Constant crashes - with logs

I’ve seen a few reports here about constant crashes on AppleTV - with Sam’s reply about a “useful” bug reports requiring logs. Unfortunately nobody seems to have taken the trouble - so I’m doing this now.

I am reliably having OSMC hang and then restart when playing higher bitrate/resolution files. The problem appears to be related to CrystalHD. I’m on an AppleTV1 with a CrystalHD (bcm70015 I believe), that has reliably worked using OpenElec (Used to run Crystalbuntu on this AppleTV too, but using a bcm70012 card). Generally the crash happens before the file even starts playing, however in the first instance that I had debug logging enabled, the file did play for a few seconds before it crashed. I’ve tried both the Confluence and OSMC skins but this doesn’t appear to make a difference.

I’ve uploaded my debug log here: http://pastebin.com/B1XwzPgJgruff

I’ve done a subsequent attempt which crashed almost immediately (without the video playback commencing) and for this one I ran the ‘grab-logs -A’ command with output here: http://paste.osmc.io/icudibazir

I’m testing this out from USB and have experienced the same symptoms using both of these builds:
2016.02-3-USB
2016.04-1-USB
(Specifically using a Sandisk 32gb cruzer micro, which again I have used reliably under OpenElec).

I can only access one of those logs. It shows CrystalHD decoder open but that’s it. It would be good to get a small sample of a file that’s problematic so I can look closely at the issue.

It should also be noted that we can swap quite aggressively on OSMC, which isn’t done quite as much on OE. This means OSMC will run a lot better on a hard disk than a USB, and should have a little better performance than OE would on an HDD. The idea behind USB installs is to use a rotational drive, or just test OSMC before you wipe the Apple OS. If you can, attempt a hard disk install.

Sam

Strange that the pastebin link died. I wonder if it was too big? (It did trigger some kind of SPAM warning when I posted it, but I completed a recaptcha which seemed to resolve this.)

I’ve tried to cut back the log so it is under 64kb, maybe this will help, though the SPAM warning was still triggered:

… and here’s a really ‘lightweight’ paste of just the last couple of hundred lines before the crash, that didn’t trigger this SPAM warning:

I’ll see if I can prepare some kind of sample file - might take me a day or two.

I do expect USB to be slow. I have an SSD in my AppleTV, but I’d prefer to know it’s going to work before I nuke my existing OpenElec install :slight_smile:

If you want to paste big logs, use paste.osmc.io

I’m also having frequent crashes.
It usually happensstopping video playback but also with video streaming.
I was running top command during these crashes and I noticed that swap free memory goes to 0, kodi becomes unresponsive and then osmc crashes.
Here is my log:
paste.osmc.io/acavudenoq

[ 3851.923276] Call Trace:
[ 3851.923298]  [<c179cc7f>] dump_stack+0x16/0x18
[ 3851.923310]  [<c10d156d>] dump_header.isra.12+0x3d/0x1b0
[ 3851.923322]  [<c179939a>] oom_kill_process.part.15+0x2e/0x297
[ 3851.923335]  [<c104fda7>] ? has_capability_noaudit+0x17/0x20
[ 3851.923345]  [<c10d1eed>] out_of_memory+0x39d/0x440
[ 3851.923355]  [<c10d754b>] __alloc_pages_nodemask+0x67b/0x6a0
[ 3851.923366]  [<c10d0b19>] filemap_fault+0x2b9/0x580
[ 3851.923375]  [<c10ce7ed>] ? filemap_map_pages+0x16d/0x280
[ 3851.923388]  [<c10f5b3f>] handle_mm_fault+0x93f/0xeb0
[ 3851.923401]  [<c103ac05>] __do_page_fault+0x155/0x410
[ 3851.923410]  [<c103aed0>] ? vmalloc_sync_all+0x10/0x10
[ 3851.923420]  [<c103aedb>] do_page_fault+0xb/0x10
[ 3851.923431]  [<c17a3e68>] error_code+0x58/0x60
[ 3851.923440]  [<c17a0000>] ? __schedule+0x130/0x790
[ 3851.923450]  [<c103aed0>] ? vmalloc_sync_all+0x10/0x10
[ 3851.923456] Mem-Info:
[ 3851.923472] active_anon:22721 inactive_anon:22729 isolated_anon:0
 active_file:170 inactive_file:238 isolated_file:0
 unevictable:0 dirty:0 writeback:2711 unstable:0
 slab_reclaimable:774 slab_unreclaimable:2038
 mapped:2436 shmem:146 pagetables:481 bounce:0
 free:4424 free_pcp:31 free_cma:0

