Re: [lxc-users] kernel crash when starting an unprivileged container
Hi, On Wed, 3 Jun 2015 13:30:32 +0700 in message Re: [lxc-users] kernel crash when starting an unprivileged container Fajar A. Nugraha-san wrote: On Wed, Jun 3, 2015 at 1:26 PM, Tomasz Chmielewski man...@wpkg.org wrote: On 2015-06-03 15:01, Tomasz Chmielewski wrote: I'm trying to start an unprivileged container on Ubuntu 14.04; unfortunately, the kernel crashes. - kernel used is 4.0.4-040004-generic from http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.0.4-wily/ Crash of unprivileged container on 4.0.x is a known issue, there are posts about it on this list. This issue have been solved on 4.0.5 kernel. I've confirmed it now. Unpriv container is worked fine. -- KATOH Yasufumi ___ lxc-users mailing list lxc-users@lists.linuxcontainers.org http://lists.linuxcontainers.org/listinfo/lxc-users
Re: [lxc-users] kernel crash when starting an unprivileged container
As a side note, you can use rsyslogs remotelogging to get the oops Am 3. Juni 2015 08:01:22 MESZ, schrieb Tomasz Chmielewski man...@wpkg.org: I'm trying to start an unprivileged container on Ubuntu 14.04; unfortunately, the kernel crashes. # lxc-create -t download -n test-container (...) Distribution: ubuntu Release: trusty Architecture: amd64 (...) # lxc-start -n test-container -F Kernel crashes at this point. It does not crash if I start the container as privileged. - kernel used is 4.0.4-040004-generic from http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.0.4-wily/ - lxc userspace: http://ppa.launchpad.net/ubuntu-lxc/stable/ubuntu # dpkg -l|grep lxc ii liblxc1 1.1.2-0ubuntu3~ubuntu14.04.1~ppa1amd64Linux Containers userspace tools (library) ii lxc 1.1.2-0ubuntu3~ubuntu14.04.1~ppa1amd64Linux Containers userspace tools ii lxc-templates 1.1.2-0ubuntu3~ubuntu14.04.1~ppa1amd64Linux Containers userspace tools (templates) ii lxcfs 0.7-0ubuntu4~ubuntu14.04.1~ppa1 amd64FUSE based filesystem for LXC ii python3-lxc 1.1.2-0ubuntu3~ubuntu14.04.1~ppa1amd64Linux Containers userspace tools (Python 3.x bindings) It's a bit hard to get the printout of the OOPS, as I'm only able to access the server remotely and it doesn't manage to write the OOPS to the log. Anyway, after a few crashes and while true; do dmesg -c ; done I was able to capture this: [ 237.706914] device vethPI4H7F entered promiscuous mode [ 237.707006] IPv6: ADDRCONF(NETDEV_UP): vethPI4H7F: link is not ready [ 237.797284] eth0: renamed from veth1OSOTS [ 237.824526] IPv6: ADDRCONF(NETDEV_CHANGE): vethPI4H7F: link becomes ready [ 237.824556] lxcbr0: port 1(vethPI4H7F) entered forwarding state [ 237.824562] lxcbr0: port 1(vethPI4H7F) entered forwarding state [ 237.928179] BUG: unable to handle kernel NULL pointer dereference at (null) [ 237.928262] IP: [8122f888] pin_remove+0x58/0xf0 [ 237.928318] PGD 0 [ 237.928364] Oops: 0002 [#1] SMP [ 237.928432] Modules linked in: xt_conntrack veth xt_CHECKSUM iptable_mangle ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack xt_tcpudp iptable_filter ip_tables x_tables bridge stp llc intel_rapl iosf_mbi x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm crct10dif_pclmul crc32_pclmul eeepc_wmi ghash_clmulni_intel aesni_intel asus_wmi sparse_keymap ie31200_edac aes_x86_64 edac_core lrw gf128mul glue_helper shpchp lpc_ich ablk_helper cryptd mac_hid 8250_fintek serio_raw tpm_infineon video wmi btrfs lp parport raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq e1000e raid1 ahci raid0 ptp libahci pps_core multipath linear [ 237.930151] CPU: 2 PID: 6568 