asus rt-ac87u: USB drive unmounts + no space left on device

[Martin - SNBuser]

Hi all,

I've got the ASUS RT-AC87U router installed with Merlin-FW and entware (for installing packages with "opkg install"). My problems:

1) Very often the USB3-stick I've inserted in either of the USB2/3-ports get automatically disconnected (unmounted). If I try to reboot, it normally isn't enough. I have to physically turn off the router and wait maybe 10-30 seconds and start up again. Then it finds the USB-drive and mounts it correctly. This is a problem - I don't know the solution. Anyone have any experience?

2) Yesterday I installed some new packages with "opkg install" (entware). I think 2 times in the middle of the process, the USB-stick was unmounted automatically. So I had to restart/turn off and turn on. Eventually, now I have no space left:

/dev/root 43.9M 43.9M 0 100% /
devtmpfs 124.7M 0 124.7M 0% /dev
tmpfs 124.8M 5.9M 119.0M 5% /tmp
/dev/mtdblock4 62.8M 2.0M 60.7M 3% /jffs
/dev/sda1 117.6G 163.9M 111.5G 0% /tmp/mnt/Router_ext2

How do I clean up the small flash drive disk and put as much as possible on the >100 GB USB flash drive?
(I would like to avoid a hard reset and to re-install everything, I hope there are some temporary files somewhere I just need to delete)... Example (I believe lack of space causes the "Input/output error"-message:


wrt54g@router:/tmp/home/root# opkg install gcc libcc boost
Installing gcc (5.3.0-3) to root...
Downloading http://pkg.entware.net/binaries/armv7/gcc_5.3.0-3_armv7soft.ipk.
Installing binutils (2.25-1) to root...
Downloading http://pkg.entware.net/binaries/armv7/binutils_2.25-1_armv7soft.ipk.
Unknown package 'libcc'.
Installing boost (1_60_0-2) to root...
Downloading http://pkg.entware.net/binaries/armv7/boost_1_60_0-2_armv7soft.ipk.
/opt/tmp/opkg-VRDGUf/boost_1_60_0-2_armv7soft.ipk: Input/output error
Collected errors:
* wfopen: //opt/lib/opkg/info/binutils.control: Input/output error.
* wfopen: /opt/bin/addr2line: Input/output error.
* wfopen: /opt/bin/as: Input/output error.
* wfopen: /opt/bin/c++filt: Input/output error.
* wfopen: /opt/bin/elfedit: Input/output error.
* wfopen: /opt/bin/gprof: Input/output error.
* wfopen: /opt/bin/ld: Input/output error.
* wfopen: /opt/bin/ld.bfd: Input/output error.
* wfopen: /opt/bin/nm: Input/output error.
* wfopen: /opt/bin/objcopy: Input/output error.
* wfopen: /opt/bin/ranlib: Input/output error.
* wfopen: /opt/bin/readelf: Input/output error.
* wfopen: /opt/bin/size: Input/output error.
* wfopen: /opt/bin/strings: Input/output error.
* wfopen: /opt/bin/strip: Input/output error.
* set_flags_from_control: Failed to open //opt/lib/opkg/info/binutils.control: No such file or directory.
* pkg_write_filelist: Failed to open //opt/lib/opkg/info/binutils.list: No such file or directory.
* opkg_install_pkg: Failed to extract data files for binutils. Package debris may remain!
* opkg_install_cmd: Cannot install package gcc.
* opkg_install_cmd: Cannot install package libcc.
* opkg_download: Failed to download http://pkg.entware.net/binaries/armv7/boost_1_60_0-2_armv7soft.ipk, wget returned 1.
* opkg_install_pkg: Failed to download boost. Perhaps you need to run 'opkg update'?
* opkg_install_cmd: Cannot install package boost.
* opkg_prep_intercepts: Failed to make temp dir /opt/tmp/opkg-VRDGUf/opkg-intercept-R8nivN: Input/output error.
* opkg_conf_write_status_files: Can't open status file //opt/lib/opkg/status: No such file or directory.
* opkg_conf_write_status_files: Can't open status file /opt/tmp//opt/lib/opkg/status: Input/output error.
* rm_r: Failed to read dir /opt/tmp/opkg-VRDGUf: Input/output error.
wrt54g@router:/tmp/home/root# opkg list
-sh: opkg: not found
wrt54g@router:/tmp/home/root#
wrt54g@router:/tmp/home/root# reboot
wrt54g@router:/tmp/home/root# Connection to router closed by remote host.
Connection to router closed.
martin@HPdesk:~$ ssh wrt54g@router
wrt54g@router's password:


ASUSWRT-Merlin RT-AC87U 380.59-0 Tue May 10 15:46:11 UTC 2016
wrt54g@router:/tmp/home/root# opkg install gcc libcc boost
-sh: opkg: not found


In the end you see that the USB drive got automatically unmounted. After reboot, it is still unmounted. When I try to use "opkg install" it says opkg is not found (because soft reboot is not enought, need to be unplugged for 10-30 seconds)...

Please help with suggestions/advice,
thanks.

Br,
M

 

[ColinTaylor]

If the USB drive is un-mounting in the way you describe then there is no point trying to diagnose any potential entware issues. But it suggests that there is some sort of physical problem with the drive rather than the router.

Post your syslog for the period you think it is un-mounting. We should be able to see something.

BTW, your USB drive has plenty of space left, 163.9M used out of 117.6G available (99.8% free). /jffs has 3% used and /tmp 5% used. / (root) is 100% used because it is a ROM.

 

[Martin - SNBuser]

If the USB drive is un-mounting in the way you describe then there is no point trying to diagnose any potential entware issues. But it suggests that there is some sort of physical problem with the drive rather than the router.

Post your syslog for the period you think it is un-mounting. We should be able to see something.

BTW, your USB drive has plenty of space left, 163.9M used out of 117.6G available (99.8% free). /jffs has 3% used and /tmp 5% used. / (root) is 100% used because it is a ROM.

I did a reset to factory defaults - and then read this... It is really good to hear about the 100% "space used" on root (that it is not a problem), thank you very much...

I still do not understand the "Input/output error"-messages, but suspect it has something to do with the USB drive disconnecting... I wrote it was unmounting, but it is actually more like it's physically disconnecting because I think I remembered that blkid did not show anything. I noticed that the USB3-drive gets really really hot, so now I bought a cable to have a distance from the USB-port to the USB-drive. I think this is better and healthier for the USB-stick/drive. Another thing I started doing is to delete the USB-stick and recreate the EXT2-filesystem:

mkfs.ext2 -ccv -L Router_ext2 /dev/sdc1
mke2fs 1.42.13 (17-May-2015)
/dev/sdc1 contains a ext2 file system labelled 'Router_ext2'
last mounted on /mnt/exchange on Sat Jul 2 17:21:16 2016
Proceed anyway? (y,n) y
fs_types for mke2fs.conf resolution: 'ext2'
Filesystem label=Router_ext2
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
7831552 inodes, 31325952 blocks
1566297 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
956 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
... etc...

(still not finished after almost 4 hours)... About the syslog : it it the file in /tmp/syslog.log ? Because that file has something from August 1st (weird!) and then July 2nd. July 2nd is today, but I think the log starts after I did the factory reset from the webinterface... I hoped maybe other with the same router had noticed the same problem and then they would write here. On monday I'll pickup a USB3-cable I ordered, to get the USB-stick away from the USB-port as something makes the USB-stick really really hot... Maybe the current is wrong inside, I'll know more about that in the coming week after I've picked up the USB3-cable. If anyone knew about the input/output-error I would also be very interested in hearing about that. Otherwise, I'm about to re-install entware also and see what happens... Thanks a lot for your comment(s), ColinTaylor...

 

[ColinTaylor]

Yes the syslog is /tmp/syslog.log. It is also the same thing that you can see in the web interface.

/tmp is stored in memory so its contents are lost when the router is rebooted. As the router doesn't have a battery backed clock, when it is rebooted the syslog will show the default date/time (August 1st) up until it can connect to the internet to get the current time.