We may want to use fewer ring buffer for the crystalhd decoder in future.

I also found this in dmesg, maybe related :
[ 3851.923198] systemd invoked oom-killer: gfp_mask=0x201da, order=0, oom_score_adj=0 [ 3851.923218] CPU: 0 PID: 1 Comm: systemd Tainted: P O 4.2.3-7-osmc #1 [ 3851.923225] Hardware name: Apple Inc. AppleTV1,1/Mac-F4228DC8, BIOS ATV11.88Z.00DA.B04.0912220919 12/22/09 [ 3851.923232] cf43b940 cf43b940 cf43dd80 c179cc7f cf43ddc4 c10d156d c18b83ac cf43bb94 [ 3851.923249] 000201da 00000000 00000000 00000004 00000002 00000007 00000006 00000005 [ 3851.923262] 00000004 00000003 00000000 000201da cf468680 cf43ddec c179939a 00000000 [ 3851.923276] Call Trace: [ 3851.923298] [<c179cc7f>] dump_stack+0x16/0x18 [ 3851.923310] [<c10d156d>] dump_header.isra.12+0x3d/0x1b0 [ 3851.923322] [<c179939a>] oom_kill_process.part.15+0x2e/0x297 [ 3851.923335] [<c104fda7>] ? has_capability_noaudit+0x17/0x20 [ 3851.923345] [<c10d1eed>] out_of_memory+0x39d/0x440 [ 3851.923355] [<c10d754b>] __alloc_pages_nodemask+0x67b/0x6a0 [ 3851.923366] [<c10d0b19>] filemap_fault+0x2b9/0x580 [ 3851.923375] [<c10ce7ed>] ? filemap_map_pages+0x16d/0x280 [ 3851.923388] [<c10f5b3f>] handle_mm_fault+0x93f/0xeb0 [ 3851.923401] [<c103ac05>] __do_page_fault+0x155/0x410 [ 3851.923410] [<c103aed0>] ? vmalloc_sync_all+0x10/0x10 [ 3851.923420] [<c103aedb>] do_page_fault+0xb/0x10 [ 3851.923431] [<c17a3e68>] error_code+0x58/0x60 [ 3851.923440] [<c17a0000>] ? __schedule+0x130/0x790 [ 3851.923450] [<c103aed0>] ? vmalloc_sync_all+0x10/0x10 [ 3851.923456] Mem-Info: [ 3851.923472] active_anon:22721 inactive_anon:22729 isolated_anon:0 active_file:170 inactive_file:238 isolated_file:0 unevictable:0 dirty:0 writeback:2711 unstable:0 slab_reclaimable:774 slab_unreclaimable:2038 mapped:2436 shmem:146 pagetables:481 bounce:0 free:4424 free_pcp:31 free_cma:0 [ 3851.923507] DMA free:1708kB min:796kB low:992kB high:1192kB active_anon:5280kB inactive_anon:5280kB active_file:0kB inactive_file:4kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15992kB managed:15912kB mlocked:0kB dirty:0kB writeback:0kB mapped:860kB shmem:92kB slab_reclaimable:144kB slab_unreclaimable:500kB kernel_stack:40kB pagetables:84kB unstable:0kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:28 all_unreclaimable? yes [ 3851.923515] lowmem_reserve[]: 0 223 223 223 [ 3851.923542] Normal free:15988kB min:11488kB low:14360kB high:17232kB active_anon:85604kB inactive_anon:85636kB active_file:680kB inactive_file:948kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:241876kB managed:229184kB mlocked:0kB dirty:0kB writeback:10844kB mapped:8884kB shmem:492kB slab_reclaimable:2952kB slab_unreclaimable:7652kB kernel_stack:936kB pagetables:1840kB unstable:0kB bounce:0kB free_pcp:124kB local_pcp:124kB free_cma:0kB writeback_tmp:0kB pages_scanned:9768 all_unreclaimable? yes [ 3851.923550] lowmem_reserve[]: 0 