Is this nuvi 2589 simply Bricked or is it a Hardware Failure?

[VintageDave]

Nuvi 2589LMT Bricked

MODEL: Nuvi 2589LMT
IC: 1792A-A4BVG03
FWVER: 8.80
HWID: 1947
No NV backup
How it got bricked: Unknown; it worked for 7 years (with periodic updates), then yesterday it just didn't.
Symptoms:
1. Garmin logo displayed forever (most tries)
"System Software Missing" (once in a while)
Same behavior on battery, USB charge cable, USB data cable, Traffic Update cable
Same behavior with microSD card in/out
2. Hard power off (depress POWER for > 8 seconds) usually reboots instead of powering off
3. Power up by connecting data cable does not show any new device in Device Manager
4. Pre-boot behavior: (invoked by thumb on G after Garmin logo appears)
SYSTEM is displayed for about 8 seconds; Windows detects an Unknown device;
Garmin Express does not react
Then LOADING... is displayed, and the Unknown device goes away
LOADING... displays forever until hard power off
Host computer:
Windows 10-64 on a Lenovo Thinkpad T430
Garmin Express installed, currently version 7.17.2.0, not running by default
Nuvi 2589 successfully updated many times in the past with this USB data cable
Cable is on a Docking Station USB port i.e. internal hub
Also have previously used a direct laptop port which is also a hub
All this DOES NOT MEAN THE DRIVERS ARE INSTALLED,
I had to explicityly download and install USBDrivers_2312.exe from Garmin
per link "Garmin: USB Drivers Updates & Downloads" ([Only registered and activated users can see links. ])
posted in thread, "IMPORTANT! If you want HELP in case of a bricked Garmin device, please follow this".

WHAT I'VE DONE

1. Download and install Garmin USB drivers 2.3.1.2 and restart computer
2. Plug in cable. If nuvi 2589 was off, it powers up to Garmin logo forever
(or occasionally SYSTEM SOFTWARE MISSING)
3. Start G7ToWin. (No special compatility settings nor Administrator.)
Select USB communication.
4. Hard power off nuvi 2589, it immediately powers up again
5. Quickly press and hold G in Garmin logo forever
6. Nuvi 2589 displays scrolling "SYSTEM" and Windows sees an Unknown USB device
7. Quickly in G7ToWin do GPS | Get GPS ID
G7ToWin spins about 10 seconds
(nuvi 2589 switches from SYSTEM to LOADING... and Windows stops seeing the device)
then G7ToWin returns a popup
"USB Error: WaitComplete: Error waiting for overlapped I/O" .
When I dismiss the popup I see
"nuvi 25x9 Software Version 8.80 ID: 1947
Unit ID: xxxxxxxxxx; USB Comm via Bulk Pipe"
8. Nuvi 2589 displays LOADING... forever and now Get GPS ID pops up "invalid handle"
and returns nothing
9. I don't know if it's in preboot mode, or just got comm-lucky partway there and
then froze.

I don't want to rush ahead and possibly wreck it worse. Given the symptoms, what do you suggest next?

[Butters]

@VintageDave
The device likely needs to load cure firmware to re-enable Mass Storage Mode so you can remove any corrupt files or as a last resort do a re-format. Read the two threads linked below:
[Only registered and activated users can see links. ]
[Only registered and activated users can see links. ]

Clearly the device can access preboot mode because G7ToWin only works to read devices in preboot. Usually devices that can access preboot can be cured.

Post again if you need more assistance. BTW, your device probably isn't getting far enough into the boot process to read from a card, in any case that file you inquired about [Only registered and activated users can see links. ] is for a device with a different problem, i.e. 'System Software Missing' permanently displayed. Yours is probably 'soft-bricked' which needs a cure fw loaded in the first instance.