Comm: lxc-start Not tainted 4.0.4-040004-generic #201505171336 [ 237.930188] Hardware name: System manufacturer System Product Name/P8B WS, BIOS 0904 10/24/2011 [ 237.930225] task: 880806970a00 ti: 8808090c8000 task.ti: 8808090c8000 [ 237.930259] RIP: 0010:[8122f888] [8122f888] pin_remove+0x58/0xf0 [ 237.930341] RSP: 0018:8808090cbe18 EFLAGS: 00010246 [ 237.930383] RAX: RBX: 880808808a20 RCX: dead00100100 [ 237.930429] RDX: RSI: dead00200200 RDI: 81f9a548 [ 237.930474] RBP: 8808090cbe28 R08: 81d11b60 R09: 0100 [ 237.930572] R13: 880806970a00 R14: 81ecd070 R15: 7ffe57fd5540 [ 237.930618] FS: 7fd448c0() GS:88082fa8() knlGS: [ 237.930685] CS: 0010 DS: ES: CR0: 80050033 [ 237.930728] CR2: CR3: 0008099c1000 CR4: 000407e0 [ 237.930773] Stack: [ 237.930809] 880806970a00 880808808a20 8808090cbe48 8121d0f2 [ 237.930957] 8808090cbe68 880808808a20 8808090cbea8 8122fa55 [ 237.931123] 880806970a00 810bb2b0 8808090cbe70 [ 237.931286] Call Trace: [ 237.931336] [8121d0f2] drop_mountpoint+0x22/0x40 [ 237.931380] [8122fa55] pin_kill+0x75/0x130 [ 237.931425] [810bb2b0] ? prepare_to_wait_event+0x100/0x100 [ 237.931471] [8122fb39] mnt_pin_kill+0x29/0x40 [ 237.931530] [8121baf0] cleanup_mnt+0x80/0x90 [ 237.931573] [8121bb52] __cleanup_mnt+0x12/0x20 [ 237.931617] [81096ad7] task_work_run+0xb7/0xf0 [ 237.931662] [8101607c] do_notify_resume+0xbc/0xd0 [ 237.931709] [817f0beb] int_signal+0x12/0x17 -- Tomasz Chmielewski http://wpkg.org ___ lxc-users mailing list lxc-users@lists.linuxcontainers.org http://lists.linuxcontainers.org/listinfo/lxc-users
Re: [lxc-users] kernel crash when starting an unprivileged container
It may be worth trying, but it won't work reliably for most kernel crashes (network, disk IO etc. may crash as well). -- Tomasz Chmielewski http://wpkg.org On 2015-06-10 14:11, Christoph Lehmann wrote: As a side note, you can use rsyslogs remotelogging to get the oops Am 3. Juni 2015 08:01:22 MESZ, schrieb Tomasz Chmielewski man...@wpkg.org: I'm trying to start an unprivileged container on Ubuntu 14.04; unfortunately, the kernel crashes. # lxc-create -t download -n test-container (...) Distribution: ubuntu Release: trusty Architecture: amd64 (...) # lxc-start -n test-container -F Kernel crashes at this point. It does not crash if I start the container as privileged. - kernel used is 4.0.4-040004-generic from http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.0.4-wily [1]/ - lxc userspace: http://ppa.launchpad.net/ubuntu-lxc/stable/ubuntu [2] # dpkg -l|grep lxc ii liblxc1 1.1.2-0ubuntu3~ubuntu14.04.1~ppa1 amd64 Linux Containers userspace tools (library) ii lxc ! 1.1.2-0ubuntu3~ubuntu14.04.1~ppa1 amd64 Linux Containers userspace tools ii lxc-templates 1.1.2-0ubuntu3~ubuntu14.04.1~ppa1 amd64 Linux Containers userspace tools (templates) ii lxcfs 0.7-0ubuntu4~ubuntu14.04.1~ppa1 amd64 FUSE based filesystem for LXC ii python3-lxc 1.1.2-0ubuntu3~ubuntu14.04.1~ppa1 amd64 Linux Containers userspace tools (Python 3.x bindings) It's a bit hard to get the printout of the OOPS, as I'm only able to access the server remotely and it doesn't manage to write the OOPS to the log. Anyway, after a few crashes and while true; do dmesg -c ; done I was able to capture this: [ 237.706914] device vethPI4H7F entered promiscuous mode [ 237.707006] IPv6: ADDRCONF(NETDEV_UP): vethPI4H7F:! link is not ready [ 237.797284] eth0: renamed from veth1OSOTS [ 237.824526] IPv6: ADDRCONF(NETDEV_CHANGE): vethPI4H7F: link becomes ready [ 237.824556] lxcbr0: port 1(vethPI4H7F) entered forwarding state [ 237.824562] lxcbr0: port 