 

[Martin - SNBuser]

I ran this command mkfs.ext2 -ccv -L Router_ext2 /dev/sdc1 and after around 6 hours it finished (128 GB USB3-stick). The -cc should mean:
-c Check the device for bad blocks before creating the file system. If this option is specified twice,
then a slower read-write test is used instead of a fast read-only test.

The result is:

root@HP:/mnt# mkfs.ext2 -ccv -L Router_ext2 /dev/sdc1
mke2fs 1.42.13 (17-May-2015)
/dev/sdc1 contains a ext2 file system labelled 'Router_ext2'
last mounted on /mnt/exchange on Sat Jul 2 17:21:16 2016
Proceed anyway? (y,n) y
fs_types for mke2fs.conf resolution: 'ext2'
Filesystem label=Router_ext2
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
7831552 inodes, 31325952 blocks
1566297 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
956 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Filesystem UUID: 58e950f9-98bf-4d2f-b3da-13e0dcd8002b
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424, 20480000, 23887872

Running command: badblocks -b 4096 -X -s -w /dev/sdc1 31325951
Testing with pattern 0xaa: done
Reading and comparing: done
Testing with pattern 0x55: done
Reading and comparing: done
Testing with pattern 0xff: done
Reading and comparing: done
Testing with pattern 0x00: done
Reading and comparing: done
Warning: the backup superblock/group descriptors at block 262144 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 393216 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 425984 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 1409024 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 1638400 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 2097152 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 2392064 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 2523136 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 2785280 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 2883584 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 3080192 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 3342336 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 3538944 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 3964928 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 3997696 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 4423680 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 4718592 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 4816896 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 5046272 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 6324224 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 6389760 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 6422528 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 6455296 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 6914048 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 7471104 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 7962624 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 8028160 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 9011200 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 9764864 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 10452992 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 11796480 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 11829248 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 12353536 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 12386304 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 12812288 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 12845056 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 13271040 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 13500416 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 13664256 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 14778368 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 15007744 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 15597568 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 16220160 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 16580608 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 17203200 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 17825792 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 19267584 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 19628032 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 19857408 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 21266432 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 21561344 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 21790720 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 22216704 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 22478848 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 23494656 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 23724032 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 24051712 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 25133056 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 25657344 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 26411008 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 26476544 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 26705920 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 27688960 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 27852800 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 28508160 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 28573696 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 28672000 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 29229056 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 29458432 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 30310400 contain
bad blocks.

Warning: the backup superblock/group descriptors at block 31031296 contain
bad blocks.

Allocating group tables: done
Writing inode tables: done
Writing superblocks and filesystem accounting information: done

root@HP:/mnt#
​

I think/believe this will mark bad blocks so hopefully that'll not be a problem in the future. The annoying thing is that it's a relatively new USB3-stick - maybe 3-4 months old. I think it might have been damaged because I've noticed it was really really hot, after running for a while and when I took out the USB-stick from the RT-AC87-router - maybe 50-60 deg. C... Maybe this high temperature can have caused bad blocks? Maybe high temperatures also caused the router to automatically disconnect the USB-stick? I'm currently using a cable to make a distance from the router to the USB-stick, I hope the problem is not too high currents from the router. With this distance, I hope the USB-stick will operate much cooler and I think this could maybe solve the problem. Maybe high USB temperature also explains the input/output errors (due to bad blocks, maybe?)...

Phew, that's a lot of work. Now I'll try to re-install entware and my packages and see what happens... Thanks a lot for advice about /tmp - as I did a factory reset the router restarted so I cannot see if there's anything in the syslog about the usb disconnecting. I'll keep and eye and write back if it happens again, within a few days... In the meanwhile, I'll try to reinstall my entware/opkg packages.

Thanks for the help!

 

[Martin - SNBuser]

Yes the syslog is /tmp/syslog.log. It is also the same thing that you can see in the web interface.

/tmp is stored in memory so its contents are lost when the router is rebooted. As the router doesn't have a battery backed clock, when it is rebooted the syslog will show the default date/time (August 1st) up until it can connect to the internet to get the current time.

