[solved]BRICKed: inside eMMC's dead, with E:Can't open /cache/recovery/log

nobody can help me ???

ben_pyett said:

Just in case, I ever get in this situation, how would I manage to achieve this, please could you PM me the answer so that it's not in thread, so that someone doesn't think it's something to try out....thanks

Sent from my HTC Desire S using XDA Premium App

Click to expand...

Click to collapse

Guys..assisting a friend regarding this same problem...can i know this method too before sending back to HTC service centre?? Thank you.

PM me, I'll send it to you.

There's a good chance that it won't work though, if the eMMC is fried it won't be able to flash correctly to it -which is most strange, because flashing back to a stock RUU and HBOOT -if it works and gets applied correctly- seems to "hide" the problem and makes the phone behave like normal..

Tried it this weekend and I couldn't find a single flaw in the behavior of the phone. It's only when you S-OFF and root it again -or try to format the affected partition via the recovery- that it's complaining like hell... :S

And this is most annoying too.. If the phone doesn't show a problem when on S-ON and stock ROM, what'd be your excuse to send it back to the service centre ? :S

Snakeforhire said:

PM me, I'll send it to you.

There's a good chance that it won't work though, if the eMMC is fried it won't be able to flash correctly to it -which is most strange, because flashing back to a stock RUU and HBOOT -if it works and gets applied correctly- seems to "hide" the problem and makes the phone behave like normal..

Tried it this weekend and I couldn't find a single flaw in the behavior of the phone. It's only when you S-OFF and root it again -or try to format the affected partition via the recovery- that it's complaining like hell... :S

And this is most annoying too.. If the phone doesn't show a problem when on S-ON and stock ROM, what'd be your excuse to send it back to the service centre ? :S

Click to expand...

Click to collapse

Have you installed EXT4 recovery and checked the types of file systems that you have on these partitions EXt2/EXT3/ETX4? etc

A different stance would be do the errors ever effect or get in the way of your daily phone use, if not - tempted to ignore them?

Bubaaak said:

I tried this before , but I am still not sure. I've got M4G2DE which is faulty one, but in adb shell I get this - no error.

Code:

dmesg | grep mmc0
<3>[    8.565704] mmc0: No card detect facilities available
<6>[    8.566223] mmc0: Qualcomm MSM SDCC at 0x00000000a0500000 irq 98,0 dma 7
<6>[    8.566467] mmc0: Platform slot type: MMC
<6>[    8.566589] mmc0: 4 bit data mode disabled
<6>[    8.566711] mmc0: 8 bit data mode enabled
<6>[    8.566955] mmc0: MMC clock 144000 -> 50000000 Hz, PCLK 96000000 Hz
<6>[    8.567077] mmc0: Slot eject status = 0
<6>[    8.567291] mmc0: Power save feature enable = 1
<6>[    8.567443] mmc0: DM non-cached buffer at ffa0f000, dma_addr 0x0ba08000
<6>[    8.567687] mmc0: DM cmd busaddr 0x0ba08000, cmdptr busaddr 0x0ba08300
<6>[    8.720886] mmc0: new high speed MMC card at address 0001
<6>[    8.722167] mmcblk0: mmc0:0001 M4G2DE 2.10 GiB

Anyway I'm very interested in that if they will repair it for someone under warranty, because I'm S-OFF yet and now I'm worrying pretty much about that...

Click to expand...

Click to collapse

Hi, my Desire S has the same symptom (same output) as yours. So please confirm if my eMMC was fried or not?

Thanks

ben_pyett said:

Have you installed EXT4 recovery and checked the types of file systems that you have on these partitions EXt2/EXT3/ETX4? etc

A different stance would be do the errors ever effect or get in the way of your daily phone use, if not - tempted to ignore them?

Click to expand...

Click to collapse

Of course, my dear man...
Actually EXT4 was installed there as soon as I rooted the phone. ^^
It's only one of the many things I tried in order to solve it, among reflashing and repartitioning the mtd0, formatting the affected partitions in EXT4 then converting to E3 and back (which makes recovery detect their sizes correctly again, whereas it detected them as "null" before), formatting them manually via 'adb shell root' (using various options like '-cc" to detect the bad blocks and automatically exclude them from the mke2fs.ext4 inode list, and such), using tune2fs to tweak the FS settings... About the only thing I didn't try yet is formatting with btrfs to see if it can circumvent such "shortcomings", which I'm looking into as of now..

"to be continued"... ^^

im having same trouble as posted on 1st post. cant mount /data and /cache. getting same error. so does this mean its an eMMC problem?

BratPAQ said:

im having same trouble as posted on 1st post. cant mount /data and /cache. getting same error. so does this mean its an eMMC problem?

Click to expand...

Click to collapse

I'm afraid that it does, although you really need to give a little more information, as post #1 contains dozens of varying commands sections and errors...

What happened just prior to you experiencing this problem?- this is part for diagnosis of your issue and part as an attempt to work out what not to do in the future!

ben_pyett said:

this is part for diagnosis of your issue and part as an attempt to work out what not to do in the future!

Click to expand...

Click to collapse

I got a ready-made and one-size-fits-all answer to that : 'don't buy HTC hardware, ever, under any circumstance'... :/

Sorry for your loss, BUT!

Snakeforhire said:

I got a ready-made and one-size-fits-all answer to that : 'don't buy HTC hardware, ever, under any circumstance'... :/

Click to expand...

Click to collapse

LOL, that comment may make you feel better , yet it isn't of much help for the rest of us that still own a working device and therefore may face the problem in the future, these being the same people that were prepared to assist you if we we could

well that's the whole point : to spare my troubles to some potential HTC buyers (RUN, ye hapless mortals ! Here there be great danger...) :s

sorry if I sound bitter, but paying 250€ for a piece of **** that don't even last for 3 weeks b4 croaking makes u a bit pissed off (and actually want to give back Taiwan to communist china. bcoz if this is taiwanese quality it'll be just as comfy there, and at least we'll pay much less for their crap... )

When is it time to say something?

Snakeforhire said:

well that's the whole point : to spare my troubles to some potential HTC buyers (RUN, ye hapless mortals ! Here there be great danger...) :s

sorry if I sound bitter, but paying 250€ for a piece of **** that don't even last for 3 weeks b4 croaking makes u a bit pissed off (and actually want to give back Taiwan to communist china. bcoz if this is taiwanese quality it'll be just as comfy there, and at least we'll pay much less for their crap... )

Click to expand...

Click to collapse

Don't feel the need to apologize, if I was in your position, I imagine that I'd wouldn't be so restrained with regard holding back my expletives....(I know the forum has rules....but...).

With the increasing frequency of this identical problem now happening I'm quite amazed that HTC haven't responded or been inundated with angry/livid users screaming at them demanding action!

Is it due to the fact that we've all S-OFF'd and feel the need to be polite as we're aware that we've knowingly broken our warranties....not that that has anything to do with the problem we have then suffered!

Perhaps it's time to start spamming their Facebook Page with questions as to the quality and reliability of said eMMC chips and see what their plans for the future are......we have seen public action change HTC strategy in the past. Whose is to say that it wouldn't happen again!?

ben_pyett said:

Don't feel the need to apologize, if I was in your position, I imagine that I'd wouldn't be so restrained with regard holding back my expletives....(I know the forum has rules....but...).

With the increasing frequency of this identical problem now happening I'm quite amazed that HTC haven't responded or been inundated with angry/livid users screaming at them demanding action!

Is it due to the fact that we've all S-OFF'd and feel the need to be polite as we're aware that we've knowingly broken our warranties....not that that has anything to do with the problem we have then suffered!

Perhaps it's time to start spamming their Facebook Page with questions as to the quality and reliability of said eMMC chips and see what their plans for the future are......we have seen public action change HTC strategy in the past. Whose is to say that it wouldn't happen again!?

Click to expand...

Click to collapse

Sounds like an idea! To find some people doing this on the Facebook page, could help us to get an answer from htc.
If there is more interest I will do it tomorrow.
Cheers


Sent from my HTC Desire S using Tapatalk

Well... after a long night... I had all of the symptoms described by everyone here, however mine came back from the dead after a stock RUU update from fastboot. I *HOPE* at least a small percentage of people were as lucky as me... details on my blog... which I can't link here as this is my first post..

search for it if you're interested (dave heavy industries) for the full recount of events, otherwise the nitty gritty is that after following the instructions to work out of your MMC is screwed, I decided to try and get rid of clockworkmod and try my luck at a warranty job. I couldn't get fastboot to flash boot.img or recovery.img so instead I ran the htc telstra RUU while the phone was in fastboot and BLAMMO! success.
Re-rooted, Re flashed, now 100% up and running... very happy (relieved) camper.

dave303 said:

Well... after a long night... I had all of the symptoms described by everyone here, however mine came back from the dead after a stock RUU update from fastboot. I *HOPE* at least a small percentage of people were as lucky as me... details on my blog... which I can't link here as this is my first post..

search for it if you're interested (dave heavy industries) for the full recount of events, otherwise the nitty gritty is that after following the instructions to work out of your MMC is screwed, I decided to try and get rid of clockworkmod and try my luck at a warranty job. I couldn't get fastboot to flash boot.img or recovery.img so instead I ran the htc telstra RUU while the phone was in fastboot and BLAMMO! success.
Re-rooted, Re flashed, now 100% up and running... very happy (relieved) camper.

Click to expand...

Click to collapse

someone should take note of this and make it sticky, as one of the solution to try, running RRUU while the phone is in fastboot.

Hmm... Me thinks you're getting your hopes up a bit too much..

I experienced the same kind of situation reversal as you did, only to find out that when re-rooted, my DS was just as quirky as before..
It's not giving any "hard errors" messages per se, it's way more insidious than this..
It's all about the phone being generally much more sluggish as before, often lingering at 90-100% CPU (and for a good cause : it has to write the same blocks/inodes X number of times before they actually get written correctly), apps mysteriously "disappearing" (=shown as "not installed" in app manager, or "on SD-card") when they're supposed to be in the internal memory, and so on.

