Lost Boot Screen

[pwyll]

The first mod I attempted was the custom cl.gif, which worked beautifully. The second mod I attempted was the "Death to Hellomoto (http://www.modmymoto.com/forums/showpost.php?p=2522471&postcount=1)" guide by jklong03, which also worked beautifully. The third mod I attempted was replacing the power-up and power-down sounds (http://modmymoto.com/forums/showthread.php?t=326741), which did not go beautifully, but which was compleated successfully (as documented here (http://modmymoto.com/forums/showpost.php?p=2484771&postcount=15)) with the help of Danation and modnmotos, and which resulted in some interesting twitches/glitches/caveats of the V3re (or at least a hw:CHWF1608AA running R3442A_G_0E.43.25R) regarding the naming of custom sounds in relation to stock or custom animations (here (http://modmymoto.com/forums/showpost.php?p=2504081&postcount=51)).

For my fourth mod I decided to go back to the cl.gif and try the changing it on the fly using the seem edits/renaming on phone method. I cannot tell if it worked (or would have) or not.

I did the seem edits and they seemed to go successfully, but once finished I could not upload the gifs to the a/mobile/picture directory. Actually, I could not upload anything. Actually, I could not get the phone to connect. It seemed to work fine (save for the lack of a cl.gif when the flip was open, which suggested to me the seem edits worked), but the only program I could (finally) get to recognize the phone was F&B3 (I tried RSDLite first, to do a backup--never saw the phone).

F&B3 *did* see the phone, though, so I loaded up my "last good" .shx and commenced to flashin'. It did the first four code groups just fine (cg, cg2, cg4 and cg5 I think--I should fire up the proggy and look for proper documentation, but laziness is why I got B's in science and only rarely write technical docs). Upon getting to CG8 (which it flashes before CG7 for some reason) the phone gave me a "Critical Error // 1337 // 510B"--twice. (I was only mildly amused that I got a "'leet" error)

The phone was in the beautiful "white screen" state, but I figured the firmware was the only thing *needed* for the drivers to work (and the flex had been successfully transferred anyway), so I fired up RSDLite and reflashed the phone--successfully and without incident. Except that there was no boot-up screen. The custom animation and custom sound start right when they should and played perfectly, it just gives me that blue-black "I got nuthin' to show you" screen during the bootloading process.

So I pulled out jklong03's guide, checked all my seem edits and fired up MotoBootScrnReplL9 to check my CG8 and everything checked out. Still no bootscreen. So I went through and redid everything according to the guide, and still no bootscreen.

I figured I'd go through this afternoon or evening and reenable all the moto ani's and restart/power-cycle the phone to see if that does anything (and then go through and disable them again whether it works or not), but I thought I'd throw the conundrum out to the population at large here to see if there were any suggestions for anything I may have missed or might want to try first. I'd really rather avoid the tedious process of editing, reediting and unediting the seems (and all the restarts and power-cycles between them) if I could (I think I mentioned that I'm lazy) even though that does seem to be the next logical step...


Thanks,

Scott

P.S. I simply want to get my bootscreen back so the phone is in a "pristine" state before trying the "changing CL's" mod again. I figure I can't really call it a failure and move on to another method without trying at least twice, but I want to start from a 'known good' state with a fresh back-up (didn't realize how much I'd edited the phone book and changed the ring-tones after the last back-up I had).

[modnmotos]

:afro:
pwyll if your flash with rsdlite was successfull back to the last good/working configuration, Im puzzeled as to why the bootscreen is not working. I assume the backup .shx was with a working bootscreen correct? Anyway I know how long some of the processes can be, but to do the mod you want, they have to be done. I cant tell you for sure what the issue is here. Just to throw a few ideas out at you, maybe a different bit, other than the one you were working with, got checked or unchecked when doing the edits. And thats interfering with the bootscreen working. Also is the bootscreen the right format? Size? Try starting from scratch like the orginal bootscreen was on there and your going to change it to the new custom one again. If you need any seem values tell me which ones I can check my bits and let you know which bits are checked and not. Im not an expert at this, I've had alot of errors :banghead: when modding myself so Im just giving ideas hopefully one works for you.
:tapdance:

[Danation]

Here's why your original backup doesn't have the bootscreen. When you added CG8 to the F&B profile, you're showing it where to back up the bootscreen, but without that, it doesn't save the boot screen. So your original backup won't have the bootscreen. But now that you've added the code group to F&B, it will keep the bootscreen in the future.

What I don't understand is why you can't go back and replace the blank white screen with jklong03's guide. That suggests that backing it up to the original missed something somehow.

[pwyll]

:afro:
pwyll if your flash with rsdlite was successfull back to the last good/working configuration, Im puzzeled as to why the bootscreen is not working. I assume the backup .shx was with a working bootscreen correct?

Yes. And I know this for sure because some of the phone book edits that *were* saved in the backup were done *after* the bootscreen replacement.

Anyway I know how long some of the processes can be, but to do the mod you want, they have to be done. I cant tell you for sure what the issue is here. Just to throw a few ideas out at you, maybe a different bit, other than the one you were working with, got checked or unchecked when doing the edits. And thats interfering with the bootscreen working. Also is the bootscreen the right format? Size? Try starting from scratch like the orginal bootscreen was on there and your going to change it to the new custom one again.

That's the problem with the seem edits, I know I didn't check or uncheck anything "extraneous," I was just wondering if anything got changed inadvertently during the process--which means I'd have no idea what it may have been (especially since everything else is working as expected). I was hoping that perhaps someone might have had a similar problem and could at least suggest a place to start looking.