0 0 0 [ 3851.923560] DMA: 5*4kB (M) 1*8kB (U) 1*16kB (M) 14*32kB (UEM) 11*64kB (UEM) 4*128kB (EM) 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1708kB [ 3851.923593] Normal: 433*4kB (UEM) 228*8kB (UEM) 97*16kB (UEM) 34*32kB (UEM) 49*64kB (EM) 12*128kB (UEM) 8*256kB (EM) 4*512kB (UM) 1*1024kB (M) 0*2048kB 0*4096kB = 15988kB [ 3851.923634] Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB [ 3851.923639] 3833 total pagecache pages [ 3851.923645] 3279 pages in swap cache [ 3851.923652] Swap cache stats: add 111779, delete 108500, find 22640/28371 [ 3851.923657] Free swap = 0kB [ 3851.923661] Total swap = 131068kB [ 3851.923666] 64467 pages RAM [ 3851.923670] 0 pages HighMem/MovableOnly [ 3851.923674] 3193 pages reserved [ 3851.923679] [ pid ] uid tgid total_vm rss nr_ptes nr_pmds swapents oom_score_adj name [ 3851.923693] [ 749] 0 749 2010 112 7 3 46 0 systemd-journal [ 3851.923704] [ 1247] 0 1247 2615 1 5 3 124 -1000 systemd-udevd [ 3851.923714] [ 1377] 0 1377 792 15 4 3 36 0 cron [ 3851.923724] [ 1378] 0 1378 880 18 4 3 47 0 systemd-logind [ 3851.923735] [ 1379] 105 1379 861 31 4 3 41 0 avahi-daemon [ 3851.923745] [ 1380] 104 1380 1312 63 5 3 66 -900 dbus-daemon [ 3851.923755] [ 1401] 105 1401 829 4 4 3 51 0 avahi-daemon [ 3851.923765] [ 1420] 0 1420 2070 12 6 3 129 0 connmand [ 3851.923775] [ 1422] 1000 1422 1743 0 6 3 90 0 udisks-glue [ 3851.923785] [ 1424] 0 1424 550 0 4 3 35 0 eventlircd [ 3851.923796] [ 1441] 0 1441 6807 60 10 3 136 0 udisks-daemon [ 3851.923806] [ 1442] 0 1442 2120 4 7 3 81 0 udisks-daemon [ 3851.923816] [ 1444] 0 1444 1965 0 6 3 121 -1000 sshd [ 3851.923826] [ 1448] 0 1448 841 79 4 3 36 0 atvclient [ 3851.923837] [ 1451] 0 1451 876 0 4 3 56 0 mediacenter [ 3851.923847] [ 1455] 0 1455 8781 56 10 3 150 0 polkitd [ 3851.923857] [ 1490] 0 1490 1034 12 4 3 63 0 rpcbind [ 3851.923867] [ 1521] 0 1521 1881 26 6 3 85 0 wpa_supplicant [ 3851.923877] [ 1583] 1000 1583 7513 85 11 3 219 0 transmission-da [ 3851.923888] [ 1686] 0 1686 4877 30 11 3 288 0 nmbd [ 3851.923898] [ 1707] 106 1707 1400 20 5 3 91 0 ntpd [ 3851.923916] [ 1727] 0 1727 8279 36 18 3 408 0 smbd [ 3851.923926] [ 2124] 0 2124 2782 8 7 3 172 0 sshd [ 3851.923936] [ 2133] 1000 2133 1151 1 5 3 101 0 systemd [ 3851.923946] [ 2138] 1000 2138 6119 11 9 3 290 0 (sd-pam) [ 3851.923956] [ 2142] 1000 2142 2782 87 7 3 150 0 sshd [ 3851.923966] [ 2145] 1000 2145 907 0 4 3 92 0 bash [ 3851.923975] [ 2150] 1000 2150 814 54 4 3 22 0 top [ 3851.923985] [ 2377] 0 2377 1346 0 5 3 89 0 sudo [ 3851.923995] [ 2384] 1000 2384 873 0 4 3 41 0 xinit [ 3851.924006] [ 2385] 1000 2385 18628 189 37 3 7659 0 Xorg [ 3851.924016] [ 2392] 1000 2392 195374 43506 246 3 20771 0 kodi.bin [ 3851.924024] Out of memory: Kill process 2392 (kodi.bin) score 686 or sacrifice child [ 3851.924035] Killed process 2392 (kodi.bin) total-vm:781496kB, anon-rss:165288kB, file-rss:8736kB

