zador.blood.stained Posted February 14, 2016 Posted February 14, 2016 I'm compiling with 4.3.3 config (will post after it finishes), if it doesn't help, this may be kernel related and it's better to ask for help in linux-sunxi google group. [deleted, does not work]
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 And one more compilation from me ... http://mirror.igorpecovnik.com/test/variant-2.zip I tried this one first and still no drive. I will try zador's now!
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 Crikey I dont get this support for paid products! OK installed the old kernel from zador and the drive still does not show up.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 I'm compiling with 4.3.3 config (will post after it finishes), if it doesn't help, this may be kernel related and it's better to ask for help in linux-sunxi google group. Edit: https://www.dropbox.com/sh/nzshly303o0zklz/AABynxOZHvLfKt-IygF8_INka?dl=0 I'll try this one now.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 I'm compiling with 4.3.3 config (will post after it finishes), if it doesn't help, this may be kernel related and it's better to ask for help in linux-sunxi google group. Edit: https://www.dropbox.com/sh/nzshly303o0zklz/AABynxOZHvLfKt-IygF8_INka?dl=0 This also failed. I am not sure what this means at this point. Over to you guys. Thanks Gary
zador.blood.stained Posted February 14, 2016 Posted February 14, 2016 This means that unless it's u-boot or boot script related, it's a regression in kernel itself. Since neither me nor Igor (I believe) don't have this device, and this issue doesn't reproduce on other devices so far, it will take some time to resolve this.
Igor Posted February 14, 2016 Posted February 14, 2016 Uff. I would really like to recreate this error but am running my Banana (M1 not PRO) from SATA drive Jessie / 4.4.1 (downloaded from my download site) Thinking.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 Is that a banana pi or pro Igor? I would be happy to donate a pro if this helps. Thanks Gary
Igor Posted February 14, 2016 Posted February 14, 2016 I don't have Lamaker Banana PRO but almost all other variants So far we never had a need but in this case it would help.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 How about I donate the cost of a pro to you?
Igor Posted February 14, 2016 Posted February 14, 2016 If you donate for a board I am going to get one but the main problem is time. It's hard to get Banana PRO in local store ... If I order, it would take up to one week, whole next week I am out of office ... http://www.pollin.de/shop/dt/MjE2NzkyOTk-/Bausaetze_Module/Entwicklerboards/Banana_Pi/Banana_Pro_Dual_Core_1_GB_DDR3_SATA_G_LAN_WLAN.html From here I got stuff in 3-4days. We need to solve this problem in less than two weeks I also count that this is not a single incident and people will come forward with the exact same problem or without a problem with same hardware combination.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 I can give you access to the board by ssh root if like. I trust you. Would this help? I will donate £40 (what it would cost me) anyways. I will do that later today.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 Another option is i post my board to you tomorrow and you should get it within a few days? I have 2. One is in use as a NAS/PLEX/VIRTUALMIN - works superbly with armbian! The other is a developemnt/test board (how i spotted the problem with the upgrade to 5.00) This board would not be missed as my NAS is working fine atm.
Igor Posted February 14, 2016 Posted February 14, 2016 I tested SATA with kernel 4.4.1 on: Banana M1, M1plus, Cubieboard 1 ... can't reproduce. Hmm, can you swap board with your live system? I'll change topic to attract testers.
zador.blood.stained Posted February 14, 2016 Posted February 14, 2016 Works on cubietruck too. There are some commits that may be responsible for this (especially since /proc/interrupts lacks SATA entry), but this needs a) confirming that it is 100% kernel related (installing only linux-image and linux-dtb on top of 4.5 image) and then using git bisect to find commit that broke it.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 I tested SATA with kernel 4.4.1 on: Banana M1, M1plus, Cubieboard 1 ... can't reproduce. Hmm, can you swap board with your live system? I'll change topic to attract testers. I Will swap it out but am running bananananian at the moment and its working fine. Eating food now. BBS with swap out results.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 SWAP UPDATE I have tried 5.00 with both of my boards and can confirm neither of them detect the SATA drive. So its definitely Banana Pro boards that don't work. 3 counts so far.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 Works on cubietruck too. There are some commits that may be responsible for this (especially since /proc/interrupts lacks SATA entry), but this needs a) confirming that it is 100% kernel related (installing only linux-image and linux-dtb on top of 4.5 image) and then using git bisect to find commit that broke it. Can I help with that?
zador.blood.stained Posted February 14, 2016 Posted February 14, 2016 Can I help with that? Yes, with the first part. Grab Armbian 4.5 image from mirror.igorpecovnik.com and manually install 5.00 linux-image and linux-dtb packages (don't update u-boot, board supoort package or anything else) and check if this breaks SATA..
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 Yes, with the first part. Grab Armbian 4.5 image from mirror.igorpecovnik.com and manually install 5.00 linux-image and linux-dtb packages (don't update u-boot, board supoort package or anything else) and check if this breaks SATA.. Okay, am just tidying some other things up. Will get onto that later. I will be away until wednesday so will get what i can done tonight.