[VintageDave]

I'll try a bunch more times, but it looks like I'm seeing the problem where Updater's progress goes straight to 100%, time passes, then it pops a vague "There is a problem" error message. The device didn't get programmed with Cure; it still hangs at the logo without USB, or (if I signal preboot with thumb on G) enables USB only for a few seconds. Thanks for all the coaching so far. I'm trying to read about similar problems, but I get overwhelmed by the sheer number of long threads that search turns up. Due to a concussion a couple years ago I can't juggle as many eggs as I used to.

I will download the region tool to support @kunix's SD Curing method (his post #33 in "Running programs from SD in nuvi").

Edit:

I included That Word in my update so I guess it got auto-moderated. (I thought it was only if it was the first word.)

I tried QuickCure3. I got progress bar instant 0-100% then the "update successful" popup but it didn't help. Hard power off still causes reboot to Garmin logo without USB, preboot (thumb on G) still enables USB briefly.

Meanwhile, GarminCure3 Cure firmware failed to update; progress bar instant 0-100%, nuvi 2589 displays SOFTWARE LOADING..., (USB drops) spin a while, then updater pops up a vague "There is a problem with this GARMIN application" error message.

I downloaded RGN_Tool, opened nuvi25x9C4_880.gcd, saved boot.bin, opened boot.bin in a hex editor, and found the first occurrence of 0C 80 A0 E1. Is that the right thing to patch? What do I do with the edited boot.bin ?

[Butters]

Yeah the "T" word here gets more adverse attention than a fox in a henhouse.

This is now seeming to be a more complex problem than a simple soft-bricking caused by a corrupt file which can't be loaded during the boot process.

The behavior you're seeing is worryingly indicative of physical flash damage affecting an essential firmware region, probably region 14 (0x0E) which is absolutely essential for not only normal operation of the device but is also needed to load essential files during bootup as well as executing the loader to start the boot process. Flash damage would explain why Cure3 fw can't be properly loaded too, because if firmware regions can't be overwritten during a flash then corrupt region data is permanent and cannot be corrected by 'soft-methods'. The fact that Windows is reading it in Device Manager as an unknown USB device isn't encouraging either, it should identify itself as GARMIN Device when connected to a PC in preboot.

Let's not give up just yet however. Sometimes severe corruption in one of the firmware regions can mimic flash damage although usually that's corrected by flashing an original fw RGN in preboot. One way to be sure is to use a ramoader (aka boot.bin/Ldr.bin) modified so that it doesn't rely at all on the firmware to initiate a flash. Seeing you have already looked into the modified boot.bin method you likely do understand the reason why we should try that. Please follow the following guide exactly:

• In a hex editor open boot.bin, find the first instance of 0C 80 A0 E1 and change E1 to E3 in that string. Save the change.
• Open the modified boot.bin in RGN_Tool and save it as an RGN file (use only safe naming of 194701000880.rgn), put it aside for now.
• Make a common text file with Notepad, you can copy and paste the text in the coded box below but make sure there are no additional spaces at line ends or carriage returns in it after you save it:

Please Login or Register to see the links

• Save the file as update.txt.
• Open the firmware file "nuvi25x9C4_880.gcd" in RGN_Tool and save the last 3 sections as 7F.bin, 9E.bin and 0E.bin respectively, naming the BINs after their region hex names and taking care not to mix up the naming (and note "0" is zero). See image below for reference:
  
  Spoiler: Click for Pic

  [Only registered and activated users can see links. ]
• Prepare a small microSD card, preferably no bigger than 8GB ensuring that it's freshly formatted FAT32 and make a root-folder named Garmin in it. In Garmin folder make a sub-folder named Updater and in Updater make another named 1947 (i.e. Garmin\Updater\1947).
• In the root of the card (with, not in, the Garmin folder) copy 7F.bin, 9E.bin and 0E.bin.
• Place update.txt file in the 1947 folder. There should be nothing else in that folder.
• Ensure the device is fully OFF, hold the power button down to force it off if necessary, then insert the microSD.
• Open 194701000880.rgn in Updater.exe, select the radio button for "USB Device" and position your mouse cursor over the OK button of the GUI.
• Connect a 'known-good' USB cable to the PC, plug in the device and place it in preboot watching the window next to "USB Device" and IMMEDIATELY it becomes populated click OK on the GUI or hit the Enter key on PC's keyboard to initiate the flash.
• Keep holding the screen firmly until you receive a popup message from Updater.exe, hopefully it shows that the flash completed successfully.

