate data blocks on this disk which may provide the key to fixing the Graph program at least.

The first four long words of Disk Block 918 ($372) at offset $72C00 are duplicated in Disk Block 329 ($149) at offset $29200 in the disk image, and it looks like the data are intact!

The next data block pointer is $134 (308) at offset $26800 but, alas, this is the LUIS 4 file header block. Unfortunately, the (reconstructed) first four long words of Disk Block 919 ($373) at offset $72E00 are unique on the disk, and that block is very badly corrupted.

This still leaves two corrupted data blocks to fix: Disk Block 116 ($74) at offset $0E800-$0E9FF in the disk image and Disk Block 919 ($373) at offset $72E00-$72FFF, belonging to the Calc and Graph programs, respectively.

[HymnsomniaK]

Originally Posted by prowler

The next data block pointer is $134 (308) at offset $26800 but, alas, this is the LUIS 4 file header block.

I can state with 90% certainty that this is a save work file and thus not part of the main software that you are trying to salvage. Luis is a name like Lewis or Louis but that you may have guessed already.

Originally Posted by prowler

I think it is probably better to try corrupted disks in different drives than with different dumping software.

I have at least 2 more A500 , 1 A500+ , 1 A600 that I can try making images from , I'll provide images from them before resorting to sending you the disks but if that is what needs to be done it will be done.

I have another external drive I can try but since the office disk has the Saddam virus I'm a bit weary of dumping it on my A1200 with a brand new CF HD...

[prowler]

Originally Posted by HymnsomniaK

I can state with 90% certainty that this is a save work file and thus not part of the main software that you are trying to salvage. Luis is a name like Lewis or Louis but that you may have guessed already.

You are quite correct, of course. What I meant so say was that the next data block in the sequence (308) would have contained valid data for Disk Block 919 with which to repair the Graph program if it had not been overwritten by the LUIS 4 file header. One more reason not to store work files on original disks!

I'm not sure what loose blocks containing duplicate parts of the genuine program files are doing on the disk, but I have seen this before on original disks which have been write-protected from new, as if the distribution disks are actually copies of the disks they were using to develop the software, and heavily fragmented too!

Originally Posted by HymnsomniaK

I have at least 2 more A500 , 1 A500+ , 1 A600 that I can try making images from , I'll provide images from them before resorting to sending you the disks but if that is what needs to be done it will be done.

I have another external drive I can try but since the office disk has the Saddam virus I'm a bit weary of dumping it on my A1200 with a brand new CF HD...

Thanks in advance for any more images you can provide from your other drives.

I can assure you that the SADDAM virus will present no threat to your A1200 provided it is running Kickstart/Workbench 2.0 or higher. This is because it hooks into vectors which are only present in Kickstart/Workbench 1.x


Earlier this evening, I copied Disk Block 329 of the Office disk image into Block 918, corrected the header before recalculating the checksum, and it matched the original checksum exactly!

I now have two WIP images of the Office disk: one containing Disk Block 116 from your first (internal drive) set of captures, and another containing Disk Block 116 from your second (external drive) set of captures. In each case, the header has been reconstructed and the checksum recalculated.

Both WIP images also contain the badly corrupted Disk Block 919, again with the header reconstructed and checksum recalculated.

All files copy out from both WIP images without errors, but running the Calc program in either case causes a Guru when it is loading, so Disk Block 116 is still bad in both images.

The File program works properly as far as I can tell. The LUIS series of files are designed to open with the Graph program, which loads and displays them despite the corrupted Disk Block 919.

The first four long words of the reconstructed headers for data blocks 116 and 919 can't be found on any Office.adf file in either set of disk captures you have provided so far, but I am hopeful that valid copies of these bad sectors may eventually be salvaged using another disk drive.

Thanks for all the data you have uploaded so far.

I am looking forward to see what you can do with your other disk drives. It's very useful indeed that you have so many.

[HymnsomniaK]

It's gonna take me awhile digging up my old hardware and set it all up. In the meantime the Zone contains a set of dumps from another external drive. I did not make dumps with rawread since they're not being useful.

Have fun!

[prowler]

Originally Posted by HymnsomniaK

It's gonna take me awhile digging up my old hardware and set it all up.

No problem, take your time. This is worth doing, so it's worth doing properly. I have plenty to keep me occupied with three other disks I haven't yet started on.

Originally Posted by HymnsomniaK

In the meantime the Zone contains a set of dumps from another external drive. I did not make dumps with rawread since they're not being useful.

Thankyou very much. I'll let you know if I make any further progress on the Office disk as a result of those dumps. Otherwise, I'll start on the Write disk next, since that's the other one damaged with the SADDAM virus, and I know I can make some immediate progress on that.

