Astrofreak85 Posted January 4, 2016 Posted January 4, 2016 Hi, my installed version is about a year old at least...so I'd like to upgrade without loosing settings and data. Is this possible? If yes, how? In doc. there is only decribte how to update the kernel, nothing else... Thanks, Astro System says: root@cubie:~# cat /etc/issue Debian GNU/Linux CT Debian 2.7 wheezy root@cubie:~# cat /proc/version Linux version 3.4.104-sunxi (root@cubie) (gcc version 4.8.2 (Ubuntu/Linaro 4.8.2-16ubuntu4) ) #1 SMP PREEMPT Fri Sep 26 18:30:39 CEST 2014
wildcat_paris Posted January 5, 2016 Posted January 5, 2016 I hope this may help http://forum.armbian.com/index.php/topic/211-kernel-update-procedure-has-been-changed/ and please, make first a backup of your sdcard / system.
Astrofreak85 Posted January 5, 2016 Author Posted January 5, 2016 thanks, I found this in Documentation, atm it doesn't work cause I have a "litle" problem with dpkg :-/...I'll try later again. How to update scripts etc.?
Igor Posted January 5, 2016 Posted January 5, 2016 Scripts are part of the board support package. For example: If you have cubietruck and wheezy: linux-wheezy-root-cubietruck http://www.armbian.com/kernel/ What's missing are possible new packages: https://github.com/igorpecovnik/lib/blob/master/debootstrap.sh#L135-L141 and this https://github.com/igorpecovnik/lib/blob/master/debootstrap.sh#L143-L154 Some of them are mandatory, so do this install too.
Astrofreak85 Posted January 5, 2016 Author Posted January 5, 2016 (edited) Great! Thanks!... Maybe someone can give me a short hint what to do with this problem: root@cubie:~# apt-get install mc W: Not using locking for read only lock file /var/lib/dpkg/lock E: dpkg was interrupted, you must manually run 'dpkg --configure -a' to correct the problem. root@cubie:~# dpkg --configure -a dpkg: error: unable to access dpkg status area: Read-only file system EDIT: Looks like SD Card is nearly dead or something :-/ : Jan 4 23:40:15 localhost kernel: [ 1300.326224] [mmc-err] smc 0 err, cmd 18, DTO Jan 4 23:40:15 localhost kernel: [ 1300.333891] [mmc-err] In data read operation Jan 4 23:40:15 localhost kernel: [ 1300.343307] [mmc-msg] found data error, need to send stop command Jan 4 23:40:15 localhost kernel: [ 1300.355359] mmcblk0: timed out sending r/w cmd command, card status 0x900 Jan 4 23:40:15 localhost kernel: [ 1300.366282] mmcblk0: not retrying timeout Jan 4 23:40:15 localhost kernel: [ 1300.376002] end_request: I/O error, dev mmcblk0, sector 4267944 Edited January 5, 2016 by Astrofreak85 1
tkaiser Posted January 5, 2016 Posted January 5, 2016 Read-only file system Check using dmesg why your rootfs is mounted read-only and compare the output of /etc/fstab and /etc/mtab. And if you're not suffering from a corrupted filesystem then maybe just a simple mount -v -o remount / might do the trick.