And "starting from scratch like the orginal bootscreen was on there" was what I meant by "So I went through and redid everything according to the guide." I was just hoping someone might have a suggestion to try before I undid and redid all the seem edits before restarting jklong's guide for a third time ;-)

If you need any seem values tell me which ones I can check my bits and let you know which bits are checked and not. Im not an expert at this, I've had alot of errors :banghead: when modding myself so Im just giving ideas hopefully one works for you.
:tapdance:

Well, if *I* had any idea which seems to check I'd have mentioned them first. But I can be remarkably uncreative sometimes, so I have no idea where to start looking for *possible* seem "mis"values. And I've just been doing this a week ;-)

If nothing else, you've confirmed what I was 90% sure of anyway--that there's probably no shortcut (yet, anyway).

_____

Here's why your original backup doesn't have the bootscreen. When you added CG8 to the F&B profile, you're showing it where to back up the bootscreen, but without that, it doesn't save the boot screen. So your original backup won't have the bootscreen. But now that you've added the code group to F&B, it will keep the bootscreen in the future.

The problem is that replacing the bootscreen was done *before* the custom sounds and (subsequent) custom animation. The fact that both were present in the backup means that the CG8 value had already been added to the F&B3 profile. In fact, the first time I fired up F&B was to replace the bootscreen, which is why I didn't have a clean backup available when I was having the problems adding the custom start-up and shut-down sounds.

What I don't understand is why you can't go back and replace the blank white screen with jklong03's guide. That suggests that backing it up to the original missed something somehow.

Well, it's not a blank white screen anymore--in my (very limited) experience, the blank white screen seem to indicate a point at which the processor seems to be ignoring the screen compleatly (as when I dropped the phone and it spontaneously reset, or when it gets semi-bricked) whereas this is the blue-black screen of *trying* to display something but having nothing to display. If you've never seen it before you can observe what I mean by removing the cl.gif from your ~/system directory (or c.gif from your ~/picture directory if you've done the mod I was attempting) and looking at the external screen while the phone is open. It tells me that the programming is working properly and it *thinks* it's displaying a boot image--it just doesn't have one to display.

It's not a major problem as it doesn't affect the functioning of the phone and I don't turn the phone off and on much (except in the process of modding)--it just bugs me because I know there *is* a problem...

:-)


Thanks to both of you,

Scott

[Danation]

BUT, did you do a FULL backup with the CG8 checked or just the one where you check only the CG8 for the mod? That could explain a lot.

Of course, as always, I might be way off, but it's worth a thought.

I'm thinking the problem probably isn't an errant seem edit. Just seems too random to me.

[modnmotos]

I had that same issue when I first attempted the change of the bootscreen. Dont remember the exact reason why I got the blank screen(blue black), but do remember after trying it a third time I finally got it working. Thats why I suggested trying it again. Boy I wish I could remember why, but I know exactly what your talking about. Thats also why I asked about format and size of pic.

And I agree with Danation about the seem I was just thrown it out there as a place to check.

[Danation]

Yeah, I know what you mean about the blank screen. I didn't mean to type "white."

Try doing jklong03's guide again and let's see what happens.

[pwyll]

BUT, did you do a FULL backup with the CG8 checked or just the one where you check only the CG8 for the mod? That could explain a lot.

Of course, as always, I might be way off, but it's worth a thought.

I'm thinking the problem probably isn't an errant seem edit. Just seems too random to me.

Checked CG8, downloaded, replaced pic, restored, checked phone (bootscreen replaced), did full backup.

I had that same issue when I first attempted the change of the bootscreen. Dont remember the exact reason why I got the blank screen(blue black), but do remember after trying it a third time I finally got it working. Thats why I suggested trying it again. Boy I wish I could remember why, but I know exactly what your talking about. Thats also why I asked about format and size of pic.

And I agree with Danation about the seem I was just thrown it out there as a place to check.

It struck me as an errant seem value *because* it was simply the bootscreen and nothing else. But I checked the "Original Hellomoto or Blue Hellomoto" seem, and like before there was no difference between having it off and on.

What I used for the bootup screen was a wallpaper that had both already been checked on the phone *and* had worked as a custom startup image. Rechecked the size anyway, but with both Irfanview and MS Paint reporting it at 176x220 I didn't see a point in resizing it. It was already in the backed-up CG8, but I replaced it and restored it.

Expected to have gone through the process by now, but fell asleep while cuddling with the daughter putting her to bed. Have a date with the wife now, so will try resetting all the stock boot-up seems and unsetting them and then going through jklong's guide from start to finish in a couple hours and report back.


Thanks all,

Scott

[Danation]

You have a knack of finding unusual problems, but I kind of like it.... because I learn a lot from these.

Dumb question: In the F&B profile, are the codegroup settings correct? I would assume that it's obvious that they are because it worked once, but my paranoia thinks you ought to check to make sure nothing got messed up.

[pwyll]

You have a knack of finding unusual problems, but I kind of like it.... because I learn a lot from these.

Yes. There's just something about me. You should have seen the fun times I had setting up my BBS back in '92--I got a national rep as the sysop of irreproducible problems on our little proprietary network. It's also one of the reasons I never ran winblows until (the year) 2000--besides it's inability to multi-task a tenth as well as DesqView, it would never install correctly and people trained by Microsloth couldn't even figure out why (this was before the MSCE days).

But it is a "talent" that has dogged me my whole life--although it does occasionally mean I can fix things other people can't...

Dumb question: In the F&B profile, are the codegroup settings correct? I would assume that it's obvious that they are because it worked once, but my paranoia thinks you ought to check to make sure nothing got messed up.

It's not a dumb question at all--in fact I did it as part of the procedure outlined below before I'd even read your post.


At this point:

1) I have reenabled all stock startup and shutdown procedures--no change in the bootscreen situation, just had all the stock stuff again. Disabled them. Again.

2) I followed jklong03's "Death to Hellomoto" guide step by step (the custom bootscreen was already in the CG8). No joy.

3) I double-checked the necessary changes to the F&B .prof file for the V3re (including the start and end addresses for each code group) and the addition to the offset.ini file for the Boot Screen Replacer. Discovered that I had added in the CG8 line before the CG7 line, explaining why it flashed CG8 before CG7, and fixed that; double-checking everything again before saving.