I have the same problem with kernel panics on reboot and/or Sad Face hangs on the OSMC screen. I’m on my 6th re-install of OSMC.

The problem for me comes after trying to play a live stream. It’s almost like a hidden bomb that goes off if I play the wrong live stream (though one time it happened on a MSNBC stream I had played before). I’m sooo tired of it and I wish this bug was fixed!!!

It may be an add-on that you are using.

Do you get the issue on a fresh install of OSMC with local playback?

2 posts were split to a new topic: Post violates forum policy

I have this issue using official Iptv simple client with a list of public legit streams.
When I stop a stream it takes ages to go back to list screen and often it crashes.
Do you plan to fix the chd issue soon? My atv is almost unusable now.

Thank you

If you want to be off any help or you want the potential issue being fixed I suggest you provide debug enable logs.
Please see How to submit a useful support request - General - OSMC for advice on how to help us.

Already did (5th post).
It is the same oom issue.

Just getting back to Sam with a sample video file that causes the problem.

I was able to re-create the problem with the 1080p “I am Legend” trailer, that can be downloaded from here:
i_am_legend-1080p_trailer.zip - Download I Am Legend - High Definition (1080p) Theatrical Trailer - dvdloc8.com (110mb)

Here is a copy of the debug log from my first crash.
http://paste.osmc.io/enijoxijix.coffee

On the screen, I see the circle of dots (frozen) for several seconds (actually it varies, and can feel like close to a minute), before eventually the OSMC splash screen re-appears and OSMC restarts.

I’ve tried this a few times with this sample video. Once or twice, I saw the same as the above but the video actually started playing after the long pause. The first time it played right through, smooth and good. The seocnd time the video froze after a few seconds.

I might add that my OSMC install is 100% vanilla. Default skin, no add-ons - the only change I’ve made to out-of-the-box is to add a Video source (I’m playing this file over UPNP).

Do you have a BCM70012 or BCM70015?

I have a BCM70015

Same chd version here.

Hello crash-buddies!

I was hoping that the may update would solve the situation with constant crashes but it did not (not for me at least). Almost seems as it got worse… :frowning:

Is there a way to downgrade to a (ie.) February release from within the “My OSMC” or perhaps from command line? Initially this platform worked very well (just as the Crystalbuntu did, still run this on my other 2 ATV1:s with no problems what so ever… )

You can download the February version from the website

However nothing significant has changed in this period, so I’m not convinced this will resolve your issue.

To get a better understanding of the problem you are experiencing we need more information, including logs from you. Please see How to submit a useful support request - General - OSMC for advice on how to help us.

I will be preparing a series a of logs for you as I am also experiencing extreme problems with stability as well across multiple apple tv’s (Version 16.X is almost unusable for me)

Amber Skin

Scrolling through my movie or TV library too quickly will result in AppleTV crashing and restarting.

Playing h.264 content greater than 720p will usually fail on the first play, either causing the apple tv to reboot, or simply not play. Second (and sometimes third) attempt will usually play the file (this is not localized to any single file, I can reproduce it with any h.264 content). Once files are playing there are no problems, but it takes 2-3 attempts to get media to play.