Astrofreak85 Posted January 5, 2016 Author Posted January 5, 2016 dmesg: [53237.773765] [mmc-msg] found data error, need to send stop command [53237.785822] mmcblk0: timed out sending r/w cmd command, card status 0x900 [53237.795748] mmcblk0: not retrying timeout [53237.804772] end_request: I/O error, dev mmcblk0, sector 803008 [53297.788709] [mmc-err] smc 0 err, cmd 18, DTO [53297.796397] [mmc-err] In data read operation [53297.805799] [mmc-msg] found data error, need to send stop command [53297.822319] mmcblk0: timed out sending r/w cmd command, card status 0x900 [53297.833150] mmcblk0: not retrying timeout [53297.842215] end_request: I/O error, dev mmcblk0, sector 803008 [53357.784933] [mmc-err] smc 0 err, cmd 18, DTO [53357.792632] [mmc-err] In data read operation [53357.802046] [mmc-msg] found data error, need to send stop command [53357.814113] mmcblk0: timed out sending r/w cmd command, card status 0x900 [53357.829171] mmcblk0: not retrying timeout [53357.838258] end_request: I/O error, dev mmcblk0, sector 803008 fstab: root@cubie:~# cat /etc/fstab # UNCONFIGURED FSTAB FOR BASE SYSTEM /dev/mmcblk0p1 / ext4 defaults,noatime,nodiratime,data=writeback,commit=600,errors=remount-ro 0 0 /dev/sda1 /opt ext4 defaults 0 2 mtab: root@cubie:~# cat /etc/mtab rootfs / rootfs rw 0 0 /dev/root / ext4 ro,noatime,nodiratime,errors=remount-ro,commit=600 0 0 devtmpfs /dev devtmpfs rw,relatime,size=1023780k,nr_inodes=182162,mode=755 0 0 sysfs /sys sysfs rw,nosuid,nodev,noexec,relatime 0 0 tmpfs /run tmpfs rw,nosuid,noexec,relatime,size=131072k,mode=755 0 0 tmpfs /run/lock tmpfs rw,nosuid,nodev,noexec,relatime,size=5120k 0 0 proc /proc proc rw,nosuid,nodev,noexec,relatime 0 0 configfs /sys/kernel/config configfs rw,relatime 0 0 tmpfs /run/shm tmpfs rw,nosuid,nodev,noexec,relatime,size=131072k 0 0 devpts /dev/pts devpts rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmode=000 0 0 /dev/sda1 /opt ext4 rw,relatime,data=ordered 0 0 tmpfs /tmp tmpfs rw,nosuid,nodev,relatime,size=1048576k 0 0 OWFS /mnt/1wire fuse.OWFS rw,nosuid,nodev,relatime,user_id=0,group_id=0,allow_other 0 0 Look like SD is dying (?!) Hopefully I can still make an image... :-/
tkaiser Posted January 5, 2016 Posted January 5, 2016 Look like SD is dying (?!) Hopefully I can still make an image... :-/ Yes. Or maybe you use a fake card (only half the capacity as advertised with) and exceeded this limit. It's alway a good idea to check every flash based media prior to using it. Check the FAQ: http://www.armbian.com/documentation/ At this point I would rely on ddrescue to try to create an image, get a new card and then proceed.
Astrofreak85 Posted January 5, 2016 Author Posted January 5, 2016 Card is a Sanddisk Card...but it's now 2 years old (or even older...), since one year systems runs from SD (since the internal NAND died..)...strange Maybe a good start for a clean system...only some little things to do, everything importand are scripts on HDD or in Backup
tkaiser Posted January 5, 2016 Posted January 5, 2016 I's no problem to get a full fake card showing a major brand logo BTW: I usually let this job run weekly on Armbian servers to store the rootfs on SD card to a connected SATA/USB disk or another system: /usr/bin/rsync -aAX --delete --delete-excluded --exclude={"/dev/*","/proc/*","/sys/*","/tmp/*","/run/*","/mnt/*","/media/*","/lost+found","/var/log.hdd","/var/swap"} /* /mnt/sda/sd-card-backup/rootfs-$i/ && touch /mnt/sda/sd-card-backup/rootfs-$i/.last_successful_run $i is either 1 or 2 in rotation and the job stops automatically triggering an email notification when either rsync exits != 0 or the first mmc error occurs parsing dmesg after the run.
PaceyIV Posted January 5, 2016 Posted January 5, 2016 Why you run a weekly backup of your sd card instead of using rootfs directly on the sata hard drive?
tkaiser Posted January 5, 2016 Posted January 5, 2016 Why you run a weekly backup of your sd card instead of using rootfs directly on the sata hard drive? Since it's a HDD I neither like the LCC problem (that many drives show: parking the heads after a few seconds of inactivity just to unramp them a few seconds later until parked to death) nor constantly spinning disks. When nothing's to do the drive has to go into standby/sleeping state (spiining down completely and only consuming 0.1W in most cases). Would it be an SSD instead the rootfs would be there (and a sync to SD card would happen). Armbian's default settings (tmpfs, ramlog, 600 seconds commit interval) are perfect to stay with the rootfs on SD card even when a HDD is connected. That does not apply to most other distros available for SBCs.
PaceyIV Posted January 5, 2016 Posted January 5, 2016 Hum... My first board was an IGEPv2 with it's original rootfs on a new SD card. The card died after only a year, and the board was slow. I move the rootfs on a notebook hard drive and I speed up my board. I haven't tried armian image directly on the SD card: i'm using rootfs on a Western Digital Red hard drive.
Recommended Posts