4) Repeated step 2.

5) Double-checked that the image was not suddenly the problem by converting the bmp backup of the original Hellomoto file to jpg and replacing the custom screen with the original. No joy.

6) Repeated jklong03's procedure, step by step, using my custom bootscreen image saved at a lower file size (sometimes size is an issue). No joy.

7) Repeated the procedure using the original moderate-resolution image just to have my "original" starting point. Checked in case. No joy.


So I'm looking at disabling all customizations dealing with the startup/shutdown procedures to see what happens. Rererenabling the stock startup/shutdown procedure to see what happens when there is no change. And then redisabling the stock and reenabling the custom procedures to have what I can get at this point.

If it *is* a seem problem, I have no idea where to start looking...


:woot:

Scott

-----
Latest development:

While exploring options with P2KAE, SEEM Changer &c. to see if they offered any new avenues to explore or triggered some new way of looking at things, I decided to go ahead and try the changeable CL's again. Worked perfectly. Did not bring back my bootscreen.

I'm thinking, at this point, of flashing an MP just to see if the bootscreen works or not there. I don't know if embedded devices and flash memory works the same way as 8086 processors, ROM and RAM did back when I was taking my little Introduction to Assembler a couple decades ago, but it's struck me that it's possibly a hardware problem. Having all the start and end addresses having to be specifically listed for each phone reminds me of dealing with the hardware registers when programming at the machine level--having to designate the specific physical address of whatever object with which you are dealing. Back then, when there were only a few hundred thousand transistors on a chip, it was not uncommon for one (or a few) transistor(s) to go bad--which is why memory testing utilities were developed.

As far as I know there are no memory testing utilities for these phones, so the only way I can think of to test for a single bad transistor (or cluster) is to flash a "known good" image and see if the problem persists (as it did flashing my "known good" backup). I'm just hesitant, for some reason, to flash an MP >shrug<

Ah well, it'll probably happen sooner or later...


Scott

[Dumba]

I only understand very little, lol, but it sounds like good info. Hope it will be figured out, :)

[pwyll]

jklong03's latest (dual AT&T/T-mobile) Monster Pack flashed fine (if a bit slowly ;) and worked quite well--except for there being no bootscreen on start-up. Tried power-cycling the phone twice just to make sure, got the same blue-black "nuthin' here" screen. Reflashed my backup.

At this point I'm prepared to diagnose it as a hardware problem.

Ah well, I've never had a phone that worked perfectly so if this is the worst problem I have, it's quite livable...


Scott

[Danation]

Wow, I thought for sure flashing jklong03's would work. I have no idea what you'd need to try from there. I'm glad you could at least get the cl.gif mod to work.

hmmm..... I don't understand as much about the hardware aspect as you do, but it seems odd that a hardware issue would only affect one specific function of the phone. Wouldn't there be a lot of problems?

[pwyll]

Not necessarily. Memory now is a lot more reliable than it used to be, so people these days are used to things either working perfectly or suffering catastrophic failure. When transistors were first being miniaturized, back in the day when 12 mhz was blazingly fast and "no one would ever need more than 640k" of memory, IC chips only consisted of a few hundred thousand transistors and it was quite easy for one (or a very few) of those transistors to go bad. This was the reason memory testing utilities were developed--they would write specific bits to each individual physical location and then read them back to see if the value was correct. They would do this dozens or hundreds of times using different information each time to test the integrity of the chip.

The technology is totally different now, and the quality control is much better, but the concept is basically the same--put millions of transistors on a single integrated chip and determining whether a specific transistor is on or off is what tells whether that "bit" is a one or zero. With computers there are redundancies and fail-safes so that any small problem can be seamlessly fixed "on the fly" and the process is totally transparent--you can't tell whether you have something wrong or not if the problem is relatively insignificant.

But "embedded systems" are much more like the personal computers from the early nineties than like the personal computers of today--most still use direct physical addressing (at least at the OS level) because you can get away with much less overhead that way. While it only has half the storage capabilites of my first computer (10 meg flash vs. my 20 meg hard drive), the V3re is much faster and more capable than the computer on which I used to run a nationally-connected BBS.

The point is, that while people today are used to the only physical problems being the loss of an entire chip, it is (or at least used to be) possible for a single component (or small group of components) to go bad within that chip without bringing the whole chip down. While I don't *know* that it's still possible, the behaviour in this specific case suggests that that's what's happening. To wit:

1) Everything dealing with the operating system or its embedded interface (firmware) has to be flashed to specific addresses or the phone will not work.

2) These addresses are different across different models, but the same or very similar within the "family" of the same model--suggesting that the importance of the addresses is as much a part of the hardware as it is the software or firmware.

3) I am having a problem with one specific aspect of the firmware that persists across not only differing configurations, but across differing *firmwares* that work on this model (my firmware is ~.25R, jk's MP is ~.04R).

4) Nothing else on my system seems to be affected, and no one else has this problem with firmwares I have tried, suggesting that the problem is not in the firmware but in my system.

5) Everything checks out good as far as the software goes.

Since the image is saved at a specific location within a file whose size is static** and whose address is fixed, and the problem seems to be within the device itself, this suggests that the memory at that physical location is unreadable to the device. It's not necessarily "bad memory" since changing the image in the CG8 writes to the phone and the image reads back fine when downloading that code group in F&B--but for some reason the phone is not reading that location.
**I have put several different jpegs in the CG8 as bootscreens, all of various sizes, and the filesize as listed in F&B never changes.

The next step in "nailing" the diagnosis is flashing my backup to another phone. When I get the chance I will flash my backup to my boss's phone--if the bootscreen shows there it will prove that the problem is *not* with my files (if it *does* create the same problem on his phone, he won't miss the bootup screen anyway).


