| 1 |
On 03/09/13 11:26, meino.cramer@×××.de wrote: |
| 2 |
> William Kenworthy <billk@×××××××××.au> [13-09-03 05:08]: |
| 3 |
>> On 03/09/13 10:45, meino.cramer@×××.de wrote: |
| 4 |
>>> walt <w41ter@×××××.com> [13-09-03 04:15]: |
| 5 |
>>>> On 09/02/2013 09:15 AM, meino.cramer@×××.de wrote: |
| 6 |
>>>>> The rootfs and $HOME of my embedded system is stored |
| 7 |
>>>>> on a 16GB SD-card (about 5GB used, rest free). The FS |
| 8 |
>>>>> is ext4. |
| 9 |
>>>>> |
| 10 |
>>>>> Since the system hangs for unknown reasons several times |
| 11 |
>>>> Does it hang at a predictable point, like during boot, or poweroff? |
| 12 |
>>>> |
| 13 |
>>>> I know almost nothing about SD cards (yet). Do they develop bad |
| 14 |
>>>> blocks like other storage media? I notice fsck.ext4 has a -c flag |
| 15 |
>>>> to check for bad blocks. |
| 16 |
>>>> |
| 17 |
>>> No, it hangs while compiling or while updateing (eix-sync; emerge ...). |
| 18 |
>>> |
| 19 |
>>> |
| 20 |
>>> I did the following now: |
| 21 |
>>> I did a binary image backup with dd of the sdcard. |
| 22 |
>>> I made a backup of the all files from the bad fs with tar. |
| 23 |
>>> I say "YES" to fsck to fix what it found. |
| 24 |
>>> I made another backup of the all files from the bad fs with tar. |
| 25 |
>>> I md5summed both tar archives and found them identical. |
| 26 |
>>> |
| 27 |
>>> Now...is the conclusion correct, that the identical md5sum |
| 28 |
>>> indicate, that the fixed error of the fs only had impact to |
| 29 |
>>> already invalidated data? |
| 30 |
>>> Or whatelse could this indicate? |
| 31 |
>>> |
| 32 |
>>> Best regards, |
| 33 |
>>> mcc |
| 34 |
>>> |
| 35 |
>>> PS: What come mind just in this moment: |
| 36 |
>>> Can I ran fsck on an binary image of the fs which I made with dd somehow? |
| 37 |
>>> |
| 38 |
>>> |
| 39 |
>>> |
| 40 |
>>> |
| 41 |
>>> |
| 42 |
>> Have you run out of inodes? - ext 4 has had very mixed success for me on |
| 43 |
>> solid state. Running out of inodes is a real problem for gentoo on |
| 44 |
>> smaller SD cards with standard settings. |
| 45 |
>> |
| 46 |
>> BillK |
| 47 |
>> |
| 48 |
>> |
| 49 |
>> |
| 50 |
> Does this error message from fsck indicate that? I am really bad in |
| 51 |
> guessing what fsck tries to cry at me ... ;) |
| 52 |
> |
| 53 |
> |
| 54 |
>>> solfire:/root>fsck.ext4 -f -p /dev/sdb2 |
| 55 |
>>> rootfs: Inodes that were part of a corrupted orphan linked list found. |
| 56 |
>>> |
| 57 |
>>> rootfs: UNEXPECTED INCONSISTENCY; RUN fsck MANUALLY. |
| 58 |
>>> (i.e., without -a or -p options) |
| 59 |
>>> [1] 18644 exit 4 fsck.ext4 -f -p /dev/sdb2 |
| 60 |
>>> |
| 61 |
>>> |
| 62 |
> Is there any way to correct the settings from the default values to |
| 63 |
> more advances ones, which respect the sdcard size of 16GB *without* |
| 64 |
> blanking it...a "correction on the fly" so to say??? |
| 65 |
> |
| 66 |
> And if not: Is there a way to backup the sdcard and playback the files |
| 67 |
> after reformatting it by preserving all three time stamps of the |
| 68 |
> files (atime is deactivated via fstab though) ? |
| 69 |
> |
| 70 |
> Best regards, |
| 71 |
> mcc |
| 72 |
> |
| 73 |
> |
| 74 |
> |
| 75 |
> |
| 76 |
> |
| 77 |
df -i - if you get 100% iUSE or near to it thats your problem ... I have |
| 78 |
seen that error message you give as a result of running out of inodes |
| 79 |
corrupting the FS. |
| 80 |
|
| 81 |
No, your only way out is to copy (I use rync) the files off, recreate |
| 82 |
the fs with max inodes ("man mke2fs") and rsync the files back. Once an |
| 83 |
ext* fs has been created with a certain number of inodes its fixed until |
| 84 |
you re-format. |
| 85 |
|
| 86 |
I get it happening regularly on 4G cards when I forget and just emerge a |
| 87 |
couple of packages without cleaning up in between packages. On 16G |
| 88 |
cards, its compiling something like glibc or gcc that uses huge numbers |
| 89 |
of inodes at times. On a single 32G card I have, the standard settings |
| 90 |
have been fine ... so far :) |
| 91 |
|
| 92 |
Billk |
Archives