Originally Posted by HymnsomniaK

Have fun!

Yes, it is fun recovering data, because each task presents a different situation and with that comes new knowledge, so I'm getting better at it all the time.

[lesta_smsc]

Prowler - once again you help restore the past!

[mai]

Originally Posted by HymnsomniaK

I think it would be best to have all four disks in the best possible shape you can get them into before uploading. The main goal is to get the best possible set so they can be contributed to TOSEC. This app is turning out to be quite rare which is bad , very bad.

Good way of thinking, instead of contributing bad dumps.

[prowler]

Hi Hymnsomniak,

The third set of dumps you uploaded contain nothing which will help recover the two bad sectors on the Office disk, unfortunately.

I have now completed mapping the blocks on that disk and I can confirm that bad Disk Blocks 116 and 919 are the only remaining problems with it.

I will now start looking at the Write disk, but will of course continue checking any future dumps you make of the Office disk as well.

[HymnsomniaK]

@mai

It is only logical.

@prowler

As soon as I get my old Amigas working (they are out just gotta find those PSU's , not stored in the same boxes for some reason) I'll dump them ASAP.

[prowler]

Originally Posted by HymnsomniaK

@prowler

As soon as I get my old Amigas working (they are out just gotta find those PSU's , not stored in the same boxes for some reason) I'll dump them ASAP.

Thanks, mate!

Earlier today, I restored all seven virus-damaged data blocks on the Write disk. All files now copy out of the disk okay, but the Disk-Validator file is once more the SADDAM virus and I'll replace that tomorrow.

I think that will complete the restoration of the disk, but I will run the Write and Spell programs, check the other files and map the blocks to see if any any files have been deleted before trying the Page disks.

If it turns out that this disk is fixed, then it'll save you making any more dumps of it.

[prowler]

Hi HymnsomniaK,

Earlier today, I replaced the infected Disk-Validator on the Write disk with a copy from a Workbench 1.3.3 disk (the same I used to replace the infected file on the Office disk).

The Write program works. The 'sij' and 'SONY V 180' documents and those in the Examples directory (except 'names.mrg' which has no default tool) open with the Write program. I'm assuming that the Spell and Dictionary applications are invoked from within Write, as they cannot be opened directly. There does seem to be some missing files (no important ones, as far as I can tell); maybe mapping the blocks will bring them to light.

So far, I have mapped half the blocks on the disk. No problems yet (and no missing files either).

I don't think you need to dump the Write disk any more. All dumps of this disk from the three sets you have uploaded so far (except the DMS dump from the internal drive I started with) are identical.

[prowler]

Hi HymnsomniaK,

I have now mapped the first thousand blocks.

It's slow going because, as I had suspected, the second half of the disk is full of deleted files which I think may be part of the original software.

The default tool for most of the files (#?.ADV and #?.ADV.info) is "df0:Advantage" or sometimes "Advantage:Advantage".

There are some similar files in the Exemplos directory on the Office disk, but I can't find the Advantage tool on any of the four disks in the set.

Do you perchance have a fifth disk entitled "Advantage" that goes with this set?

[HymnsomniaK]

Hi prowler

Unfortunately there is no 5th disk and no way of knowing exactly how many disks were because they aren't numbered. Never heard of any Advantage software neither.
As for the mystery of the missing PSU's I can't find them at my place so they might still be in a box somewhere in my parents house.

Thanks for all you've done so far.

[gulliver]

Originally Posted by prowler

Hi HymnsomniaK,

I have now mapped the first thousand blocks.

It's slow going because, as I had suspected, the second half of the disk is full of deleted files which I think may be part of the original software.

The default tool for most of the files (#?.ADV and #?.ADV.info) is "df0:Advantage" or sometimes "Advantage:Advantage".

There are some similar files in the Exemplos directory on the Office disk, but I can't find the Advantage tool on any of the four disks in the set.

Do you perchance have a fifth disk entitled "Advantage" that goes with this set?

Easy answer:

You can get Advantage, The (1989)(Gold Disk)(Disk 1 of 2).zip, 353.30 KB and its related files from Amiga Applications in TOSEC 0.23
I am sure it will help you to rebuild the set.

Anyway, great job

[prowler]

Originally Posted by HymnsomniaK

Hi prowler

Unfortunately there is no 5th disk and no way of knowing exactly how many disks were because they aren't numbered. Never heard of any Advantage software neither.