Scott

[Danation]

Well, you make a convincing argument. Still seems rather odd, but you know more about hardware than I do.

Good idea about flashing it to your boss's phone.

[pwyll]

Yes, it does seem odd. I haven't really seen anything act like this since about '93 or '94. That is what led me to this (still hypothetical) conclusion--the last time I saw anything act like this (absolute failure of one, and only one, aspect) was "back in the day." And there are still inconsistencies since if it was a problem with the physical memory at that address writing to it might not be a problem but anything read from it should be corrupted somehow and that doesn't seem to be the case.

I actually don't know anything about *this* hardware so I'm really flying blind--it's just that this is the best conclusion *I* can come up with based on hardware I *do* know a little about and what I know of embedded systems in general.

I'll get a chance to flash my boss' phone sometime this week, and that will tell me whether it's my backup causing some sort of fault or my hardware--but I'm really just guessing as to what the hardware problem would be...


Scott

[Danation]

I had another thought that might be relevant. There are areas of the phone that are not affected by a regular flash. For example, your IMEI number won't change if you flash somebody else's MP (if so, we'd all be felons, lol.) Perhaps there is an error in one of those protected zones.

Just another theory, but I thought I'd throw that out there.

[pwyll]

That's entirely possible. It seems to me, though, that if that were the case there would be other effects as well. But it's definitely something to keep in mind.

As far as that goes, I do not use the video functions, haven't tried the camera since this happened, don't use mms or picture messages, &c. &c. So I can't say for sure that there aren't other issues I just haven't found >shrug<

Okay, just tried the camera and it works as well as ever ;>


At this point I'm heading toward "live with it and keep an eye out for similar issues and possible solutions" mode. Just waiting now for the chance to flash the boss' phone to see what happens...


Scott

[Danation]

Yeah, that's true. Just throwing ideas out there. Maybe something will end up working.

[modnmotos]

I also believed the same as Danation:
but it seems odd that a hardware issue would only affect one specific function of the phone. Wouldn't there be a lot of problems?
But after your explaination I stand corrected.

[pwyll]

Yeah, that's true. Just throwing ideas out there. Maybe something will end up working.

New ideas are *always* welcomed :-)

I also believed the same as Danation:
but it seems odd that a hardware issue would only affect one specific function of the phone. Wouldn't there be a lot of problems?
But after your explaination I stand corrected.

Not necessarily "corrected," but maybe "better informed." The fact is that with the state of chips today, it *would* be much more common for chip failure to result in multiple issues rather than just one. It's just that my theory is quite possible considering the state of *most* embedded hardware these days...


:-)


Scott

[modnmotos]

Clearly you know more than me in this area and this maybe a silly question, but here it is: Is each chip dedicated to one function or can one chip serve multiple functions? (with our phones)

[pwyll]

That is precisely what I don't know about *these phones.* With the 8086/8088 and 80286 computers, every chip had its function, and sometimes a function would take more than one chip. Starting with the '386, some chips started serving multiple functions because the expense of manufacturing a more complicated chip started to undercut the cost of materials for multiple chips. You can really see the difference in comparing the guts of a first generation personal computer, in which the "mother board" was often three different boards connected by ribbon cables (and adding 256 kilobytes of memory meant shoving in a four inch wide by 10 inch long card), and a microITX motherboard of today sporting a 1ghz C-7 processor, a gig of memory and has integrated video, sound and networking but is only slightly larger than the RAZR .

The problem with answering your question is that the design of embedded systems depends entirely on intended use and size. You could fit a '486 processor, video and sound unit, i/o interface and a few meg of memory into a single chip the size of a postage stamp, but it would have to be purpose-built and be relatively expensive for it's capabilities. If you had plenty of room, though, you could use commodity chips, from things like watches, calculators, toasters &c., that only had one function each but would be free compared to the cost of the manufacturing process and marketing. But the more single-function commodity chips are used the easier it is to hack and/or modify.

Based on the size of the RAZR's and Motorola's vested interest in keeping their "secrets", I'd say it's a combination. The phone's way too big to be based on a single-chip design (look at the quad-band cellular watches), but if *every* chip had only a single function we'd be able to unlock the phones by shorting a couple solder points or clipping an IC leg or two...


Scott

[Danation]

I agree, it's gotta be a combo between dedicated hardware and chips with multiple functions.

[modnmotos]

Thanks for the detail pwyll!

[pwyll]

No problem :-)

Well, there *is* a problem, but it's very bizarre and for once it's not mine. It's complicated, so I'll start at the seemingly unrelated beginning.

Sometime this week my boss stopped by his T-Mobile store and picked up a new Sony Ericsson. While he was there he asked what it would take to get a new sim card as he likes to trade it between multiple phones and was worrying about it wearing out. The woman behind the counter said "nothing, I can give you one now" took the sim out of his phone, popped it in a programmer (from his description), evidently duplicated his sim to a new card, and then gave both back to him. He said it worked fine except that it now had her name on it instead of his and she gave him a new phone number. He put it in his RAZR to show me, and when I had him check the phone book on the *sim* none of his numbers were there.

I explained that she had given him someone else's sim and that she'd either kept the duplicate of his, or had copied something from the machine's memory instead of his sim, and he needed to go back to the store and make sure someone else wasn't walking around with access to his phone book and account.

When I finally had a chance to talk to him about wanting to flash my backup to his phone to see what happened and then flash his backup back on it, I powered cycled his phone to show him the bootscreen that was missing on mine, and lo and behold it's now missing on his. Even stranger was that the startup animation is now T-Mobile's rather than Motorola's. Even though I *know* that the sim card couldn't affect the phone's startup procedures, I had him change back to his original sim.

