I've had my Pine running Ubuntu for a while now. Out of nowhere I noticed some programs not running correctly. While the general system boots up fine and MOST things look OK. I get this error on a few different programs:
Code: [ 45.593149] sunxi-mmc 1c0f000.sdmmc: smc 0 p0 err, cmd 18, RD DTO !!
[ 45.605412] sunxi-mmc 1c0f000.sdmmc: data error, sending stop command
[ 45.618610] mmcblk0: timed out sending r/w cmd command, card status 0x900
[ 45.631346] mmcblk0: not retrying timeout
[ 45.640936] blk_update_request: 115 callbacks suppressed
[ 45.640942] end_request: I/O error, dev mmcblk0, sector 14137176
[ 45.652834] appdaemon[987]: unhandled level 3 translation fault (7) at 0x00417508, esr 0x92000007
[ 45.652840] pgd = ffffffc033186000
[ 45.661887] [00417508] *pgd=0000000074a36003, *pmd=0000000074b69003, *pte=0000000000000000
Even trying to run as simple as lsb_release gives me a bus error:
Code: ubuntu@pine64:~$ lsb_release -a
Bus error
Which also causes a similar error in dmesg:
Code: [ 180.946771] sunxi-mmc 1c0f000.sdmmc: smc 0 p0 err, cmd 18, RD DTO !!
[ 180.959150] sunxi-mmc 1c0f000.sdmmc: data error, sending stop command
[ 180.971652] mmcblk0: timed out sending r/w cmd command, card status 0x900
[ 180.984500] mmcblk0: not retrying timeout
[ 180.994239] end_request: I/O error, dev mmcblk0, sector 14137176
[ 181.006184] lsb_release[1429]: unhandled level 3 translation fault (7) at 0x00417508, esr 0x92000007
[ 181.006194] pgd = ffffffc0348d3000
[ 181.015216] [00417508] *pgd=00000000790e2003, *pmd=00000000740a0003, *pte=0000000000000000
[ 181.029721] CPU: 3 PID: 1429 Comm: lsb_release Tainted: G C O 3.10.102-2-pine64-longsleep #66
[ 181.029728] task: ffffffc03310a940 ti: ffffffc0363e0000 task.ti: ffffffc0363e0000
I'm pretty much at a loss for what else I can try. Its odd because most of the system is working but I can't seem to update anything. Any ideas?
(05-11-2018, 06:29 PM)fsa317 Wrote: Code: [ 180.946771] sunxi-mmc 1c0f000.sdmmc: smc 0 p0 err, cmd 18, RD DTO !!
[ 180.959150] sunxi-mmc 1c0f000.sdmmc: data error, sending stop command
[ 180.971652] mmcblk0: timed out sending r/w cmd command, card status 0x900
[ 180.984500] mmcblk0: not retrying timeout
[ 180.994239] end_request: I/O error, dev mmcblk0, sector 14137176
I'm pretty much at a loss for what else I can try. Its odd because most of the system is working but I can't seem to update anything. Any ideas?
Have you tried re-seating the emmc card?
It may also be failing.
Is the EMMC card the SD card? I've taken the SD card out a few times and even tried inserting into my laptop and running disk check on it. No change.
(05-11-2018, 07:00 PM)fsa317 Wrote: Is the EMMC card the SD card? I've taken the SD card out a few times and even tried inserting into my laptop and running disk check on it. No change.
Emmc is always mmcblk0 and sdcard is always mmcblk1
It looks like one of these:
https://www.pine64.org/?product=16gb-emmc
(05-11-2018, 07:11 PM)evilbunny Wrote: (05-11-2018, 07:00 PM)fsa317 Wrote: Is the EMMC card the SD card? I've taken the SD card out a few times and even tried inserting into my laptop and running disk check on it. No change.
Emmc is always mmcblk0 and sdcard is always mmcblk1
It looks like one of these:
https://www.pine64.org/?product=16gb-emmc
First, I appreciate your help, I'm going nuts here.
I'm on a pine64, I dont think I have an emmc card. Output of /etc/fstab appears to show mmcblk0 with 2 partitions, pretty sure this is my SD card.
Code: ubuntu@pine64:/$ more /etc/fstab
# <file system> <dir> <type> <options> <dump> <pass>
/dev/mmcblk0p1 /boot vfat defaults 0 2
/dev/mmcblk0p2 / ext4 defaults,noatime 0 1
(05-11-2018, 07:46 PM)fsa317 Wrote: Output of /etc/fstab appears to show mmcblk0 with 2 partitions, pretty sure this is my SD card.
Code: ubuntu@pine64:/$ more /etc/fstab
# <file system> <dir> <type> <options> <dump> <pass>
/dev/mmcblk0p1 /boot vfat defaults 0 2
/dev/mmcblk0p2 / ext4 defaults,noatime 0 1
If that's the case your sdcard may be on the way out, have you tried another card?
(05-11-2018, 08:13 PM)evilbunny Wrote: (05-11-2018, 07:46 PM)fsa317 Wrote: Output of /etc/fstab appears to show mmcblk0 with 2 partitions, pretty sure this is my SD card.
Code: ubuntu@pine64:/$ more /etc/fstab
# <file system> <dir> <type> <options> <dump> <pass>
/dev/mmcblk0p1 /boot vfat defaults 0 2
/dev/mmcblk0p2 / ext4 defaults,noatime 0 1
If that's the case your sdcard may be on the way out, have you tried another card?
Would I need to start with a new fresh OS image on a new card or I could I try to clone this card to a new physical SD card (this way I don't need to reconfigure everything).
(05-11-2018, 08:20 PM)fsa317 Wrote: Would I need to start with a new fresh OS image on a new card or I could I try to clone this card to a new physical SD card (this way I don't need to reconfigure everything).
You might be able to clone it, or at least copy most of the data, I'm unsure about other OS, but there is a number of linux utils which will try to rescue everything on the card.
Code: gddrescue - GNU data recovery tool
myrescue - rescue data from damaged disks
safecopy - data recovery tool for problematic or damaged media
05-12-2018, 04:11 AM
(05-11-2018, 07:11 PM)evilbunny Wrote: Emmc is always mmcblk0 and sdcard is always mmcblk1
It looks like one of these:
https://www.pine64.org/?product=16gb-emmc
Now now @ evilbunny... I'm the one who's supposed to confuse the boards... unless he has the LTS board, there's no eMMC on the pine64 
@ fsa317 In the first instance, I'd just start from a new image on a new card, preferably the same one you started from before, so that you can firmly point the finger at the old SD card probably failing. Then you can try recovering the card as evilbunny said. I'd also add testdisk to the list as a possible rescue tool. When you do get it up and running again, keep in mind that even though SD cards (or at least the better ones) have wear leveling build in to the SD card control chip, that they do have a limited number of write cycles, so you can to minimise unnecessary writes to the SD card. So things like log2ram can help a lot. Also not having databases stored on the SD card if possible - shove them on a hard drive if you can - even on another machine via NFS or SMB if you have to.
Thanks for all the help. Here is what worked for me.
I took the sd card and loaded into another machine (actual a windows machine running virtualbox) and ran fsck.ext4 on the partition. There were a bunch of errors it tried to fix.
Upon reloading it seems to be working although I'm concerned about this happening again. What is the best way to "backup" the SD card image periodically?
|