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 root@bananapi:~# apt-get install linux-image-next-sunxi Reading package lists... Done Building dependency tree Reading state information... Done The following packages will be upgraded: linux-image-next-sunxi 1 upgraded, 0 newly installed, 0 to remove and 76 not upgraded. Need to get 13.0 MB of archives. After this operation, 11.6 MB of additional disk space will be used. Get:1 http://apt.armbian.com/ jessie/main linux-image-next-sunxi armhf 5.00 [13. 0 MB] Fetched 13.0 MB in 15s (831 kB/s) (Reading database ... 49566 files and directories currently installed.) Preparing to unpack .../linux-image-next-sunxi_5.00_armhf.deb ... Unpacking linux-image-next-sunxi (5.00) over (4.5) ... Setting up linux-image-next-sunxi (5.00) ... root@bananapi:~# apt-get install linux-dtb-next-sunxi Reading package lists... Done Building dependency tree Reading state information... Done The following packages will be upgraded: linux-dtb-next-sunxi 1 upgraded, 0 newly installed, 0 to remove and 75 not upgraded. Need to get 63.3 kB of archives. After this operation, 494 kB of additional disk space will be used. Get:1 http://apt.armbian.com/ jessie/main linux-dtb-next-sunxi armhf 5.00 [63.3 kB] Fetched 63.3 kB in 0s (107 kB/s) (Reading database ... 50291 files and directories currently installed.) Preparing to unpack .../linux-dtb-next-sunxi_5.00_armhf.deb ... Unpacking linux-dtb-next-sunxi (5.00) over (4.5) ... Setting up linux-dtb-next-sunxi (5.00) ... root@bananapi:~# blkid /dev/mmcblk0: PTUUID="000380a6" PTTYPE="dos" /dev/mmcblk0p1: UUID="b5d952cd-46ba-4d79-b669-ba961ec9559f" TYPE="ext4" PARTUUID="000380a6-01" /dev/sda1: UUID="38f93141-1d36-434f-9f53-ca0d6916fb0e" TYPE="ext4" PARTUUID="1bd8dd3e-01" root@bananapi:~# reboot Broadcast message from root@bananapi (pts/0) (Sun Feb 14 20:30:48 2016): The system is going down for reboot NOW! root@bananapi:~# login as: root root@192.168.0.8's password: ____ | __ ) __ _ _ __ __ _ _ __ __ _ | _ \ / _` | '_ \ / _` | '_ \ / _` | | |_) | (_| | | | | (_| | | | | (_| | |____/ \__,_|_| |_|\__,_|_| |_|\__,_| Last login: Sun Feb 14 20:26:07 2016 from 192.168.0.100 Load: 0.57, 0.20, 0.07 - Board: 38.9°C - Memory: 966Mb root@bananapi:~# blkid /dev/mmcblk0: PTUUID="000380a6" PTTYPE="dos" /dev/mmcblk0p1: UUID="b5d952cd-46ba-4d79-b669-ba961ec9559f" TYPE="ext4" PARTUUID="000380a6-01" root@bananapi:~# That broke SATA!!
zador.blood.stained Posted February 14, 2016 Posted February 14, 2016 Ok, and what if you install kernel 4.3.3? http://apt.armbian.com/pool/main/l/linux-4.3.3-sunxi/
Gary Blake ヅ Posted February 14, 2016 Author Posted February 14, 2016 Ok, and what if you install kernel 4.3.3? http://apt.armbian.com/pool/main/l/linux-4.3.3-sunxi/ Thats a working SATA drive after update. What does that mean?
zador.blood.stained Posted February 14, 2016 Posted February 14, 2016 This means that we have 2 reference points: "good" 4.3.3 and "bad" 4.4.1. I'll prepare environment and make proper scripts for further testing when I have some free time and will post new build here, if there won't be new info by that time.
nihilista Posted February 15, 2016 Posted February 15, 2016 Got the same problem here, first i thought my fault was to install 4.4.1 kernel first and got a non-booting system (system on sata ssd). I tried copying kernel files back (4.1.0rc6 as i used before) but couldn't get system to boot. Btw, after some fiddling at boot partition sytem now boots again with 4.1.0rc6 kernel. I tested armbian 5.0 image on a fresh sd card, system booted but sata ssd not present. If someone builds a new kernel image i can safely test that safely on my fresh installed second sd card and report here.
Igor Posted February 15, 2016 Posted February 15, 2016 Thank you. We need to do this systematically so it will take some time to prepare. Currently we know that 4.3.3 and older works while 4.4.1 breaks ... and only on Banana PRO.
Alexio Posted February 15, 2016 Posted February 15, 2016 I think I have the same issue on a Banana pi (non pro). Last weekend I started building my 'Banana-Pi-openmediavault' server from scratch with V5. In openmediavault I'm unable to mount the attached sata drive. When mounting OMV returns the following error: Failed to mount '*ID*': mount: wrong fs type, bad option, bad superblock on /dev/sda1, missing codepage or helper program, or other error In some cases useful info is found in syslog - try dmesg | tail or so I've tried formatting the HDD to ext3 and ext4, both to no avail.
Gary Blake ヅ Posted February 15, 2016 Author Posted February 15, 2016 That's a different problem alexia. You would not get the chance to mount with this problem. 1
nihilista Posted February 15, 2016 Posted February 15, 2016 I think I have the same issue on a Banana pi (non pro). Last weekend I started building my 'Banana-Pi-openmediavault' server from scratch with V5. In openmediavault I'm unable to mount the attached sata drive. When mounting OMV returns the following error: Failed to mount '*ID*': mount: wrong fs type, bad option, bad superblock on /dev/sda1, missing codepage or helper program, or other error In some cases useful info is found in syslog - try dmesg | tail or so I've tried formatting the HDD to ext3 and ext4, both to no avail. Try lsblk, if your drive is present its definitively other problem...
zador.blood.stained Posted February 15, 2016 Posted February 15, 2016 OK, so (anyone) please test kernels from subdirectories v4.3.3 and v4.4.1 to confirm that we didn't miss anything, preferrably on clean Armbian 4.5 or 5.00 images. [Removed, not needed anymore]
(Phone).thumb.jpg.f60309bb32d95d32f028a6642a35c6ea.jpg)
Recommended Posts