Still no bootscreen, still T-Mobile's animation. Once it had finished booting and locked on to the carrier signal, I *#9999#'ed it to check that it still had his original flex and firmwareit does. We then went looking through the menus for the "Software Upgrade" option (since it was there before and wasn't where I thought it was this time) to see what that would do, and at some point during the menu surfing it started asking for his security code after exiting *every menu* (rather than just when powering up or changing an option, as it had before). So we power-cycled the phone to see if that would change anything, and when it finished booting the carrier signal indicated "Unregistered."

I went to the network settings and checked available networks--it found both T-Mobile and AT&T. I selected New Network: "Searching" --> "Unregistered." Power-cycled the phone: "Unregistered."

He swears the Hellomoto and Motorola animation were working when he walked into the store because he'd been using the PEBL over the weekend and had just put the sim in the RAZR and turned it on before walking into the store. I explained that he would have to redo all his customization, but the only thing we could try at that point (it was too late to set up the laptop and reflash his phone) was to give it a hard reset...

Master reset: no Hellomoto --> T-Mobile animation --> Unregistered, totally stock-from-the-box (otherwise) phone. He's taking it back to T-Mobile to see if it's still under warranty (I doubt it) and I'm going to take in a stock MP tonight.

I haven't done anything to his phone except back it up, Master Clear for a stock pack, and reflash his original data. And I don't see how a sim card could possibly affect the UI and basic functions of the phone--am I wrong?

Anyway, the bad news is that it looks like I *don't* have a guinea-pig phone on which to test my theory; and the worse news is that now I'm afraid that my phone may be headed in the same direction as his since we watched it go from perfectly fine to totally unusable in about 15 minutes--and it started with the same seemingly insignificant problem mine has...

:/


Scott

[Danation]

Holy cow! Is it possible that somehow the actual phones got switched, not just the SIM? Did he give her the phone as well as the SIM? That's the only thing I can think of that would cause all of that oddness. I don't think a SIM would change the phone's settings at all, though I could be wrong.

Also, did you try your SIM in his phone? I'm wondering if the "unregistered" part is just a SIM screwup. Then again, you tried his original SIM with no effect.... so that's pretty bizarre. Yeah, I'd try to flash a stock T-Mobile MP onto it.

[Dumba]

Hope that there isn't such a thing as a virus, which can be carried on the Sim. Keep us stated please.

[pwyll]

Holy cow!

That was my reaction almost exactly.

Is it possible that somehow the actual phones got switched, not just the SIM? Did he give her the phone as well as the SIM? That's the only thing I can think of that would cause all of that oddness. I don't think a SIM would change the phone's settings at all, though I could be wrong.

Phones couldn't have been switched. He *did* give her the phone, but he has one of those form-fitting leather cases and it would be too hard for her to pry both parts off his phone and wiggle them onto another without him noticing. And the *phone* still had his contact information and photos of his daughter in the memory.

Also, did you try your SIM in his phone? I'm wondering if the "unregistered" part is just a SIM screwup. Then again, you tried his original SIM with no effect.... so that's pretty bizarre. Yeah, I'd try to flash a stock T-Mobile MP onto it.

Did not try my sim in his phone, but did put both of his sims in the SE and both worked fine (at least as far as acquiring signal).

And upon further reflection (and a few hours sleep) I'm pretty sure that *his* missing bootscreen was replaced by the white "holy sh!t it's a brick" screen rather than the blue-black "I got nuthin' for ya" screen mine show--which would indicate at least a different cause.

Depending on what's happened today, I should be flashing a stock MP tonight. I'm almost certain the phone's out of warranty since I believe he's had it over a year, but T-Mobile has remarkably good cutomer service (at least in this area) and it wouldn't surprise me if he walked through the door tonight with a brand new RAZR. He's gotten at least one of his upgrades that way before--taking in a water-damaged phone to see what it would cost to fix and being told "well, it's out of warranty and we don't carry that model anymore, so take this one instead."


Scott

[modnmotos]

pwyll sim cards can cause funny things to happen, but not to the extent of what you just explained. I do not believe the bootscreen issue could be caused by a bad/faulty sim or the phone being unable to read the sim, or a empty sim.

Here is what happened to me just recently though:

I recieved my subsidy code from at&t and was trying to find a way for the phone to "prompt" for the code without having to have a different providers sim.(for those of you who dont have a different providers sim sitting around) So since I own 4 v3res I thought I would insert one of the "dead" unregistered sims in my phone. Power cycled. Waited. Not successful on getting phone to prompt for code. I dont know why I did but I had put my good(registered) sim into daughters phone and power cycled it. Before switching sims back around I fliped open my daughters phone and had some of the same issues your boss had.

--and when I had him check the phone book on the *sim* none of his numbers were there.---same here.
--network settings and checked available networks--it found both T-Mobile and AT&T. I selected New Network: "Searching" --> "Unregistered." Power-cycled the phone: "Unregistered."--same here.
--looking through the menus for the "Software Upgrade" option (since it was there before and wasn't where I thought it was this time)--same here.

And it was not just this item when looking at mutiple menu items things were missing. My first thought was 'crap my sim is damaged and so is the phone'. Opened phone took sim out and reinstalled, power cycled and still had issues. As my heart sank and I was preparing for having to get a new sim I tried one more time reinstalling sim and it worked this time! I was relieved and chalked it up to the contacts and card not matching up correctly. By the way this is a pretty new sim. We had other issues and went to store to get help with fixing them, and they suggested upgrading to new 3g sims(which did not fix our problems) and that was within the last 3 months.

And Dumba you reminded me of a news cast I saw about 8 months ago:
Hope that there isn't such a thing as a virus, which can be carried on the Sim.
I remember them saying that it was possible and there were unconfirmed reports of people who had phones infected, and warned people to be careful. Dont remember any other details about this, and have never heard anything else since, about virus infections on cellular devices.

[Danation]

A dark day it will be, when viruses are common on cell phones. I really hope it doesn't happen.