If that fails please post a copy of the text in "update.log" if it has been generated in 1947 folder.

An explanation of the text commands: The first one will delete a GCD file from the device if there's one present. The second will save a copy of the device's non-volatile memory as a BIN file into the root of the microSD card. The 3rd, 4th & 5th will attempt to write good data to the device's 3 firmware regions. Final one will attempt a reboot.

Good luck. Ask in advance if you're unsure of anything before doing this and be aware it's all at your entire risk although i've been as careful as possible in my instructions. It shouldn't make anything worse however. If the read/write command/s fail/s it will definitively show physical flash damage.

[VintageDave]

Beautifully explained, Butters. (And I'm sure I've seen similar/exact same before but I couldn't have pulled it together without handholding.)

Two questions.

1. My smallest microSD card is 16GB SDHC. If this card is contraindicated, I'll buy a tiny card and report back after I've tried it.

2. I don't think I can put my nuvi 2589 in a stable power-down state; it powers right back up every time. Can you suggest a sequence? Here's my guess:
a. Starting with the instrument in hung state;
b. Open 194701000880.rgn (patched boot.bin only) in Updater.exe, select "USB Device" and hover on OK button;
c. Connect USB cable;
d. insert microSD;
e. Position thumb ready to touch G;
f. Press and hold Power button until screen goes blank;
g. The instant the Garmin logo reappears, touch and hold G;
h. The instant the device appears in Updater, click OK;

Do I need another power cycle to execute the new boot.bin and in turn, the region bins in the SD card?

If you believe the "hung" device is waiting to mangle the SD card as soon as it sees it, what can I do? That "device available" window is literally just a few seconds - if I insert the card then, I'll have no chance to start Updater. Wish I had an assistant... or four hands.

[Butters]

The 16GB card should be ok to use, generally up to 32GB sized cards of most brands can be read by Garmin devices for flashing and even larger cards, i.e. 64GB> provided they're reformatted from exFAT to FAT32. I've had less problems overall using small cards when flashing however.

The uncontrolled boot-looping is another troubling sign TBH. Holding the power button down until the screen goes black should not result in it re-starting spontaneously, it should be hard-off. Regardless, let's press on.

Your suggested sequence is exactly what you need to do. With boot-looping devices i've sometimes had to use my elbow to hit the Enter key, i know exactly that feeling of needing a third hand to click OK fast enough on the GUI.

Even boot-looping devices that can be shut down completely are tricky because they will immediately start looping again when powered on either by the button or a 5V USB connection. Spontaneously powering on adds usually more workload than 2 hands can handle easily so practice the keyboard elbow method. Be persistent because it might take numerous attempts to catch the right moment but provided it's a software problem, once data transmission commences from Updater.exe to the device the boot process will be suspended and the flash will finish without interruption. Unfortunately flashing will always fail if it's a hardware fault causing such behavior. Good luck.

PS: Almost forgot about this:

Do I need another power cycle to execute the new boot.bin and in turn, the region bins in the SD card?

If you believe the "hung" device is waiting to mangle the SD card as soon as it sees it, what can I do? That "device available" window is literally just a few seconds - if I insert the card then, I'll have no chance to start updater .....

If the flash starts and isn't interrupted each command will be executed in turn and if one cmd fails the next will be attempted. However if the flash starts and then fails before finishing there may be a "last_id.bin" file written to the 1947 folder on the card. If so you must remove that BIN before another flash or it's presence will not allow another attempt. If you mean do you need to start from scratch on each flash attempt, then yes. The device must be in a 'power-off' state at least momentarily before each attempt. Your chances of catching it between boot-loops are slim to none so powering it off by holding the button down and then quickly connecting the USB gives you a decent chance. BTW, if the device stays off as long as you hold the button down that'll give you the best chance of all.