Try connecting to your phone via an adb rootshell console, and do a manual cache wipe on the affected partition(s) (mke2fs.ext4 with your own options if you like, like '-cc', 'writeback' and such), and see if it gives you any errors when formatting it.
I'd bet one of my testi..fingers that it will, it's almost guaranteed.

I'd really love it to be different and to be able to give you hope about this, but sadly, once an eMMC is fried there's no magic "back" button to go back in time... :/

Previously when i face the same problem, event running the stock RUU update from fastboot doesn't help. It hang halfway.

Snakeforhire said:

Hmm... Me thinks you're getting your hopes up a bit too much..

I experienced the same kind of situation reversal as you did, only to find out that when re-rooted, my DS was just as quirky as before..
It's not giving any "hard errors" messages per se, it's way more insidious than this..
It's all about the phone being generally much more sluggish as before, often lingering at 90-100% CPU (and for a good cause : it has to write the same blocks/inodes X number of times before they actually get written correctly), apps mysteriously "disappearing" (=shown as "not installed" in app manager, or "on SD-card") when they're supposed to be in the internal memory, and so on.

Try connecting to your phone via an adb rootshell console, and do a manual cache wipe on the affected partition(s) (mke2fs.ext4 with your own options if you like, like '-cc', 'writeback' and such), and see if it gives you any errors when formatting it.
I'd bet one of my testi..fingers that it will, it's almost guaranteed.

I'd really love it to be different and to be able to give you hope about this, but sadly, once an eMMC is fried there's no magic "back" button to go back in time... :/

Click to expand...

Click to collapse

I'm not saying that my emmc magically un-fried itself (thats crazy!), I'm suggesting that mine was never fried in the first place - although according to the tests provided at the beginning of this thread, it was. I'm also suggesting that something about the htc provided RUU has helped, and if you happen to be in this boat, it certainly wouldn't hurt to try fyi - the phone has been operating for about 7 hours without trouble so far. I'll advise if this changes over the next few days.

Snakeforhire:--

~ # mke2fs -j /dev/block/mmcblk0p27
mke2fs -j /dev/block/mmcblk0p27
mke2fs 1.40.8 (13-Mar-2008)
/dev/block/mmcblk0p27 is apparently in use by the system; will not make a filesy
stem here!
~ # umount /cache
umount /cache
~ # mke2fs -j /dev/block/mmcblk0p27
mke2fs -j /dev/block/mmcblk0p27
mke2fs 1.40.8 (13-Mar-2008)
Filesystem label=
OS type: Linux
Block size=1024 (log=0)
Fragment size=1024 (log=0)
76912 inodes, 307196 blocks
15359 blocks (5.00%) reserved for the super user
First data block=1
Maximum filesystem blocks=67633152
38 block groups
8192 blocks per group, 8192 fragments per group
2024 inodes per group
Superblock backups stored on blocks:
8193, 24577, 40961, 57345, 73729, 204801, 221185

Writing inode tables: done
Creating journal (8192 blocks): done
Writing superblocks and filesystem accounting information: done

This filesystem will be automatically checked every 35 mounts or
180 days, whichever comes first. Use tune2fs -c or -i to override.

~ # e2fsck /dev/block/mmcblk0p27
e2fsck /dev/block/mmcblk0p27
e2fsck 1.41.6 (30-May-2009)
/dev/block/mmcblk0p27: clean, 11/76912 files, 20003/307196 blocks
~ # e2fsck /dev/block/mmcblk0p27 -f
e2fsck /dev/block/mmcblk0p27 -f
e2fsck 1.41.6 (30-May-2009)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
/dev/block/mmcblk0p27: 11/76912 files (0.0% non-contiguous), 20003/307196 blocks

dave303 said:

Re-rooted, Re flashed, now 100% up and running... very happy (relieved) camper.

Click to expand...

Click to collapse

Glad to hear it mate. There's far too many scaremongers on here willing to wave the "your emmc chip is fried" in peoples' faces without thinking.

Peace

I have fired eMMc chip.

opumps said:

hi guys, this problem was solved, unexpectedly.



after read guhl99's post in Thread: PSA: Rooting methods and recently discovered dangersthread,
i finally confirm that my htc DS was bricked by an accidentally freeze and the immediately battery pulled-out-&-pushed-in unstably.
the eMMC-chip(marked as SAMSUNG104, KLM4G2DEJE A001, FDAE15PC) was completely dead,
i made a replacement with a new one(made by Sandisk serial:SEM04G), and then everything goes on good.


View attachment 644685 View attachment 644686



"Thanks to" the design of the battery compartment door, the battery pullout-pushin becomes running-risks.
Make sure there's NOT any tiny battery move when you flipping the compartment door from the side.
Make sure disconnect the battery and DS itself quickly & stably as soon as you could.
Make sure there's at least 10 seconds between battery pullout and pushin.


"How can I be sure if it's a hw problem? I mean, it looks like one, but I would like to be sure..."
please check #27 http://xdaforums.com/showpost.php?p=16135893&postcount=27

Click to expand...

Click to collapse

Please help me. Which Sandisk chip i have to buy (below is the URL) and what is the cost?
http://www.sandisk.com/business-solutions/inand-embedded-flash-drives