Bug 6820

Created attachment 2190 [details]
Runtime tests log

Changed title to EdgeRouter and attached the runtime tests log.

I can not reproduce this issue on my board. Can you tell me the file system type for sda3 in your board ?  Maybe you should reformat the sda3.

############################
root@edgerouter:~# fdisk -l /dev/sda

Disk /dev/sda: 3.6 GiB, 3880452096 bytes, 7579008 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x668a0900

Device    Boot     Start       End  Blocks  Id System
/dev/sda1           2048    292863  145408   c W95 FAT32 (LBA)
/dev/sda2         292864   5175295 2441216  83 Linux
/dev/sda3        5175296   7579007 1201856  83 Linux

root@edgerouter:~# dmesg | grep sda
[    4.262048] sd 0:0:0:0: [sda] 7579008 512-byte logical blocks: (3.88 GB/3.61 GiB)
[    4.268571] sd 0:0:0:0: [sda] Write Protect is off
[    4.272121] sd 0:0:0:0: [sda] Mode Sense: 23 00 00 00
[    4.272474] sd 0:0:0:0: [sda] No Caching mode page found
[    4.276495] sd 0:0:0:0: [sda] Assuming drive cache: write through
[    4.284395] sd 0:0:0:0: [sda] No Caching mode page found
[    4.288417] sd 0:0:0:0: [sda] Assuming drive cache: write through
[    4.294487]  sda: sda1 sda2 sda3
[    4.300045] sd 0:0:0:0: [sda] No Caching mode page found
[    4.304154] sd 0:0:0:0: [sda] Assuming drive cache: write through
[    4.308958] sd 0:0:0:0: [sda] Attached SCSI removable disk
[    9.309781] EXT4-fs (sda3): mounted filesystem with ordered data mode. Opts: (null)
[    9.371508] EXT4-fs (sda2): mounted filesystem with ordered data mode. Opts: (null)
[    9.489658] FAT-fs (sda1): Volume was not properly unmounted. Some data may be corrupt. Please run fsck.
root@edgerouter:~# 
root@edgerouter:~# mount | grep sda
/dev/sda3 on /run/media/sda3 type ext4 (rw,relatime,data=ordered)
/dev/sda2 on /run/media/sda2 type ext4 (rw,relatime,data=ordered)
/dev/sda1 on /run/media/sda1 type vfat (rw,relatime,gid=6,fmask=0007,dmask=0007,allow_utime=0020,codepage=437,iocharset=iso8859-1,shortname=mixed,errors=remount-ro)
root@edgerouter:~# 
#############################

Can't reproduce this issue. Need more info.

Lucian, can you try again in our HW rack and see if that reproduces?

Thanks!

This error still reproduces on the EdgerRouter in our rack, but it may be a problem with that particular system. Unfortunately I don't have another one at the moment to test but untill I get one I will assign this to myself.

After a little search I believe that this is not really an error but a warning that the kernel probes for ext2 and ext3 before ext4 on the partition. But it mounts it successfully after that:

[   25.354011] EXT3-fs (sda2): error: couldn't mount because of unsupported optional features (240)
[   25.363414] EXT2-fs (sda2): error: couldn't mount because of unsupported optional features (240)
[   25.389842] EXT4-fs (sda2): mounted filesystem with ordered data mode. Opts: (null)
[   27.728541] EXT4-fs (sda2): re-mounted. Opts: (null)


The partition is ext4 and working fine:

root@edgerouter:~# mount | grep sda
/dev/sda2 on / type ext4 (rw,relatime,data=ordered)

A possible fix for this could be using the kernel parameter CONFIG_EXT4_USE_FOR_EXT23=y.

I think this was fixed in oe-core ad9f54a5014f2d997165d0cfd6ad7c05e62f315b.

This issue is reproducible on Cherryhill with corei7-64-lsb image.

BUILD: meta-intel: 158516f3b89eafa222e64de90de383cabc8c2594
ENVIRONMENT: corei7 on Cherryhill
Image: core-image-lsb-sdk

The log can be found at this link: 
http://yocto-lava.rb.intel.com/dashboard/attachment/6778/view

This was REOPENED

Possibly relevant:

commit 474f66c59ee641a503b5fd62ef87436e4a62d4f5
Author: Bruce Ashfield <bruce.ashfield@windriver.com>
Date:   Fri Aug 21 14:18:12 2015 -0400

    linux-yocto/4.1: enable DRM_CIRRUS_QEMU and EXT4_USE_FOR_EXT23

    Updating the meta SRCREV to enable cirrus graphics emulation and wider
    compatibility for the ext4 driver.

    [YOCTO: #7348]
    [YOCTO: #6667]

    (From OE-Core rev: db02cf0734ce059b7bf47edeb1107e23e20e7160)

Further review of this seems to indicate that while these look like errors they 
are more informational when mounting an actual ext4 FS, we are correctly 
setting the EXT4_USE_FOR_EXT23 config.

What happens is that from the initial boot the rootfs comes from rootfs.img on 
a vfat mount. Because the rootfs.img is a loopmount the rootfstype is not used 
at it's no an actual rootfs, I know this sounds strange, but true. Since the 
mount command reads the filesystem types and then tries each type until it 
works we get these messages since it tries ext3 and ext2 before actually 
getting to the ext4 and mounts rootfs correctly.

We can safely white list these.

Hi Alex,
Can you retry this with M2.rc4 on CherryHill and Edgerouter?

Cheers,
AlexG

Fixed in oe-core 4e59b6f66eeae256d4e62f4741725cdebb53cbf7.