There should be no 'mangling' of the card caused by the device boot-looping (i guess you mean causing corruption).

[VintageDave]

The darn thing won't stay off even if I hold the power button. It's definitely not the way that button should work.

I will try tomorrow morning when I'm fresh. I propose to reuse the 16GB card currently holding my map. It's already FAT32, I'll just backup and erase, and copy in the new payload with the patched boot.bin .

I started to ask about flashing one region at a time, in separate attempts, but if as you say a successful start of script processing captures the system until the whole script has run through, I don't see any point in trying piecemeal if the whole-thing-at-once fails. What do you think?

[Butters]

The darn thing won't stay off even if I hold the power button. It's definitely not the way that button should work.

No it's not how it should work, and it's also indicative of a hardware problem.

I will try tomorrow morning when I'm fresh. I propose to reuse the 16GB card currently holding my map. It's already FAT32, I'll just backup and erase, and copy in the new payload with the patched boot.bin .

Don't put the patched boot.bin in the card! The flash in this instance is initiated by the RGN file made from it. There should be nothing else in the 1947 folder with the update.txt file. See post #350, one of the step points in that list is this: "Place update.txt file in the 1947 folder. There should be nothing else in that folder.". I suppose i should have highlighted that more.

I started to ask about flashing one region at a time, in separate attempts, but if as you say a successful start of script processing captures the system until the whole script has run through, I don't see any point in trying piecemeal if the whole-thing-at-once fails. What do you think?

Technically there's no reason not to run separate commands but it's probably pointless from a practical aspect. Likely either the flash will start and successfully finish, or it'll fail immediately after it initiates.

[VintageDave]

Right, I misspoke. I did not put the patched boot.bin on the card.

I saved the patched boot.bin to boot.rgn then renamed that to 194701000880.rgn in updater.exe's directory.
I installed the MicroSD card containing 0E.bin 7F.bin 9E.bin in the root directory and Update.txt in Garmin\Updater\1947 .
I started updater.exe .
I power-cycled the device and put it in preboot.
When the device appeared in the USB box, I clicked OK then YES. (YES to the prompt asking should it go ahead since the version was the same.)

Nothing visible happened after updater.exe finished "successfully". I did a hard power off, the device powered right back up and hung at the logo as usual. I waited a while, then pulled the card and examined it on my computer. It was unchanged, no log file, no 154 file.

As far as I can tell, the device didn't execute the script. At least, it didn't write anything to the card, and it didn't reboot. Same old behavior of hanging at the logo on a normal boot or hanging at LOADING... on a preboot.

I tried a cut-down script that just copies out region 154 and reboots, then an even more cut-down script that just reboots. No change in behavior, and no log file.

Is there anything else to try, or is the patient dead?

[Butters]

So no commands were executed or even initiated, not even the simple ones, i.e. the read cmd to dump non-vol nor the delete one for gupdate.gcd removal either probably.

It is definitely flash damage. That's now unmistakably 100% confirmed by the results. The use of the modded RGN negates involvement of the device firmware in the initiation of the flash. So Updater has just assumed the requested flashing has happened and falsely reports success. All the other previous indications were circumstantial evidence of such damage but also could have been due to severe region data corruption. Sorry.

Garmin does have an exchange service (they don't repair individual devices) but my experience is that the cost of that is more than buying a similar working used device on the open market. It's possible to replace the flash chip but specialized equipment and skills are needed for that. You probably should just strip it for parts and sell them on eBay. The screen and battery in particular have some second-hand value, even other bits like the USB socket and the case parts maybe. If you sell the whole unit make sure to divulge in the listing that it has physical damage to the flash and only it's useful to strip for parts otherwise someone might buy it thinking it's just soft-bricked and it'll bob up here or another similar forum again - trust me, that's happened more than once and wasted a lot more time.