1(vethPI4H7F) entered forwarding state [ 237.928179] BUG: unable to handle kernel NULL pointer dereference at (null) [ 237.928262] IP: [8122f888] pin_remove+0x58/0xf0 [ 237.928318] PGD 0 [ 237.928364] Oops: 0002 [#1] SMP [ 237.928432] Modules linked in: xt_conntrack veth xt_CHECKSUM iptable_mangle ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack xt_tcpudp iptable_filter ip_tables x_tables bridge stp llc intel_rapl iosf_mbi x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm crct10dif_pclmul crc32_pclmul eeepc_wmi ghash_clmulni_intel aesni_intel asus_wmi sparse_keymap ie31200_edac aes_x86_64 edac_core lrw gf128mul glue_helper shpchp lpc_ich ablk_helper cryptd mac_hid 8250_fintek serio_raw tpm_infineon video wmi btrfs lp parport raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq e1000e raid1 ahci raid0 ptp libahci pps_core multipath linear [ 237.930151] CPU: 2 PID: 6568 Comm: lxc-start Not tainted 4.0.4-040004-generic #201505171336 [ 237.930188] Hardware name: System manufacturer System Product Name/P8B WS, BIOS 0904 10/24/2011 [ 237.930225] task: 880806970a00 ti: 8808090c8000 task.ti: 8808090c8000 [ 237.930259] RIP: 0010:[8122f888] [8122f888] pin_remove+0x58/0xf0 [ 237.930341] RSP: 0018:8808090cbe18 EFLAGS: 00010246 [ 237.930383] RAX: RBX: 880808808a20 RCX: dead00100100 [ 237.930429] RDX: RSI: dead002! 00200 RDI: 81f9a548 [ 237.930474] RBP: 8808090cbe28 R08: 81d11b60 R09: 0100 [ 237.930572] R13: 880806970a00 R14: 81ecd070 R15: 7ffe57fd5540 [ 237.930618] FS: 7fd448c0() GS:88082fa8() knlGS: [ 237.930685] CS: 0010 DS: ES: CR0: 80050033 [ 237.930728] CR2: CR3: 0008099c1000 CR4: 000407e0 [ 237.930773] Stack: [ 237.930809] 880806970a00 880808808a20 8808090cbe48 8121d0f2 [ 237.930957] 8808090cbe68 880808808a20 8808090cbea8 8122fa55 [ 237.931123] 880806970a00 810bb2b0 8808090cbe70 [ 237.931286] Call Trace: [ 237.931336] [8121d0f2] drop_mountpoint+0x22/0x40 [ 237.931380] [8122fa55] pin_kill+0x75/0x130 [ 237.931425] [810bb2b0] ? prepare_to_wait_event+0x100/0x100 [ 237.931471] [8122fb39] mnt_pin_kill+0x29/0x40 [ 237.931530] [8121baf0] cleanup_mnt+0x80/0x90 [ 237.931573] [8121bb52] __cleanup_mnt+0x12/0x20 [ 237.931617] [81096ad7] task_work_run+0xb7/0xf0 [ 237.931662] [8101607c] do_notify_resume+0xbc/0xd0 [ 237.931709] [817f0beb] int_signal+0x12/0x17 -- Diese Nachricht wurde von meinem Android-Mobiltelefon mit K-9 Mail gesendet. ___ lxc-users mailing list lxc-users@lists.linuxcontainers.org http://lists.linuxcontainers.org/listinfo/lxc-users ___ lxc-users
Re: [lxc-users] kernel crash when starting an unprivileged container
On 2015-06-03 15:01, Tomasz Chmielewski wrote: I'm trying to start an unprivileged container on Ubuntu 14.04; unfortunately, the kernel crashes. # lxc-create -t download -n test-container (...) # lxc-start -n test-container -F Kernel crashes at this point. It does not crash if I start the container as privileged. - kernel used is 4.0.4-040004-generic from http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.0.4-wily/ The issue was a bit weird: - I've updated the kernel to 4.1-rc6, no longer crashing - still, the container was not starting on 4.1-rc6 - it turned out that lxc-create -t download ... created the container with all files being 0-bytes for some reason (so, 0-byte /sbin/init and all other files being 0-byte) - exec file format (0-byte /sbin/init) was causing 4.0.4 kernel crash? Anyway, problem solved. -- Tomasz Chmielewski http://wpkg.org ___ lxc-users mailing list lxc-users@lists.linuxcontainers.org http://lists.linuxcontainers.org/listinfo/lxc-users