[pwyll]

And Dumba you reminded me of a news cast I saw about 8 months ago:
Hope that there isn't such a thing as a virus, which can be carried on the Sim.
I remember them saying that it was possible and there were unconfirmed reports of people who had phones infected, and warned people to be careful. Dont remember any other details about this, and have never heard anything else since, about virus infections on cellular devices.

A virus has to be executable, and as far as I know there is nothing executable on a sim card--it's basically a database which is merely read. That being said, it's really dependent on the phone's OS as to *how* it uses that information, so it's at least theoretically possible. It is, in fact, the first thing I thought of--but, like you, I have never heard of anything like that and it's something that I'm sure at least The Register would have a field day with if it had happened.

On the other hand, the rest of your post has some very valuable information for the situation, so I will go to work armed with more options to try.

Thanks!


Scott

EDIT: As for the "unconfirmed reports" of virii on phones, I'd bet that they were WM phones. There are virii in the wild for *all* versions of windows, but they're carried via the normal means--downloading infected files or running unknown scripts. And I can see how someone might be able to take advantages of vulnerabilities in particular phones by putting specific information in specific places on a sim, like your Windoze "security vulnerabilities,"--but this would be so hit'n'miss that it would either have to be a specific attack against a specific person that the sim provider knew had a vulnerable phone, or would be one of those unfortunate coincidences of have the wrong card and wrong phone at the right time...

[modnmotos]

EDIT: As for the "unconfirmed reports" of virii on phones, I'd bet that they were WM phones. There are virii in the wild for *all* versions of windows, but they're carried via the normal means--downloading infected files or running unknown scripts. And I can see how someone might be able to take advantages of vulnerabilities in particular phones by putting specific information in specific places on a sim, like your Windoze "security vulnerabilities,"--but this would be so hit'n'miss that it would either have to be a specific attack against a specific person that the sim provider knew had a vulnerable phone, or would be one of those unfortunate coincidences of have the wrong card and wrong phone at the right time...
I agree! Thats what I was thinking but as I said I couldnt remember details. Mind is not what it used to be.(before partyn days) And since we are on the topic, correct me if I'm wrong, but wouldnt it be pretty easy to send a virus to say like an iphone that can surf the web and download files with email?

[pwyll]

In theory, yes. But it would have to be written specifically for the phone's OS and so the practicality would depend on how the phone executed commands and ran files. In the case of the iPhone specifically, it would likely be extremely difficult since the core of it's OS is the same as the core of MacOS X (Darwin), which is a "variation" of UNIX. Not only is it more difficult to write a virus for POSIX systems, it fairly easy to have the system run in such a way that a virus would not work *anyway*.

On the other hand, what most people *call* viruses these days are simply scripts delivered as Trojan Horses in emails, and there's no telling how easy *that* would be to get running on an iPhone since that would depend on the proprietary stuff in Darwin and Apple seems to be following M$'s path now.

As far as virii and phones go, though, I think the place to be watching (read: wary) is with Android. It's just enough of the Linux kernel to run a JRE, but the actual phone apps and functions are all java applets. Unlike the current batch of "java phones" where the OS is capable of loading a JRE, Android actually uses JAVA as the phone's OS--and there are already viable JAVA virii in the wild...


Scott


Oh, and back to the topic of the thread--

My boss didn't get by the T-Mob store yesterday, but he *did* take my advice and leave the phone powered off for a few hours to see what happened. When he powered it back up it connected to T-Mobile immediately. Still no bootscreen, but that's obviously a different issue now...

[modnmotos]

Again thankyou for the detail, and thats the main route I would think viriuses would be spread is through e-mail as far as phones go.

[Danation]

I think you could probably piggyback a virus onto a game, and if you made it a doomsday virus, it would be hard to figure out the specific source.

However, though there are viable viruses in Java, Sun created Jave with the specific intent to prevent viruses. Obviously, people can find ways around that, but it's quite difficult. (Trust me, I've played around with with Java. I even tried to write a virus to see if it would work. Like I said, it's quite difficult.) Just my two cents' worth.

I wonder if the fact that you don't have a bootscreen came from when you flashed your boss's backup onto your phone. Did he have a bootscreen before?

[pwyll]

I think you could probably piggyback a virus onto a game, and if you made it a doomsday virus, it would be hard to figure out the specific source.

One problem here may be the definition of "virus" we are using. To be a virus, "technically," it has to both replicate itself and attempt to pass itself along somehow. If it simply harms your system without infecting anything else, it's malicious software being spread as a Trojan Horse.

Of course my terminology comes from my being around when virii were still programs rather than VB scripts and Outlook was a calendar program that couldn't even read email...

I wonder if the fact that you don't have a bootscreen came from when you flashed your boss's backup onto your phone. Did he have a bootscreen before?

Yes he did. After I do anything, successful or not, I power-cycle whatever I'm working on because you never know for sure that something was successful until you've seen it after turning the unit on. You wouldn't believe how many times over the past couple of decades I've worked on configuring something and testing and double-testing to make sure it worked, turned it off for the night, only to have everything back to the "original state" the next time I turn it on. A modification is not successful until it's survived a reboot :-) So I saw his bootscreen before I backed his phone up and after clearing it and reflashing the original backup.

And I had a bootscreen after flashing back to my own backup. In fact, I had a bootscreen not more than 10 minutes before doing the initial seem edit that apparently caused the problem as I had switched out the battery to one with a full charge before starting that round of modding...


Scott

[Danation]

Well, this may be semantics, but games have the intent of distribution and therefore replicate. You sent the game to your friends. They get it. Spread it, etc. And if you make it a doomsday virus, all of the sudden 10,000 phones quit working one day.

Similar to what the Michaelangelo virus did a decade or so ago.

[pwyll]

Ah, but that's the difference. The games are merely a means of distribution and if you download an infected game, tell your friends about it and download it, it's spread but there's been no replication. It's merely malicious code that's disguised as, or hidden in, something else and once that "something else" is gone, so is your malicious software.

