Jump to content

arox

Members
  • Posts

    391
  • Joined

  • Last visited

Posts posted by arox

  1. Possibly more glaringly it could be a security issue for those unaware or not paying attention while using things like openvpn.

    Well, for MY security, I wouldn't rely on automatic network configuration inherited from the black box of my internet provider ...

  2. Thanks for all your answers.

     

    Yes, the system is installed on a Sandisk Ultra card.

    I mount simply by using the mount command. No automounter script or something.

     

     

    Here is the armbianmonitor output. It is not directly after the disk disappears, but i think there is a lot going on in the last bit with ata1 / SDA / Sata / Ext4.  

    I guess the problem starts at around 6652.232267. 

    http://sprunge.us/FCiR

    I also have ata link reset but dont find failed reset as in your logs :

     

    [ 6707.788508] ata1: hard resetting link

    [ 6710.006605] ata1: SATA link down (SStatus 1 SControl 310)

    [ 6715.016886] ata1: hard resetting link

    [ 6717.238422] ata1: SATA link down (SStatus 1 SControl 310)

    [ 6722.247381] ata1: hard resetting link

    [ 6724.469028] ata1: SATA link down (SStatus 1 SControl 310)

     

    And the no "ata1.00: disabled".

     

    You should perhaps disable spin down to verify if it is really related.

  3. For now I try everything for learning purposes ... I want to make a Web server to show information from temperature sensors. I also want to control the outputs via web page ... I want to install lighttpd. Is it a good choice?

    If you know javascript, have a look at "node.js" (and modules like express.js to serve static files). For personal use, you surely dont need Apache.

     

    The "trick" with SATA is that H3 are "attractive" because people think a computer is just CPU power, and think they will do the same things with a 10W powered computer without cooling than with a 500W with fans and giant heatsinks everywhere. So it is somehow a commercial trick, and users loves to be tricked !

     

    But IO for a data server performance is a question of "bus". And for IO, H3 (unusable 1.5 GHz quad core) is a far worse choice than A20 (usable 1GHz bi core). Anyway, if you are the only one using the server, and do not manipulate Gigabytes files you can use poor SATA usb-bridged implementation - but probably dont need more storage than eMMC (unless using intensively updating databases).

  4. First device with Allwinner's R40 is unfortunately made by one very careless vendor. It's called Banana Pi M2 Ultra to confuse customers: http://forum.banana-pi.org/latest (just have a look through the forums to get the idea what to expect).

     

    I would better look for this Kickstarter here if you want fast SATA and fast GbE: http://www.cnx-software.com/2016/09/23/marvell-espressobin-board-with-gigabit-ethernet-sata-pcie-and-usb-3-0-to-launch-for-39-and-up-crowdfunding/

    Yes this card seems well targeted for server use cases where there is not much solutions available. But "kickstarter" means "perhaps in 6 months or a year" if I am not mistaken. And "development board" means "use at your own risks" and "we dont provide support for software" ...

  5. If I where you ...

     

    I would install /boot and root fs on eMMC (with script= and mount HDD on /home (by editing fstab). I think you should get faster boot and you could the boot your system by inserting the original SD in case of problem.

     

    You may have power problem : if you power the hdd threw the card and run at full cpu load with wifi, usb devices, desktop, hdmi-vga converter ... If you have bad contacts, bad PSU and your Amperage become too high when writing to disk ... you may have to reinstall every other day ...

  6. Related read (about Firefox but applies to Chromium too): https://www.servethehome.com/firefox-is-eating-your-ssd-here-is-how-to-fix-it/

    And that is for IO. If you use ff on a badly cooled card with bad throttling protection, always have a "cpu meter" on your desk if you dont want ff to burn your board. There are a lot of pages with looping, experience enhancing, sniffing javascript. And no protection or easy way to kill these scripts or hinder periodic recall of the script.

  7. Patience is the key.

    Hope this helps.

    I saw by googling, that users with similar problem had found it may resume after 15 minutes ...

     

    But :

    - if you got no specific response from armbian gurus, they probably think it is yet another problem they answered hundreds of times in this forum (mainly PSU, SD card quality and SD card flashing). They are doing a great job, but a bit overwhelmed by hardware problems or software problems relevant to the debian mother distro.

    - your boot process hang in kernel init somewhere in disp or sd setup. I got the same with another card because of power problem as the card needs suddenly much power. It is not just a problem of PSU, but also cables, contacts or internal card design, and as the problem is transient it is hard to prove. How did you get the log ? If you have a serial console, try to boot with nothing attached, always disconnect serial and ethernet after a boot hang, and check your barrel power connector.

  8. Hi! I have a Banana Pi Pro running last Armbian v5.20 via SD card. There is 2 usb HD 2TB connected.I create a software RAID1 /dev/md0 with the 2 usb hd dev/sda1 & dev/sdb1 mounted as /mnt/md0. Up to now all is working fine.

    My goal, put the root on that raid1.