Hi ColinTaylor (or another wise person),

Oh, please help... Now it happened again with the USB3-drive (Samsung Flash Drive FIT, the first line below)... From my /tmp/syslog.log:

Jul 3 12:10:08 kernel: scsi 0:0:0:0: Direct-Access Samsung Flash Drive FIT 1100 PQ: 0 ANSI: 6
Jul 3 12:10:08 kernel: sd 0:0:0:0: Attached scsi generic sg0 type 0
Jul 3 12:10:08 kernel: sd 0:0:0:0: [sda] 250609664 512-byte logical blocks: (128 GB/119 GiB)
Jul 3 12:10:08 kernel: sd 0:0:0:0: [sda] Write Protect is off
Jul 3 12:10:08 kernel: sd 0:0:0:0: [sda] Assuming drive cache: write through
Jul 3 12:10:08 kernel: sd 0:0:0:0: [sda] Assuming drive cache: write through
Jul 3 12:10:08 kernel: sd 0:0:0:0: [sda] Assuming drive cache: write through
Jul 3 12:10:08 kernel: sd 0:0:0:0: [sda] Attached SCSI removable disk
Jul 3 12:10:09 usb: USB ext2 fs at /dev/sda1 mounted on /tmp/mnt/Router_ext2.
Jul 3 12:10:09 rc_service: hotplug 16152:notify_rc restart_nasapps
Jul 3 12:10:09 iTunes: daemon is stopped
Jul 3 12:10:09 FTP Server: daemon is stopped
Jul 3 12:10:09 Samba Server: smb daemon is stopped
Jul 3 12:10:09 kernel: gro disabled
Jul 3 12:10:09 Timemachine: daemon is stopped
Jul 3 12:10:09 kernel: gro enabled with interval 2
Jul 3 12:10:11 Samba Server: daemon is started
Jul 3 12:10:41 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:10:41 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:10:41 kernel: usb 2-2: device descriptor read/64, error -71
.....
(keeps going on around 3 pages, same error -71 message)
(keeps going on around 3 pages, same error -71 message)
(keeps going on around 3 pages, same error -71 message)
.....
Jul 3 12:14:09 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:14:10 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:14:14 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:14:16 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:14:17 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:14:17 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:15:04 kernel: usb 2-2: device descriptor read/64, error -110
Jul 3 12:15:04 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:15:04 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:15:04 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:15:05 kernel: usb 2-2: device not accepting address 2, error -71
Jul 3 12:15:06 kernel: usb 2-2: device not accepting address 2, error -71
Jul 3 12:15:06 kernel: end_request: I/O error, dev sda, sector 23092248
Jul 3 12:15:06 kernel: Buffer I/O error on device sda1, logical block 2886296
Jul 3 12:15:06 kernel: lost page write due to I/O error on sda1
Jul 3 12:15:06 kernel: Buffer I/O error on device sda1, logical block 2886337
Jul 3 12:15:06 kernel: lost page write due to I/O error on sda1
Jul 3 12:15:06 kernel: end_request: I/O error, dev sda, sector 23092752
Jul 3 12:15:06 kernel: Buffer I/O error on device sda1, logical block 2886452
Jul 3 12:15:06 kernel: lost page write due to I/O error on sda1
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=720913, block=2885665
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=720914, block=2885665
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=720915, block=2885665
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=720916, block=2885665
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=720917, block=2885665
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=720918, block=2885665
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=720919, block=2885665
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=573442, block=2293762
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=573441, block=2293762
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=679938, block=2719746
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=679937, block=2719746
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=598017, block=2392608
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=630785, block=2525378
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=6709251, block=26836994
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:06 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:15:06 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:15:07 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:07 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:07 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:07 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:07 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:15:07 kernel: usb 2-2: device descriptor read/64, error -71
Jul 3 12:15:08 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:08 hotplug[17094]: USB partition unmounted from /tmp/mnt/Router_ext2 fail. (return -1, Bad file descriptor)
Jul 3 12:15:08 rc_service: hotplug 17094:notify_rc stop_nasapps
Jul 3 12:15:08 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:08 kernel: usb 2-2: device not accepting address 5, error -71
Jul 3 12:15:08 kernel: usb 2-2: device not accepting address 6, error -71
Jul 3 12:15:08 kernel: hub 2-0:1.0: unable to enumerate USB device on port 2
Jul 3 12:15:08 iTunes: daemon is stopped
Jul 3 12:15:08 FTP Server: daemon is stopped
Jul 3 12:15:09 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:09 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=6709250, block=26836994
Jul 3 12:15:09 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:09 kernel: EXT2-fs (sda1): error: ext2_get_inode: unable to read inode block - inode=6709251, block=26836994
Jul 3 12:15:09 syslog: USB partition unmounted from /tmp/mnt/Router_ext2
Jul 3 12:15:09 rc_service: hotplug 17094:notify_rc restart_nasapps
Jul 3 12:15:09 rc_service: waitting "stop_nasapps" via hotplug ...
Jul 3 12:15:09 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:09 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:09 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:09 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:09 kernel: EXT2-fs (sda1): previous I/O error to superblock detected
Jul 3 12:15:09 kernel: usb 3-2: device descriptor read/64, error -62
Jul 3 12:15:09 kernel: usb 3-2: device descriptor read/64, error -62
....

A lot of USB errors.... I would REALLY REALLY appreciate any help with these USB-errors, thanks!

 

[octopus]

When i have that kind of trouble i refomat with "Paragon Partition Magic"
1: Delete Partition
2: Create Partition (ext2) Primary
3: Change volyme label
4: Check File System Integrity
5: Done

 

[Martin - SNBuser]

So you think it's the usb, not the router? It's a very new usb3-stick... 128 GB. I spend around 6 hours on this: mkfs.ext2 -ccv -L Router_ext2 /dev/sdc1 , so I thought all the bad sectors where labelled as bad in the partition table...

 

[octopus]

So you think it's the usb, not the router? It's a very new usb3-stick... 128 GB. I spend around 6 hours on this: mkfs.ext2 -ccv -L Router_ext2 /dev/sdc1 , so I thought all the bad sectors where labelled as bad in the partition table...

Yes I think your problem is on your USB-drive. Worth a try to see if it get fixed.

 

[Martin - SNBuser]

Yes I think your problem is on your USB-drive. Worth a try to see if it get fixed.

One thing is that I received these bad blocks (according to mkfs.ext2 -ccv -L Router_ext2 /dev/sdc1)... But I've never had/tried a malfunctioning USB-drive. Why should it not be possible to just mark the bad blocks in the partition table and re-use the USB-stick? It's causing me a lot of pain. But maybe I should try with another USB-stick, this time with long cable to avoid excessive heat from the router... I'm not happy that I should try to return the USB-stick, if I'm not convinced this is the problem... I'm not sure I can get the USB-stick replaced as it's around 7 months old and warranty is 6 months - but I've never tried a malfunctioning USB-stick - never heard of it, too.

 

[ColinTaylor]

I doubt that marking the bad blocks will help as I suspect that the address of the "bad" blocks will keep changing. It's possible that the USB drive is heating up to much and pushing the RAM chips beyond their operational range. Or there might be fractures in the soldered joints and the change in temperature is causing the connections to become intermittent.

The only problem with this theory is that you said you had the same problem with a different USB drive.

I think you need to eliminate the computer you have been using to do the formatting/testing in case there's a problem with that. If you have another PC running Windows, try just formatting it as NTFS and then doing a chkdsk. If that works OK try a surface test as suggested by @octopus.

 

[octopus]

The USB memory becomes hot is usually a sign of bad blocks.
It may be sufficient to reformat according to my proposal.
@Martin - SNBuser

 

[john9527]

Just a couple more possibilities over a bad USB....
- The router USB drivers may just be incompatible with the chipset/firmware on the USB. A hint of this is in the syslog
Jul 3 12:10:41 kernel: usb 2-2: device descriptor read/64, error -71
This has happened before with the latest rev of the Sandisk Ultra-fit where people are reporting problems having it recognized. I've also personally had a problem with a Lexar 32GB flash where it would 'just disappear' when used on the router, but worked fine under Windows.
- The router USB drivers may just be unable to deal with a 128GB flash chip. I don't have any to test with (largest I have is an older version of the Sandisk at 64GB which works).

