Problem with SD Card Read/Write

[adepssimius]

I have been having trouble getting an install of xbian to work on both of my pi's (I have been using 1.0a4). I have a 256 and a 512, and after several reboots they always start to give me journal I/O errors and fail to boot.

The weird thing is it only occurs when I am using an SD card. I copied an installation to an external hdd and I now just use the SD card to boot the root partition stored on the hdd and I have incurred no problems since then.

I have tried three different brands of SD cards but still no luck. I am using an iPad charger as a power supply (rated for 2.3A) so I'm pretty sure that isn't the issue. I am also careful to use the sync command before powering down or unplugging the pi. The issue occurs at the default clock speed as well as when overclocked.

SD Cards tried:
1. Amazon Basics 16GB Class 10
2. Sandisk 16GB UHS Class 1
3. PNY 2GB unknown speed, but slow

Any ideas?

[Koenkk]

Can you remove/change the overclock from your /boot/config.txt? Start with changing core_freq=375 to core_freq=250, if that doesn't solve the problem change arm_freq=840 to arm_freq=800 or even arm_freq=700.

[adepssimius]

It seems that changing the core freq to 250 stopped the errors. I have had no errors through the last 2 days and 20+ reboots. Can you give me some insight as to why that would prevent the errors? Also, how could I speed it back up? I don't want the pi to be running slower than it should.

[Koenkk]

@adepssimius , can you try core_freq=300 and test that? (We are planning to lower the core_freq in beta 1 )

[adepssimius]

I will try it out and get back to you.

Beta 1!?

[Koenkk]

@adepssimius, alpha 5 first

[Richdotward]

Thanks for the tip. Same problem on my pi. Works fine for ages, then a few reboots and refuses to start again.

Will try 700/250.

Update : 2 weeks as of 26th feb and no crashes and no more sd corruption. Pi has been running 24/7 and rebooted a few times.

Update : corruption again so no good. Went with running from laptop HD and working for about one month with no issues.

Thanks

Rich

[adepssimius]

@Koenkk, I have experienced no more journaling issues after running the Rpi at a reduced core speed of 325. Lowering the arm frequency alone didn't have any effect, and it seems to run fine at 840.

[Koenkk]

Good news, beta 1 will use core_freq=300 by default

[hrabeX]

I have the same problem - running crashplan on xbian causes I/O errors and remounting filesystem as read only.
dmesg:
[ 302.569535] EXT4-fs (mmcblk0p2): error count: 9
[ 302.569565] EXT4-fs (mmcblk0p2): initial error at 1365969698: ext4_remount:4584
[ 302.569581] EXT4-fs (mmcblk0p2): last error at 1365969735: ext4_remount:4584
[36046.897283] mmc0: Timeout waiting for hardware interrupt - cmd25.
[36046.901223] mmcblk0: error -110 transferring data, sector 170656, nr 24, cmd response 0x900, card status 0x0
[36046.901735] end_request: I/O error, dev mmcblk0, sector 170664
[36046.901757] end_request: I/O error, dev mmcblk0, sector 170672
[36046.911910] Aborting journal on device mmcblk0p2-8.
[36046.914573] EXT4-fs error (device mmcblk0p2) in ext4_reserve_inode_write:4550: Journal has aborted
[36047.476950] journal commit I/O error
[36047.492303] EXT4-fs error (device mmcblk0p2): ext4_journal_start_sb:349: Detected aborted journal
[36047.492339] EXT4-fs (mmcblk0p2): Remounting filesystem read-only
[36047.496455] EXT4-fs error (device mmcblk0p2): ext4_journal_start_sb:349: Detected aborted journal
[36048.136254] EXT4-fs error (device mmcblk0p2) in ext4_reserve_inode_write:4550: Journal has aborted
[36048.136400] EXT4-fs error (device mmcblk0p2) in ext4_reserve_inode_write:4550: Journal has aborted

[Koenkk]

did you already try the above solutions?

[hrabeX]

Lowerred core_freq to 300 helps, 325 produces rare I/O errors

[CurlyMo]

Please notice that the core_freq will change to 250 instead of 300.