    The script nand-sata-install does not reconize 'md' device, only 'sd' device so i did a very litte modification to the script to reconize 'md'. The script run fine and it transfert all from the SD card to the 'md' (raid) but after the reboot,,,,, nothing, the system wont start. :(

    I am not a linux expert but i think that the problem is that the raid must be start from /boot (still on SD card) before the system try to mount root?

    note: running nand-sata-install to put root on a none-raid partition on thoses drives is working very well but putting root on a raid partition, ex: 'md0' is not working.

    Can somebody help me?

     

    Daniel

    Well, if you are not a linux expert, you take great risk in installing the system on a mirror if it is not handled by the distro ! (and even if you are a linux expert and this is handled by the distro)

     

    For that to be possible, you probably need a specific initrd (this is a compressed fs used to load in ram the mandatory drivers for complex/unusual configurations).

  9. custom version + other images = not compatible

    debian (armbian) + speeding up boot = not compatible

    appliance (mp3 player) + debian = not compatible

    custom + give a try = not compatible

     

    A linux kernel needs about 2 seconds to init. I had stripped down a raspbian with a cam to make an appliance able to start in about that time. For that, you will find that you need to remove udev (and of course the init sytem (sysV, upstart, initng or systemd), which means that your system is no more compatible with any distro.

  10. ...

    if you have configuration like this then definitely you should use other, dedicated method of backup, not simple interactive script).

    ...

     

    In fact, I now avoid complicated configurations. I am an adept of the KISS philosophy (keep it simple stupid).

     

    BTW, (@tkaiser) what about compressed fs for archives ? Compression, reliability, performance ? (Not for physical archives but for rsync copies).

  11. For "slices and overlays" I have no experience (so you may elaborate). What I can guess that overlay file must exist on mounted rw block device (or any other rw device/network).

     

    The script has very precise purpose - create an image of unmounted block device (disk with partitions). Here is an example that loop devices with partitions will not be visible:

     

    root@PKSERVER:/data/temp/device# ll
    total 7696
    -rw-r--r-- 1 root root 268435456 Sep 21 11:08 block_device.img
    root@PKSERVER:/data/temp/device# losetup -f -P block_device.img
    root@PKSERVER:/data/temp/device# lsblk
    NAME         MAJ:MIN RM   SIZE RO TYPE MOUNTPOINT
    sda            8:0    1 115.5G  0 disk
    └─sda1         8:1    1 115.5G  0 part /data
    loop0          7:0    0   256M  0 loop
    └─loop0p1    259:0    0 252.9M  0 loop
    mmcblk0boot0 179:16   0     4M  1 disk
    mmcblk0boot1 179:32   0     4M  1 disk
    mmcblk0      179:0    0  14.6G  0 disk
    └─mmcblk0p1  179:1    0  14.4G  0 part /
    root@PKSERVER:/data/temp/device# backup_block_device.sh
    List of block devices:
    NAME         MAJ:MIN RM   SIZE RO TYPE MOUNTPOINT
    sda            8:0    1 115.5G  0 disk
    └─sda1         8:1    1 115.5G  0 part /data
    loop0          7:0    0   256M  0 loop
    └─loop0p1    259:0    0 252.9M  0 loop
    mmcblk0boot0 179:16   0     4M  1 disk
    mmcblk0boot1 179:32   0     4M  1 disk
    mmcblk0      179:0    0  14.6G  0 disk
    └─mmcblk0p1  179:1    0  14.4G  0 part /
    
    No devices available!
    
    root@PKSERVER:/data/temp/device# fdisk -lu /dev/loop0
    
    Disk /dev/loop0: 256 MiB, 268435456 bytes, 524288 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: 0x66d10c8d
    
    Device       Boot Start    End Sectors   Size Id Type
    /dev/loop0p1       2048 520000  517953 252.9M 83 Linux
    
    root@PKSERVER:/data/temp/device#
    
    

    You know, all I said is because I am a moron, and although I got professional experience on the matter, I always make archive and forget about it, what there were for, and if it is useful to keep them.

     

    I just had a rapid sight at your script and I dont want to tell stupid things. AFAIK, a block device is just a pointer on a driver instance that **can** handle block operations on a raw file/device and cache blocks. It can as well point to an oracle DB file, a filesystem a full disk with boot block, partition table and partitions, a part of a raid/lvm system, a filesystem that is used to compose another filesystem, on your etc/passwd copied to a partition. I dont even know what lsblk would report in the last case but I would guess "part/data".

     

    Nevertheless, you will certainly know what you used your disks or partitions for. But if you are just half as forgetful as I am, you wont remember it next year. So I just recommend to save information when you create big archives that will encumber your disks and may need very different handling to restore them. It is only practical experience : I loose (or may have loose - I dont even know) data because I had not well documented the structure and drives used in a RAID/LVM installation some years ago. Physical archives are more difficult to manage than logical ones (files, or db exports)

     

    BTW, in 1435503102ce64000028_2016-09-19_1474302693, what is 1474302693 ? Is that the compressed size (that you already can know by the file size) ? You can of course retrieve the part size by decompressing threw wc, but there is useful information that you only can store by entering threw keyboard.