If, once you triggered the code, the virus copied itself into other files, that's a virus. It's replicated itself and getting rid of the original carrier now won't make a difference at all. The difference is that in the first case merely deleting the game "cleans" your phone, but in the case of the virus you would have to get rid of anything the virus "touched"--and in the specific case we're theorizing about that might mean compleatly ridding the phone of the ability to start a JRE at all.

The Michelangelo virus from 1991 is the perfect example. Once released into the wild, it spread on it's own. It was a boot-sector virus which means that once a computer was infected, every executable file on that computer, every floppy disk put into the computer, every writeable media of any kind, became infected. If you downloaded a "clean" game and then gave that game to a friend you passed on Michelangelo because it replicated itself in everything it possibly could.

With a trojan, you have to worry about everyone that obtains that file, so a trojan with a doomsday payload could conceiveably bring down 10,000 phones one day. With a virus you have to worry about every file that's been in contact with every phone that's been in contact with any file that's been in contact with the virus. If a successful virus is written for *these* phones and it *does* somehow incorporate a doomsday payload, you're talking about having to worry about every P2K phone in circulation--10,000 phones would end up being a pretty lame hit in that case...

:eek:


Scott

[Danation]

Well, it looks like I need to update my definition of "virus." I guess I'm from the "new" crowd that just calls anything that hurts your computer a virus.

Hmm... that would be a problem, if all p2k phones were infected.

But I was theorizing a more realististic scenario. It would be almost impossible to write a true virus in Java (with all of the replication, etc.) that would infect all p2k phones. But a game with a doomsday payload (do I have the correct terminology there?) is within the realm of possibility.

I must say you are very knowledgeable about this sort of thing. I'm impressed.

[pwyll]

Well, it looks like I need to update my definition of "virus." I guess I'm from the "new" crowd that just calls anything that hurts your computer a virus.

It doesn't help that none of the anti-virus programs differentiate between a "true" virus and the "new" virus. When M$ added email functionality to Outlook and then included that part (Outlook Express) with windoze, OE became the 'default' mail reader for the internet. Since M$ had it set to execute VB scripts automatically, a new form of malware was born that delivered its payload *and* sent a copy of itself to everyone in your address book. It was technically not a virus because it didn't actually replicate (infect other files), but the term "virus" had finally entered the mainstream and caused automatic concern so it was used as an easy way to emphasize the threat. And no other term is more accurate since it *does* use a viral form of transmission. And since most adware and spyware were delivered this way at first (before it was discovered how to load files onto your computer without you having to download them) most people just started calling *everything* a virus--so you're not alone :-)


But I was theorizing a more realististic scenario. It would be almost impossible to write a true virus in Java (with all of the replication, etc.) that would infect all p2k phones. But a game with a doomsday payload (do I have the correct terminology there?) is within the realm of possibility.

I must say you are very knowledgeable about this sort of thing. I'm impressed.

Yes, luckily virii are very unlikely so long as phones are either proprietary devices or are running proprietary software. The more phones are based on windoze, though, the more likely it is that virii will pop up since the windoze pda's have had virii for a while and the WM is based on the pda software (as far as I know). There is a proof-of-concept Blackberry trojan (BBProxy) but it is a difficult to load intentionally and almost impossible to do so "accidentally"--and once loaded is usually ineffective. A doomsday payload in a game (or even just a malicious program) is quite possible (depending on how much access to the unit the JRE allows the program--this is one of the strength of Java in this area, it's basically a "virtual machine" so the programs can rarely have direct access to anything but the screen), but with so many people across the world modding and playing and learning *how the phones work*, it's still unlikely. The great "advantage" to windoze in the early days was that you didn't have to know anything about a computer to use one, just turn it on and start clicking icons. This was also what allowed virii to become so prevelant and spread so easily--without knowing how the programs interacted with the hardware people had no idea virii were even possible, let alone how to prevent them. On the other hand, it seems to me that this is what makes virii and malware on phones very unlikely, since it would probably have to infect the PDS area of the phone to be effective and the only likely method of transmission (on most phone) would be via Java which will not allow access to sensitive areas. Unless an exploit is found in the JavaME engine.

And don't be impressed--its a function of being old ;-) Once you've had to clean a couple virii off your machine, and had to clean other people's machines a few times, you learn everything you can...


Scott

Article I found while posting this message: How Cell-phone Viruses Work (http://electronics.howstuffworks.com/cell-phone-virus.htm)

[Danation]

Hm.... lots of food for thought there. You're absolutely right about a virus likely trying to attack the PDS area. That would do the most damage. But even a seem edit or two would the phone useless to most users. Remap all the keys to the off button or something like that. Even deleting a few key files would work.

I hope our faith in Java's security is not misplaced. One well placed piece of malware could cause major major problems.

[modnmotos]

Could you use java to 'seem edit' or remap keys? Is that possible?

[Danation]

I know that there's a z3 corelet that messes around with softkeys, so it wouldn't surprise me. In fact, that's one thing I want to look into doing intentionally. Maybe I could get a corelet to store texts in a different place so that I can save more... but that's just pure speculation at this point

[modnmotos]

I have a few more questions about this I was just tired last night sorry. Does any one else have any info they could provide in this area? This is an area of interest for me, and obvisiouly for Danation.

[pwyll]

... But even a seem edit or two would the phone useless to most users. Remap all the keys to the off button or something like that. Even deleting a few key files would work.

The first sentence and last sentence only get a "yep" from me, but that second sentence brings back memories of ansi bombs and someone could throw in a whole lot of frustration and fear before (or instead of, depending on their sense of humour) making a :bricked:. The biggest difference is that with an ansi bomb all you had to do was powerdown the box without touching the keyboard...

I hope our faith in Java's security is not misplaced. One well placed piece of malware could cause major major problems.