Don't worry, I have made some interesting discoveries online this evening:

The Gold Disk "Office" software comprises four disks and includes the applications Write (word processor), Page (WYSIWYG page layout), Calc (spreadsheet), File (database) and Graph (chart program). An installer program is provided for hard disk installation.

Gold Disk "Advantage" is a separate power spreadsheet application on two disks. I have attached a picture below. (There's a copy for sale on eBay ATM.)

I suspected as much when I discovered several remnants of an "Examples" Root Block on the Write disk, having already found suggestions of a separate "Advantage" disk yesterday.

It turns out that the two disks in the Gold Disk Advantage package are entitled "Advantage" and "Examples".

I have now mapped 1500 blocks on the Write disk and should finish it tomorrow. The majority of the deleted files appear to be intact, so then will begin the task of undeleting them.

Originally Posted by HymnsomniaK

As for the mystery of the missing PSU's I can't find them at my place so they might still be in a box somewhere in my parents house.

There's no rush here at the moment (with two disks still untouched), but let's hope you can find them eventually!

Originally Posted by gulliver

You can get Advantage, The (1989)(Gold Disk)(Disk 1 of 2).zip, 353.30 KB and its related files from Amiga Applications in TOSEC 0.23
I am sure it will help you to rebuild the set.

Thanks, Gulliver!

I downloaded them both (disks 1 and 2) from Planet Emu a short while ago. I can see many (but not all) of the deleted files on those disks, but it's a start.

[prowler]

Hi HymnsomniaK,

I have now completed mapping the blocks on the Write disk.

The second half of the disk is filled to a large extent with deleted files. Most are salvageable, but some are not. I will have to undelete the salvageable ones so that I can access them properly. This must be done by editing the disk blocks, because the original directory header has been overwritten.

There is a previous Bitmap on the disk which is different from the current Bitmap only inasmuch as which Bitmap is marked used, so I shall use that previous Bitmap Block for the new directory.

There is, in fact, a third Bitmap which has all of the first half and much of the second half of the disk marked used, and this may indicate more closely what the disk originally looked like. This one I shall keep intact.

I will, of course, save a copy of the latest WIP Write disk image in case I am following a dead end, but if these files are found to be different to those on the Advantage disks, then they will be worth preserving.

However, before I continue with this, I am first going to have a look at the Page and PageUtil disks to see if I can save you the trouble of dumping those disks again.

I have today downloaded a set of four Advantage program disks and two Examples disks from TOSEC. All three disks from Planet Emu are duplicated here, but there are two very different versions of each disk in this set.

[HymnsomniaK]

Hi prowler

I will be in possession of my PSU's this weekend and will be able to dump whatever disks you need.

[prowler]

Originally Posted by HymnsomniaK

Hi prowler

I will be in possession of my PSU's this weekend and will be able to dump whatever disks you need.

That's great! Thanks for letting me know.

The Page and PageUtil disks are in much better shape than the other two I have looked at so far.

I think I will be able to fix them both using the images you have dumped so far, and I am hoping I will be in a position to confirm that by tomorrow evening.

If all goes well, you will be able to concentrate on the Office disk, which still has two bad sectors to salvage.

No further dumps of the Write disk will help, because the image I have is already as good as you will get. There are no bad sectors - just some deleted and overwritten files, but the word processor functions, basically.

I will upload an image for you to try. It's all in Portuguese. There are some error messages which pop up in certain situations, probably because of missng files. If you can tell me what they are saying, it might give me an idea what I should be looking for. I have found some other disk images which might help with that.

[prowler]

Hi HymnsomniaK,

I am within a whisker of fully recovering the Page disk!

Again, I started with the original DMS dump you made of the disk using your internal drive.

Although, as you have said, only the images made using your internal drive would boot, this image was, in fact, in very poor condition with very many blocks zeroed out and a few were corrupted.

In the end I had to use data from all three sets of dumps you uploaded, and I needed to repair the preferences file using one from a Workbench 1.3.2 disk.

The only problem with the disk now is that there should be a ReadMe.info file in the Root directory, which isn't showing up.

The file exists, but either there isn't a Root directory Hash table entry for it or it is missing from the appropriate hash-chain, and it's header and data blocks are probably marked unused in the Bitmap.

For this reason, the disk is modified when you boot with it (the Root directory .info file is modified because the ReadMe.info file is missing).

It will be a trivial fix, but when that is done, the disk will be 100% restored!

There will be no need for you to dump this disk any more now.

I didn't have time to check the PageUtil disk today, but I'm hoping to get that done tomorrow. I'm not expecting any major problems with that.

If you get any time for disk dumping over the weekend, it would be a great help if you could make DMS and DiskRipper dumps of the Office disk in as many drives as you can. Thanks.