 

[sfx2000]

I've also personally had a problem with a Lexar 32GB flash where it would 'just disappear' when used on the router, but worked fine under Windows.

FWIW - I've had a couple of Lexar 32GB's that behaved oddly under Linux - they're odd drives in USB3, work a bit better under USB2... so I would consider that to be a Lexar problem, not an AsusWRT problem, at least in my case

 

[Martin - SNBuser]

The USB memory becomes hot is usually a sign of bad blocks.
It may be sufficient to reformat according to my proposal.
@Martin - SNBuser

But the USB stick was brand new when I started... Anyway, I hear you - I'll try to download "Paragon Partition Magic" and 1: Delete Partition, 2: Create Partition (ext2) Primary, 3: Change volyme label, 4: Test surface, 5: Done (never heard of Paragon or surface test before, I'll try it one of these days). Thanks...

 

[Martin - SNBuser]

Just a couple more possibilities over a bad USB....
- The router USB drivers may just be incompatible with the chipset/firmware on the USB. A hint of this is in the syslog
Jul 3 12:10:41 kernel: usb 2-2: device descriptor read/64, error -71
This has happened before with the latest rev of the Sandisk Ultra-fit where people are reporting problems having it recognized. I've also personally had a problem with a Lexar 32GB flash where it would 'just disappear' when used on the router, but worked fine under Windows.
- The router USB drivers may just be unable to deal with a 128GB flash chip. I don't have any to test with (largest I have is an older version of the Sandisk at 64GB which works).

Thanks... Yes, maybe you're right... By "disappearing" I mean, it doesn't even unmount (if it did/does, I would still be able to see it with blkid, and therefore I discovered it "disappeared" although it was still connected physically)... Thanks for your ideas... Still not easy to know exactly the cause... I can add that today I inserted a USB-3 cable to get a distance from the router to the USB-stick. Until now (3-4 hours the USB-stick didn't disappear and there were no heat problems)... But I still think it can happen one of the coming days, if I play around with opkg so I'll watch out and see...

 

[sfx2000]

I ran this command mkfs.ext2 -ccv -L Router_ext2 /dev/sdc1 and after around 6 hours it finished (128 GB USB3-stick). The -cc should mean:
-c Check the device for bad blocks before creating the file system. If this option is specified twice,
then a slower read-write test is used instead of a fast read-only test.

Shouldn't be using ext2 with a flash drive - might consider ext3 (with journaling) or ext4...

 

[ColinTaylor]

Shouldn't be using ext2 with a flash drive - might consider ext3 (with journaling) or ext4...

Personally I'd agree with RMerlin and say that ext2 is preferred over ext3/4 for flash drives.

http://www.snbforums.com/threads/usb-file-system.15134/#post-102395

 

[octopus]

I use ext4 with journaling off and write back mode enabled. I got great speed.

To turn off journal and speed things up on a ext4 partition follow these steps:

Example uses sda1 device, this does not mean you should do it on your main disk,
replace sda1 with the disk you are preparing, data loss will happen!

Create ext4 fs on /dev/sda1 disk

mkfs.ext4 /dev/sda1

To enable writeback mode use this.
This mode will typically provide the best ext4 performance.

tune2fs -o journal_data_writeback /dev/sda1

To delete journal remove has_journal option,
this will disable etx4 journal

tune2fs -O ^has_journal /dev/sda1

Check if everything is ok

e2fsck -f /dev/sda1

mount
/dev/sda1 on /tmp/mnt/rt-ac68 type ext4 (rw,nodev,relatime,barrier=1,data=writeback)

Untested: (tested 16-09-2016 working)
mount -o remount,noatime /dev/sda1 /mnt/your-mount-point-here

Running a 400MB file write
Average (W):     17,59 MB/sec
-----------------------------
Running a 400MB file read
Average (R):     24,21 MB/sec
-----------------------------

@Martin - SNBuser

 

[mattiL]

To reduce wear I think you should also use the option 'noatime' when mounting the USB stick.
It disables writing last file access information.