I think Java's security is pretty tight. I think the only problem would be if an exploit was found that either bypassed the security or caused the engine itself to do something malicious. This is only my opinion, of course, and I am not an authority. But I have been "following" Java since Sun first introduced it as a platform-independent programming tool, and one thing I've seen (at least on the "full-sized" Java) is that it shuts itself down rather than behave in an unexpected way.

And even if something was developed that could get around the Java security, it would still likely be model-specific and so not worth the effort...

[Danation]

Could be. I was able to do some things with Java a few years ago that could theoretically cause harm, but I didn't really pursue it much (I'm not a virus writer, lol.) But good point about it shutting down rather than acting in an unexpected manner. I had so much grief trying to do legitimate things and accidentally causing errors.

What do you think about the corelet idea I've been kicking around? I haven't had time to research it (I'm at work, so when things are slow I can post stuff, but I rarely have time to do anything in depth.)

But the concept, at least, seems very intriguing.

[modnmotos]

So java has a fail safe built in that will not allow anything malicious to run. Yes that is tight security.
"And even if something was developed that could get around the Java security, it would still likely be model-specific and so not worth the effort... " Well that would be worth the effort for him to acomplish what he is wanting to do. He wouldnt have to reprogram the phone.

[pwyll]

... I had so much grief trying to do legitimate things and accidentally causing errors.

From what I understand, that's the biggest "learning curve" of Java--figuring out what works in practice. As far as I can tell, it's harder to figure out what correct code will not cause failures than it is to learn the 'language' itself.

What do you think about the corelet idea I've been kicking around? I haven't had time to research it (I'm at work, so when things are slow I can post stuff, but I rarely have time to do anything in depth.)


I'm very interested in finding out what you can figure out/discover/learn but you won't get any *help* from me--that's beyond me. I haven't had the opportunity to purse any sort of programming since my early college years, and that was before Windoze 3.1

I follow a lot of stuff, clinging to the illusion that I'll be able to get back into it (read: actually start) but at this point I really doubt it will happen. :-/

[Danation]

I hear you there.

I was programming Java when it was just barely out of Beta testing. I've looked at it a little recently, but not very deeply. But I'm pretty motivated on this corelet idea. It has many possibilities. For example, myfaves is a corelet (at least it is on a z3.)

So if I can actually get some time, I'll see what I can do. I'll post any results I can find.

Did you ever find out anything about the bootscreen deal?

btw, nice avatar pic.

[pwyll]

So java has a fail safe built in that will not allow anything malicious to run. Yes that is tight security.

Sort of. Java is kind of like an interpreter (like basic) in that jar files are not actually executable--they're just very extensive and flexible scripts. The JAR tells the JRE what it wants to do, but it's the Java engine that actually does things, and it's built not to do anything malicious. And in order to guard against possible exploits (which have been found, but all of them I've heard of have been with the M$ implementation of Java [didn't affect Linux, BSD or Mac OS's] and the only vulnerability in a phone discovered so far has been in the Nokia system 40's as referenced here (http://blogs.zdnet.com/security/?p=1772)), the Java engine will shut down as soon as it does something it doesn't think it's supposed to do.

"And even if something was developed that could get around the Java security, it would still likely be model-specific and so not worth the effort... " Well that would be worth the effort for him to acomplish what he is wanting to do. He wouldnt have to reprogram the phone.

Yes. But virus writers are looking for a different kind of payoff--either delivering a payload (in the case of spyware/adware/hijackers) or the thrill of looking at a mass reaction and saying "I did that." Being able to accomplish something with your phone that you couldn't otherwise do would justify the trouble--but *possibly* getting your spyware on three phone across the country or managing to bring down 150 phones across the world is not going to provide a payoff justifying the effort of a *malicious* code writer to spend the time figuring out how to do it or if it can be done. IMHO :-)


Scott

I hear you there.

I was programming Java when it was just barely out of Beta testing. I've looked at it a little recently, but not very deeply. But I'm pretty motivated on this corelet idea. It has many possibilities. For example, myfaves is a corelet (at least it is on a z3.)

So if I can actually get some time, I'll see what I can do. I'll post any results I can find.

Then when you started with Java was when I was convinced it was too difficult for me to start on. Found out later that I might have actually been able to "keep up" if I'd actually investigated rather than over-complicating what I was reading. And it makes sense, but what you can do is likely as limited by the implementation of Java in the phone as much as the Java itself. In the link I posted above, Sun mentions that the vulnerablity they've confirmed is limited to the System 40 Nokias, not even other Nokia Java phones. Have you looked at Motorola's development pages to see what they have available as far as that goes?

Did you ever find out anything about the bootscreen deal?

btw, nice avatar pic.

It's the end of the year, so work is ramping up. It's the holidays, so we've got all the extra crap going on for that. And a friend's computer is going wonky again so I have to diagnose whether his hard drive is going bad or the IDE controller is freaking out--I haven't been able to do anything on the bootscreen front at all. Right now it's only broken stuff that's getting attention and the phone works. I may be able to do a little looking at work the next couple of days, but if not it'll be sometime in January before I get back to it :-/

Thanks. It's one of the few pictures Of myself I like--and it's due entirely to the amazingly poor button design of Samsung phones ;-)


Scott

[Danation]

It's the end of the year, so work is ramping up.

Yeah, I know what you mean. Which is why I haven't been able to get much skinning done, something I've been trying to get to for a while.

Thanks. It's one of the few pictures Of myself I like--and it's due entirely to the amazingly poor button design of Samsung phones ;-)


Haha, yeah, I read that, that's awesome.

Java is kind of like an interpreter

The Java VM is indeed an interpreter. Which means nobody's going to be writing the next first person shooter on it, but it provides a lot of cross-platform possibilities.

Have you looked at Motorola's development pages to see what they have available as far as that goes?

Good idea. I'll see if I can do that soon. I'm also going to take a look at the java development section here at m3.