  12. Well ...

     

    - you dont provide a usage manual for your script.

    - raw archives need a full understanding of partitions and filesystems !

    - when you copy and compress a fs, you loose some crucial information : the size of partition needed to restore it. You need to store that somehow because you will end up with archives which you dont know why you have done it, and that require long and tedious operation to restore or simply identify.

    - it is always much more difficult to restore and manage archives than to make archives ...

  13. "By adding a simple USB-Ethernet adapter and using this as WAN port you're able to implement 'router behaviour'"

     

    In fact, with my Internet box, I dont even need an USB/Ethernet adapter as it can do IP over USB with speed greater than DSL link. (OTH I dont rely on standards boot scripts to ensure security and want to have full control over it).

     

    I was interested by R1 and I could use it even with its flawed design. But in fact the sale price is then double that what it should be, so I never bought one.

     

    And by the way, can it be correctly cooled when you enclose it with a SATA drive in an acrylic box ?

  14. Benchmark are crap !

     

    They do not represent real usage case and in any case I am not going to install a 250 DB database on an 8 GB SD card or eMMC. For me, what import is the response of the OS and the use as login directory.

     

    Even the more crappy SD card has better response time than a hard disk drive (for limited short file access) because of the rotational latency of the drive. The problem of IO (disk, net, or other) is not the access time but the service time (queuing + operation) when the queue is saturated.

     

    When a mono-server get just the number of requests it can theoretically handle, the wait queue become infinite and so the service time ! Does any benchmark simulate a consequent number of independent random operations and test the effect of a heavy transfer which saturate the queue at the expense of the other processes ? I never saw such reports although it is what import really. And it depends of the device, the controllers, the bus and the drivers.

     

    I have an SSD on an intel Atom board : the system responsiveness is excellent. I have a (cheap) eMMC on a bpi m2+ : the responsiveness is good (but I dont use heavy graphical desktop or even browser - so maybe it is bad in those cases)

  15. It looks super nice, but dunno so you needed 3 arm boards. Maybe its better to buy 1 intel-pc-stick  :) , what do you think  :D.

    - 1 RPI and 1 Lamobo = 2

    - I would bet RPI for image processing capabilities and Lamobo for SATA

     

    Thats the problem with cheap cards : you have to make a choice for available hardware and must do with what exists.

  16.      1) does it exist an HDMI to sound 3.5mm jack adapter ?

        

         2) which USB bluetooth dungle works perfectly on ARMBian I may use to pair this board to a portable bluetooth speaker ?

     

     

    1) HDMI VGA converters have a jack output (some, not all have it). Beware to verify that they dont need to much power ! Better by a cheap usb DAC as recommended by IGOR or a good one if you will connect to a good sound system.

     

    2) I use Logilink BT0015A (ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle) USB 2.0 Bluetooth 4.0 +EDR Micro Class 1 (6 euros)

     

    Bluetooth and "Works perfectly" is a good joke ! But I use it for pan service, with a good range and reliable connection with other nodes with class 1 dongles. It should of course work for a2dp "IF" you get a version of bluez where this has not be broken.

     

    Tips :

    - on my bpi m2+ I had to disable internal BT in script.bin in order to allow bluetoothd to reference the usb dongle correctly.

    - internal BT card may work - but not for me with pan service anyway (bluez bug)

    - the bt0015 is a class 1 device (100m) : the range should be a little better than with class 2, but your speaker is certainly class 2, so you will be lucky if you get 5m range.

  17. If only midori/epiphany would work properly, that would definitely be an improvement. Sadly, it segfaults randomly.

    Its not just a question of stability. When you browse the web, you want to be able to display every pages. The browser has to implement most of the moronic standards in use to format and style pages and execute javascript code. And even then, if the browser respects standards, you will only have a usable browser if websites test their pages against that browser.

     

    So, you have firefox or chromium. I never could use chromium because of memory leaks (the tabs where killed). You have to do with firefox :

     

    - use the card with the fastest CPU and huge heatsink (mulicore not usable with ff - 2 cores usable for sys/net/io/display AND js)

    - go in preference and "about:config" to disable all that is not needed

    - install extension to filter pubs and sniffing js (uBlock, disconnect, noscript ...) or use some frontend proxy like privoxy.

    - assure you have fast disk io for .mozilla and .cache

    - forget about video and perhaps remove pluggins or block them

    - compile it from source with optimization and disable what you can

    - wait for a multi-threaded stable version

     

    Edited on a BPI M1 (A20 2x1 GHz 1Go DDR3) - web filtered against js abuse. Quiet usable (but I browse youtube with another machine : atom 1.6 GHz and ff compiled with gentoo)

×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines