Closed rye-stripe closed 7 months ago
@rye-stripe and I were able to track this down using a combination of tracing virtiofsd
via logging, eBPF, and the kernel debugger. One thing to note here is that the fuse_inode.nodeid
gets used by multiple mount namespaces. In the container, it's associated with the s_root
of the submount's mountpoint. Outside the container, in the initial mount namespace of the guest, it's just an automount dentry in the kataShared
virtiofs
mount.
After figuring out how to trigger this, we looked up our container's mnt root for this filesystem:
crash> struct super_block.s_root ffff88800960e800
s_root = 0xffff888008a78900,
crash> struct dentry.d_inode 0xffff888008a78900
d_inode = 0xffff888008ae2300,
crash> struct inode.i_count 0xffff888008ae2300
i_count = {
counter = 1
},
crash> struct fuse_inode.nodeid 0xffff888008ae2300
nodeid = 8,
but when we watched with eBPF, it was a fuse inode on a different sb->s_inodes
list getting killed by the shrinker:
forgetting nodeid 8: 158 bash sb 92df800
fuse_evict_inode+5
iput+290
dentry_unlink_inode+187
__dentry_kill+223
shrink_dentry_list+128
prune_dcache_sb+89
super_cache_scan+313
do_shrink_slab+327
shrink_slab+519
shrink_node+712
do_try_to_free_pages+212
try_to_free_pages+221
__alloc_pages_slowpath.constprop.0+905
__alloc_pages+482
alloc_pages_vma+149
__handle_mm_fault+1933
handle_mm_fault+197
do_user_addr_fault+444
exc_page_fault+112
asm_exc_page_fault+39
which turned out to be this mount in the guest:
ffff888009332800 ffff8880092df800 virtiofs kataShared /run/kata-containers/shared/containers
After a bit of digging, it turns out that the mounts in the container are set up via fuse_fill_super_submount()
in Linux. That code presently creates its own copy of the fuse_inode
with the borrowed nodeid
from the previous copy of the mount. It ends up with an essentially separate fuse_inode
structure that refers to the same nodeid
, but it never sends a lookup to the virtiofsd
server to increment that reference count on the server side.
The patch that we sent to linux attempts to solve the problem by ensuring that there's a separate LOOKUP sent to the server for each submount created, that way even if the dentry shrinker kills one of these inodes in a superblock in another mount namespace, it continues to be valid until all references have been forgotten.
After a bit of iterating, the patch series from @kmjohansen was accepted in Linux 6.7. The fixing commit is c4d361f. It has been backported to 6.6, 6.1, and 5.15 series.
@rye-stripe does kata 3.2 or 3.3 use a fixed kernel version?
Users encountering this problem need to upgrade or patch their guest kernel.
The fix is in Linux kernels >= 6.7, >= 6.6.8, >= 6.1.70, and >= 5.15.146.
Expanding on what @kmjohansen said above, none of the kernel versions included in the kata-static tarballs (v3.2.0 or v3.3.0alpha) have a fix for this bug. However, there currently is an open PR which upgrades the default kernel to 6.6.16, which does include a fix for this bug.
Description of problem
We recently ran into an issue where a host mount directory becomes inaccessible after any process in the container gets killed due to OOM. After the OOM any attempt to access files or list directory contents results in EBADF error code to
open
syscalls.Version information
This issue is reproducible with out of the box kata-containers version 3.1.3 with default configuration (as downloaded from the releases page). Version 3.1.0 is also affected, probably others too.
Show kata-collect-data.sh details
# Meta details Running `kata-collect-data.sh` version `3.1.3 (commit ee57732fe08504773b1b5474f2248834ae1fbd66)` at `2023-09-21.20:45:46.592441702+0200`. ---
Runtime
Runtime is `/opt/kata/bin/kata-runtime`. # `kata-env`
---
/opt/kata/bin/kata-runtime kata-env
```toml [Kernel] Path = "/opt/kata/share/kata-containers/vmlinux-5.19.2-100" Parameters = "systemd.unit=kata-containers.target systemd.mask=systemd-networkd.service systemd.mask=systemd-networkd.socket scsi_mod.scan=none" [Meta] Version = "1.0.26" [Image] Path = "/opt/kata/share/kata-containers/kata-ubuntu-latest.image" [Initrd] Path = "" [Hypervisor] MachineType = "q35" Version = "QEMU emulator version 7.2.0 (kata-static)\nCopyright (c) 2003-2022 Fabrice Bellard and the QEMU Project developers" Path = "/opt/kata/bin/qemu-system-x86_64" BlockDeviceDriver = "virtio-scsi" EntropySource = "/dev/urandom" SharedFS = "virtio-fs" VirtioFSDaemon = "/opt/kata/libexec/virtiofsd" SocketPath = "" Msize9p = 8192 MemorySlots = 10 PCIeRootPort = 0 HotplugVFIOOnRootBus = false Debug = false [Runtime] Path = "/opt/kata/bin/kata-runtime" GuestSeLinuxLabel = "" Debug = false Trace = false DisableGuestSeccomp = true DisableNewNetNs = false SandboxCgroupOnly = false [Runtime.Config] Path = "/opt/kata/share/defaults/kata-containers/configuration-qemu.toml" [Runtime.Version] OCI = "1.0.2-dev" [Runtime.Version.Version] Semver = "3.1.3" Commit = "ee57732fe08504773b1b5474f2248834ae1fbd66" Major = 3 Minor = 1 Patch = 3 [Host] Kernel = "6.2.0-32-generic" Architecture = "amd64" VMContainerCapable = true SupportVSocks = true [Host.Distro] Name = "Ubuntu" Version = "22.04" [Host.CPU] Vendor = "AuthenticAMD" Model = "AMD Ryzen 7 5800H with Radeon Graphics" CPUs = 16 [Host.Memory] Total = 14158736 Free = 6124704 Available = 10461560 [Agent] Debug = false Trace = false ```
Runtime config files
# Runtime config files ## Runtime default config files ``` /etc/kata-containers/configuration.toml /opt/kata/share/defaults/kata-containers/configuration.toml ``` ## Runtime config file contents Config file `/etc/kata-containers/configuration.toml` not found
Config file `/usr/share/defaults/kata-containers/configuration.toml` not found
---
cat "/opt/kata/share/defaults/kata-containers/configuration.toml"
```toml # Copyright (c) 2017-2019 Intel Corporation # Copyright (c) 2021 Adobe Inc. # # SPDX-License-Identifier: Apache-2.0 # # XXX: WARNING: this file is auto-generated. # XXX: # XXX: Source file: "config/configuration-qemu.toml.in" # XXX: Project: # XXX: Name: Kata Containers # XXX: Type: kata [hypervisor.qemu] path = "/opt/kata/bin/qemu-system-x86_64" kernel = "/opt/kata/share/kata-containers/vmlinux.container" image = "/opt/kata/share/kata-containers/kata-containers.img" # initrd = "/opt/kata/share/kata-containers/kata-containers-initrd.img" machine_type = "q35" # rootfs filesystem type: # - ext4 (default) # - xfs # - erofs rootfs_type="ext4" # Enable confidential guest support. # Toggling that setting may trigger different hardware features, ranging # from memory encryption to both memory and CPU-state encryption and integrity. # The Kata Containers runtime dynamically detects the available feature set and # aims at enabling the largest possible one, returning an error if none is # available, or none is supported by the hypervisor. # # Known limitations: # * Does not work by design: # - CPU Hotplug # - Memory Hotplug # - NVDIMM devices # # Default false # confidential_guest = true # Choose AMD SEV-SNP confidential guests # In case of using confidential guests on AMD hardware that supports both SEV # and SEV-SNP, the following enables SEV-SNP guests. SEV guests are default. # Default false # sev_snp_guest = true # Enable running QEMU VMM as a non-root user. # By default QEMU VMM run as root. When this is set to true, QEMU VMM process runs as # a non-root random user. See documentation for the limitations of this mode. # rootless = true # List of valid annotation names for the hypervisor # Each member of the list is a regular expression, which is the base name # of the annotation, e.g. "path" for io.katacontainers.config.hypervisor.path" enable_annotations = ["enable_iommu"] # List of valid annotations values for the hypervisor # Each member of the list is a path pattern as described by glob(3). # The default if not set is empty (all annotations rejected.) # Your distribution recommends: ["/opt/kata/bin/qemu-system-x86_64"] valid_hypervisor_paths = ["/opt/kata/bin/qemu-system-x86_64"] # Optional space-separated list of options to pass to the guest kernel. # For example, use `kernel_params = "vsyscall=emulate"` if you are having # trouble running pre-2.15 glibc. # # WARNING: - any parameter specified here will take priority over the default # parameter value of the same name used to start the virtual machine. # Do not set values here unless you understand the impact of doing so as you # may stop the virtual machine from booting. # To see the list of default parameters, enable hypervisor debug, create a # container and look for 'default-kernel-parameters' log entries. kernel_params = "" # Path to the firmware. # If you want that qemu uses the default firmware leave this option empty firmware = "" # Path to the firmware volume. # firmware TDVF or OVMF can be split into FIRMWARE_VARS.fd (UEFI variables # as configuration) and FIRMWARE_CODE.fd (UEFI program image). UEFI variables # can be customized per each user while UEFI code is kept same. firmware_volume = "" # Machine accelerators # comma-separated list of machine accelerators to pass to the hypervisor. # For example, `machine_accelerators = "nosmm,nosmbus,nosata,nopit,static-prt,nofw"` machine_accelerators="" # Qemu seccomp sandbox feature # comma-separated list of seccomp sandbox features to control the syscall access. # For example, `seccompsandbox= "on,obsolete=deny,spawn=deny,resourcecontrol=deny"` # Note: "elevateprivileges=deny" doesn't work with daemonize option, so it's removed from the seccomp sandbox # Another note: enabling this feature may reduce performance, you may enable # /proc/sys/net/core/bpf_jit_enable to reduce the impact. see https://man7.org/linux/man-pages/man8/bpfc.8.html #seccompsandbox="on,obsolete=deny,spawn=deny,resourcecontrol=deny" # CPU features # comma-separated list of cpu features to pass to the cpu # For example, `cpu_features = "pmu=off,vmx=off" cpu_features="pmu=off" # Default number of vCPUs per SB/VM: # unspecified or 0 --> will be set to 1 # < 0 --> will be set to the actual number of physical cores # > 0 <= number of physical cores --> will be set to the specified number # > number of physical cores --> will be set to the actual number of physical cores default_vcpus = 1 # Default maximum number of vCPUs per SB/VM: # unspecified or == 0 --> will be set to the actual number of physical cores or to the maximum number # of vCPUs supported by KVM if that number is exceeded # > 0 <= number of physical cores --> will be set to the specified number # > number of physical cores --> will be set to the actual number of physical cores or to the maximum number # of vCPUs supported by KVM if that number is exceeded # WARNING: Depending of the architecture, the maximum number of vCPUs supported by KVM is used when # the actual number of physical cores is greater than it. # WARNING: Be aware that this value impacts the virtual machine's memory footprint and CPU # the hotplug functionality. For example, `default_maxvcpus = 240` specifies that until 240 vCPUs # can be added to a SB/VM, but the memory footprint will be big. Another example, with # `default_maxvcpus = 8` the memory footprint will be small, but 8 will be the maximum number of # vCPUs supported by the SB/VM. In general, we recommend that you do not edit this variable, # unless you know what are you doing. # NOTICE: on arm platform with gicv2 interrupt controller, set it to 8. default_maxvcpus = 0 # Bridges can be used to hot plug devices. # Limitations: # * Currently only pci bridges are supported # * Until 30 devices per bridge can be hot plugged. # * Until 5 PCI bridges can be cold plugged per VM. # This limitation could be a bug in qemu or in the kernel # Default number of bridges per SB/VM: # unspecified or 0 --> will be set to 1 # > 1 <= 5 --> will be set to the specified number # > 5 --> will be set to 5 default_bridges = 1 # Default memory size in MiB for SB/VM. # If unspecified then it will be set 2048 MiB. default_memory = 2048 # # Default memory slots per SB/VM. # If unspecified then it will be set 10. # This is will determine the times that memory will be hotadded to sandbox/VM. #memory_slots = 10 # Default maximum memory in MiB per SB / VM # unspecified or == 0 --> will be set to the actual amount of physical RAM # > 0 <= amount of physical RAM --> will be set to the specified number # > amount of physical RAM --> will be set to the actual amount of physical RAM default_maxmemory = 0 # The size in MiB will be plused to max memory of hypervisor. # It is the memory address space for the NVDIMM devie. # If set block storage driver (block_device_driver) to "nvdimm", # should set memory_offset to the size of block device. # Default 0 #memory_offset = 0 # Specifies virtio-mem will be enabled or not. # Please note that this option should be used with the command # "echo 1 > /proc/sys/vm/overcommit_memory". # Default false #enable_virtio_mem = true # Disable block device from being used for a container's rootfs. # In case of a storage driver like devicemapper where a container's # root file system is backed by a block device, the block device is passed # directly to the hypervisor for performance reasons. # This flag prevents the block device from being passed to the hypervisor, # virtio-fs is used instead to pass the rootfs. disable_block_device_use = false # Shared file system type: # - virtio-fs (default) # - virtio-9p # - virtio-fs-nydus shared_fs = "virtio-fs" # Path to vhost-user-fs daemon. virtio_fs_daemon = "/opt/kata/libexec/virtiofsd" # List of valid annotations values for the virtiofs daemon # The default if not set is empty (all annotations rejected.) # Your distribution recommends: ["/opt/kata/libexec/virtiofsd"] valid_virtio_fs_daemon_paths = ["/opt/kata/libexec/virtiofsd"] # Default size of DAX cache in MiB virtio_fs_cache_size = 0 # Default size of virtqueues virtio_fs_queue_size = 1024 # Extra args for virtiofsd daemon # # Format example: # ["--arg1=xxx", "--arg2=yyy"] # Examples: # Set virtiofsd log level to debug : ["--log-level=debug"] # # see `virtiofsd -h` for possible options. virtio_fs_extra_args = ["--thread-pool-size=1", "--announce-submounts"] # Cache mode: # # - never # Metadata, data, and pathname lookup are not cached in guest. They are # always fetched from host and any changes are immediately pushed to host. # # - auto # Metadata and pathname lookup cache expires after a configured amount of # time (default is 1 second). Data is cached while the file is open (close # to open consistency). # # - always # Metadata, data, and pathname lookup are cached in guest and never expire. virtio_fs_cache = "auto" # Block storage driver to be used for the hypervisor in case the container # rootfs is backed by a block device. This is virtio-scsi, virtio-blk # or nvdimm. block_device_driver = "virtio-scsi" # aio is the I/O mechanism used by qemu # Options: # # - threads # Pthread based disk I/O. # # - native # Native Linux I/O. # # - io_uring # Linux io_uring API. This provides the fastest I/O operations on Linux, requires kernel>5.1 and # qemu >=5.0. block_device_aio = "io_uring" # Specifies cache-related options will be set to block devices or not. # Default false #block_device_cache_set = true # Specifies cache-related options for block devices. # Denotes whether use of O_DIRECT (bypass the host page cache) is enabled. # Default false #block_device_cache_direct = true # Specifies cache-related options for block devices. # Denotes whether flush requests for the device are ignored. # Default false #block_device_cache_noflush = true # Enable iothreads (data-plane) to be used. This causes IO to be # handled in a separate IO thread. This is currently only implemented # for SCSI. # enable_iothreads = false # Enable pre allocation of VM RAM, default false # Enabling this will result in lower container density # as all of the memory will be allocated and locked # This is useful when you want to reserve all the memory # upfront or in the cases where you want memory latencies # to be very predictable # Default false #enable_mem_prealloc = true # Enable huge pages for VM RAM, default false # Enabling this will result in the VM memory # being allocated using huge pages. # This is useful when you want to use vhost-user network # stacks within the container. This will automatically # result in memory pre allocation #enable_hugepages = true # Enable vhost-user storage device, default false # Enabling this will result in some Linux reserved block type # major range 240-254 being chosen to represent vhost-user devices. enable_vhost_user_store = false # The base directory specifically used for vhost-user devices. # Its sub-path "block" is used for block devices; "block/sockets" is # where we expect vhost-user sockets to live; "block/devices" is where # simulated block device nodes for vhost-user devices to live. vhost_user_store_path = "/var/run/kata-containers/vhost-user" # Enable vIOMMU, default false # Enabling this will result in the VM having a vIOMMU device # This will also add the following options to the kernel's # command line: intel_iommu=on,iommu=pt #enable_iommu = true # Enable IOMMU_PLATFORM, default false # Enabling this will result in the VM device having iommu_platform=on set #enable_iommu_platform = true # List of valid annotations values for the vhost user store path # The default if not set is empty (all annotations rejected.) # Your distribution recommends: ["/var/run/kata-containers/vhost-user"] valid_vhost_user_store_paths = ["/var/run/kata-containers/vhost-user"] # The timeout for reconnecting on non-server spdk sockets when the remote end goes away. # qemu will delay this many seconds and then attempt to reconnect. # Zero disables reconnecting, and the default is zero. vhost_user_reconnect_timeout_sec = 0 # Enable file based guest memory support. The default is an empty string which # will disable this feature. In the case of virtio-fs, this is enabled # automatically and '/dev/shm' is used as the backing folder. # This option will be ignored if VM templating is enabled. #file_mem_backend = "" # List of valid annotations values for the file_mem_backend annotation # The default if not set is empty (all annotations rejected.) # Your distribution recommends: [""] valid_file_mem_backends = [""] # -pflash can add image file to VM. The arguments of it should be in format # of ["/path/to/flash0.img", "/path/to/flash1.img"] pflashes = [] # This option changes the default hypervisor and kernel parameters # to enable debug output where available. And Debug also enable the hmp socket. # # Default false #enable_debug = true # Disable the customizations done in the runtime when it detects # that it is running on top a VMM. This will result in the runtime # behaving as it would when running on bare metal. # #disable_nesting_checks = true # This is the msize used for 9p shares. It is the number of bytes # used for 9p packet payload. #msize_9p = 8192 # If false and nvdimm is supported, use nvdimm device to plug guest image. # Otherwise virtio-block device is used. # # nvdimm is not supported when `confidential_guest = true`. # # Default is false #disable_image_nvdimm = true # VFIO devices are hotplugged on a bridge by default. # Enable hotplugging on root bus. This may be required for devices with # a large PCI bar, as this is a current limitation with hotplugging on # a bridge. # Default false #hotplug_vfio_on_root_bus = true # Before hot plugging a PCIe device, you need to add a pcie_root_port device. # Use this parameter when using some large PCI bar devices, such as Nvidia GPU # The value means the number of pcie_root_port # This value is valid when hotplug_vfio_on_root_bus is true and machine_type is "q35" # Default 0 #pcie_root_port = 2 # If vhost-net backend for virtio-net is not desired, set to true. Default is false, which trades off # security (vhost-net runs ring0) for network I/O performance. #disable_vhost_net = true # # Default entropy source. # The path to a host source of entropy (including a real hardware RNG) # /dev/urandom and /dev/random are two main options. # Be aware that /dev/random is a blocking source of entropy. If the host # runs out of entropy, the VMs boot time will increase leading to get startup # timeouts. # The source of entropy /dev/urandom is non-blocking and provides a # generally acceptable source of entropy. It should work well for pretty much # all practical purposes. #entropy_source= "/dev/urandom" # List of valid annotations values for entropy_source # The default if not set is empty (all annotations rejected.) # Your distribution recommends: ["/dev/urandom","/dev/random",""] valid_entropy_sources = ["/dev/urandom","/dev/random",""] # Path to OCI hook binaries in the *guest rootfs*. # This does not affect host-side hooks which must instead be added to # the OCI spec passed to the runtime. # # You can create a rootfs with hooks by customizing the osbuilder scripts: # https://github.com/kata-containers/kata-containers/tree/main/tools/osbuilder # # Hooks must be stored in a subdirectory of guest_hook_path according to their # hook type, i.e. "guest_hook_path/{prestart,poststart,poststop}". # The agent will scan these directories for executable files and add them, in # lexicographical order, to the lifecycle of the guest container. # Hooks are executed in the runtime namespace of the guest. See the official documentation: # https://github.com/opencontainers/runtime-spec/blob/v1.0.1/config.md#posix-platform-hooks # Warnings will be logged if any error is encountered while scanning for hooks, # but it will not abort container execution. #guest_hook_path = "/usr/share/oci/hooks" # # Use rx Rate Limiter to control network I/O inbound bandwidth(size in bits/sec for SB/VM). # In Qemu, we use classful qdiscs HTB(Hierarchy Token Bucket) to discipline traffic. # Default 0-sized value means unlimited rate. #rx_rate_limiter_max_rate = 0 # Use tx Rate Limiter to control network I/O outbound bandwidth(size in bits/sec for SB/VM). # In Qemu, we use classful qdiscs HTB(Hierarchy Token Bucket) and ifb(Intermediate Functional Block) # to discipline traffic. # Default 0-sized value means unlimited rate. #tx_rate_limiter_max_rate = 0 # Set where to save the guest memory dump file. # If set, when GUEST_PANICKED event occurred, # guest memeory will be dumped to host filesystem under guest_memory_dump_path, # This directory will be created automatically if it does not exist. # # The dumped file(also called vmcore) can be processed with crash or gdb. # # WARNING: # Dump guest’s memory can take very long depending on the amount of guest memory # and use much disk space. #guest_memory_dump_path="/var/crash/kata" # If enable paging. # Basically, if you want to use "gdb" rather than "crash", # or need the guest-virtual addresses in the ELF vmcore, # then you should enable paging. # # See: https://www.qemu.org/docs/master/qemu-qmp-ref.html#Dump-guest-memory for details #guest_memory_dump_paging=false # Enable swap in the guest. Default false. # When enable_guest_swap is enabled, insert a raw file to the guest as the swap device # if the swappiness of a container (set by annotation "io.katacontainers.container.resource.swappiness") # is bigger than 0. # The size of the swap device should be # swap_in_bytes (set by annotation "io.katacontainers.container.resource.swap_in_bytes") - memory_limit_in_bytes. # If swap_in_bytes is not set, the size should be memory_limit_in_bytes. # If swap_in_bytes and memory_limit_in_bytes is not set, the size should # be default_memory. #enable_guest_swap = true # use legacy serial for guest console if available and implemented for architecture. Default false #use_legacy_serial = true # disable applying SELinux on the VMM process (default false) disable_selinux=false # disable applying SELinux on the container process # If set to false, the type `container_t` is applied to the container process by default. # Note: To enable guest SELinux, the guest rootfs must be CentOS that is created and built # with `SELINUX=yes`. # (default: true) disable_guest_selinux=true [factory] # VM templating support. Once enabled, new VMs are created from template # using vm cloning. They will share the same initial kernel, initramfs and # agent memory by mapping it readonly. It helps speeding up new container # creation and saves a lot of memory if there are many kata containers running # on the same host. # # When disabled, new VMs are created from scratch. # # Note: Requires "initrd=" to be set ("image=" is not supported). # # Default false #enable_template = true # Specifies the path of template. # # Default "/run/vc/vm/template" #template_path = "/run/vc/vm/template" # The number of caches of VMCache: # unspecified or == 0 --> VMCache is disabled # > 0 --> will be set to the specified number # # VMCache is a function that creates VMs as caches before using it. # It helps speed up new container creation. # The function consists of a server and some clients communicating # through Unix socket. The protocol is gRPC in protocols/cache/cache.proto. # The VMCache server will create some VMs and cache them by factory cache. # It will convert the VM to gRPC format and transport it when gets # requestion from clients. # Factory grpccache is the VMCache client. It will request gRPC format # VM and convert it back to a VM. If VMCache function is enabled, # kata-runtime will request VM from factory grpccache when it creates # a new sandbox. # # Default 0 #vm_cache_number = 0 # Specify the address of the Unix socket that is used by VMCache. # # Default /var/run/kata-containers/cache.sock #vm_cache_endpoint = "/var/run/kata-containers/cache.sock" [agent.kata] # If enabled, make the agent display debug-level messages. # (default: disabled) #enable_debug = true # Enable agent tracing. # # If enabled, the agent will generate OpenTelemetry trace spans. # # Notes: # # - If the runtime also has tracing enabled, the agent spans will be # associated with the appropriate runtime parent span. # - If enabled, the runtime will wait for the container to shutdown, # increasing the container shutdown time slightly. # # (default: disabled) #enable_tracing = true # Comma separated list of kernel modules and their parameters. # These modules will be loaded in the guest kernel using modprobe(8). # The following example can be used to load two kernel modules with parameters # - kernel_modules=["e1000e InterruptThrottleRate=3000,3000,3000 EEE=1", "i915 enable_ppgtt=0"] # The first word is considered as the module name and the rest as its parameters. # Container will not be started when: # * A kernel module is specified and the modprobe command is not installed in the guest # or it fails loading the module. # * The module is not available in the guest or it doesn't met the guest kernel # requirements, like architecture and version. # kernel_modules=[] # Enable debug console. # If enabled, user can connect guest OS running inside hypervisor # through "kata-runtime exec" command
#debug_console_enabled = true
# Agent connection dialing timeout value in seconds
# (default: 30)
#dial_timeout = 30
[runtime]
# If enabled, the runtime will log additional debug messages to the
# system log
# (default: disabled)
#enable_debug = true
#
# Internetworking model
# Determines how the VM should be connected to the
# the container network interface
# Options:
#
# - macvtap
# Used when the Container network interface can be bridged using
# macvtap.
#
# - none
# Used when customize network. Only creates a tap device. No veth pair.
#
# - tcfilter
# Uses tc filter rules to redirect traffic from the network interface
# provided by plugin to a tap interface connected to the VM.
#
internetworking_model="tcfilter"
# disable guest seccomp
# Determines whether container seccomp profiles are passed to the virtual
# machine and applied by the kata agent. If set to true, seccomp is not applied
# within the guest
# (default: true)
disable_guest_seccomp=true
# vCPUs pinning settings
# if enabled, each vCPU thread will be scheduled to a fixed CPU
# qualified condition: num(vCPU threads) == num(CPUs in sandbox's CPUSet)
# enable_vcpus_pinning = false
# Apply a custom SELinux security policy to the container process inside the VM.
# This is used when you want to apply a type other than the default `container_t`,
# so general users should not uncomment and apply it.
# (format: "user:role:type")
# Note: You cannot specify MCS policy with the label because the sensitivity levels and
# categories are determined automatically by high-level container runtimes such as containerd.
#guest_selinux_label="system_u:system_r:container_t"
# If enabled, the runtime will create opentracing.io traces and spans.
# (See https://www.jaegertracing.io/docs/getting-started).
# (default: disabled)
#enable_tracing = true
# Set the full url to the Jaeger HTTP Thrift collector.
# The default if not set will be "http://localhost:14268/api/traces"
#jaeger_endpoint = ""
# Sets the username to be used if basic auth is required for Jaeger.
#jaeger_user = ""
# Sets the password to be used if basic auth is required for Jaeger.
#jaeger_password = ""
# If enabled, the runtime will not create a network namespace for shim and hypervisor processes.
# This option may have some potential impacts to your host. It should only be used when you know what you're doing.
# `disable_new_netns` conflicts with `internetworking_model=tcfilter` and `internetworking_model=macvtap`. It works only
# with `internetworking_model=none`. The tap device will be in the host network namespace and can connect to a bridge
# (like OVS) directly.
# (default: false)
#disable_new_netns = true
# if enabled, the runtime will add all the kata processes inside one dedicated cgroup.
# The container cgroups in the host are not created, just one single cgroup per sandbox.
# The runtime caller is free to restrict or collect cgroup stats of the overall Kata sandbox.
# The sandbox cgroup path is the parent cgroup of a container with the PodSandbox annotation.
# The sandbox cgroup is constrained if there is no container type annotation.
# See: https://pkg.go.dev/github.com/kata-containers/kata-containers/src/runtime/virtcontainers#ContainerType
sandbox_cgroup_only=false
# If enabled, the runtime will attempt to determine appropriate sandbox size (memory, CPU) before booting the virtual machine. In
# this case, the runtime will not dynamically update the amount of memory and CPU in the virtual machine. This is generally helpful
# when a hardware architecture or hypervisor solutions is utilized which does not support CPU and/or memory hotplug.
# Compatibility for determining appropriate sandbox (VM) size:
# - When running with pods, sandbox sizing information will only be available if using Kubernetes >= 1.23 and containerd >= 1.6. CRI-O
# does not yet support sandbox sizing annotations.
# - When running single containers using a tool like ctr, container sizing information will be available.
static_sandbox_resource_mgmt=false
# If specified, sandbox_bind_mounts identifieds host paths to be mounted (ro) into the sandboxes shared path.
# This is only valid if filesystem sharing is utilized. The provided path(s) will be bindmounted into the shared fs directory.
# If defaults are utilized, these mounts should be available in the guest at `/run/kata-containers/shared/containers/sandbox-mounts`
# These will not be exposed to the container workloads, and are only provided for potential guest services.
sandbox_bind_mounts=[]
# VFIO Mode
# Determines how VFIO devices should be be presented to the container.
# Options:
#
# - vfio
# Matches behaviour of OCI runtimes (e.g. runc) as much as
# possible. VFIO devices will appear in the container as VFIO
# character devices under /dev/vfio. The exact names may differ
# from the host (they need to match the VM's IOMMU group numbers
# rather than the host's)
#
# - guest-kernel
# This is a Kata-specific behaviour that's useful in certain cases.
# The VFIO device is managed by whatever driver in the VM kernel
# claims it. This means it will appear as one or more device nodes
# or network interfaces depending on the nature of the device.
# Using this mode requires specially built workloads that know how
# to locate the relevant device interfaces within the VM.
#
vfio_mode="guest-kernel"
# If enabled, the runtime will not create Kubernetes emptyDir mounts on the guest filesystem. Instead, emptyDir mounts will
# be created on the host and shared via virtio-fs. This is potentially slower, but allows sharing of files from host to guest.
disable_guest_empty_dir=false
# Enabled experimental feature list, format: ["a", "b"].
# Experimental features are features not stable enough for production,
# they may break compatibility, and are prepared for a big version bump.
# Supported experimental features:
# (default: [])
experimental=[]
# If enabled, user can run pprof tools with shim v2 process through kata-monitor.
# (default: false)
# enable_pprof = true
# WARNING: All the options in the following section have not been implemented yet.
# This section was added as a placeholder. DO NOT USE IT!
[image]
# Container image service.
#
# Offload the CRI image management service to the Kata agent.
# (default: false)
#service_offload = true
# Container image decryption keys provisioning.
# Applies only if service_offload is true.
# Keys can be provisioned locally (e.g. through a special command or
# a local file) or remotely (usually after the guest is remotely attested).
# The provision setting is a complete URL that lets the Kata agent decide
# which method to use in order to fetch the keys.
#
# Keys can be stored in a local file, in a measured and attested initrd:
#provision=data:///local/key/file
#
# Keys could be fetched through a special command or binary from the
# initrd (guest) image, e.g. a firmware call:
#provision=file:///path/to/bin/fetcher/in/guest
#
# Keys can be remotely provisioned. The Kata agent fetches them from e.g.
# a HTTPS URL:
#provision=https://my-key-broker.foo/tenant/
```
Containerd shim v2
Containerd shim v2 is `/opt/kata/bin/containerd-shim-kata-v2`.
---
containerd-shim-kata-v2 --version
``` Kata Containers containerd shim (Golang): id: "io.containerd.kata.v2", version: 3.1.3, commit: ee57732fe08504773b1b5474f2248834ae1fbd66 ```
KSM throttler
# KSM throttler ## version ## systemd service
Image details
# Image details ```yaml --- osbuilder: url: "https://github.com/kata-containers/kata-containers/tools/osbuilder" version: "unknown" rootfs-creation-time: "2023-07-04T06:39:58.063976384+0000Z" description: "osbuilder rootfs" file-format-version: "0.0.2" architecture: "x86_64" base-distro: name: "ubuntu" version: "focal" packages: default: - "chrony" - "dbus" - "init" - "iptables" - "libseccomp2" extra: agent: url: "https://github.com/kata-containers/kata-containers" name: "kata-agent" version: "3.1.3" agent-is-init-daemon: "no" ``` ---
Initrd details
# Initrd details No initrd ---
Logfiles
# Logfiles ## Runtime logs
## Throttler logs
## Kata Containerd Shim v2 logs
---
Runtime logs
No recent runtime problems found in system journal.
Throttler logs
No recent throttler problems found in system journal.
Kata Containerd Shim v2
Recent problems found in system journal: ``` time="2023-09-21T20:19:58.977705941+02:00" level=error msg="createContainer failed" error="rpc error: code = Internal desc = the file bash was not found" name=containerd-shim-v2 pid=1077035 sandbox=test-kata source=virtcontainers subsystem=kata_agent time="2023-09-21T20:19:58.977932867+02:00" level=error msg="container create failed" container=test-kata error="rpc error: code = Internal desc = the file bash was not found" name=containerd-shim-v2 pid=1077035 sandbox=test-kata source=virtcontainers subsystem=container time="2023-09-21T20:19:58.977975077+02:00" level=warning error="no such file or directory" name=containerd-shim-v2 pid=1077035 sandbox=test-kata share-dir=/run/kata-containers/shared/sandboxes/test-kata/mounts/test-kata/rootfs source=virtcontainers subsystem=mount time="2023-09-21T20:19:58.978002658+02:00" level=warning msg="Could not remove container share dir" error="no such file or directory" name=containerd-shim-v2 pid=1077035 sandbox=test-kata share-dir=/run/kata-containers/shared/sandboxes/test-kata/mounts/test-kata source=virtcontainers subsystem=fs_share time="2023-09-21T20:19:59.306534924+02:00" level=warning msg="failed to cleanup network" error="failed to get netns /var/run/netns/cnitest-ff638666-78ec-ecdb-2c03-595b7942348d: failed to Statfs \"/var/run/netns/cnitest-ff638666-78ec-ecdb-2c03-595b7942348d\": no such file or directory" id=/var/run/netns/cnitest-ff638666-78ec-ecdb-2c03-595b7942348d name=containerd-shim-v2 pid=1077035 sandbox=test-kata source=katautils time="2023-09-21T20:36:24.565215851+02:00" level=error msg="failed to cleanup the &{%!s(*v2.Manager=&{/sys/fs/cgroup /sys/fs/cgroup/vc}) /vc %!s(*specs.LinuxCPU=) [{%!s(bool=true) c %!s(*int64=0xc00009d838) %!s(*int64=0xc00009d840) rwm} {%!s(bool=true) c %!s(*int64=0xc00009d868) %!s(*int64=0xc00009d870) rwm} {%!s(bool=true) c %!s(*int64=0xc00009d898) %!s(*int64=0xc00009d8a0) rwm} {%!s(bool=true) c %!s(*int64=0xc00009d9e8) %!s(*int64=0xc00009d9f0) rwm} {%!s(bool=true) c %!s(*int64=0xc00009da48) %!s(*int64=0xc00009da50) rwm} {%!s(bool=true) c %!s(*int64=0xc00009daa8) %!s(*int64=0xc00009dab0) rwm} {%!s(bool=true) c %!s(*int64=0xc00009dad8) %!s(*int64=0xc00009dae0) rwm} {%!s(bool=true) c %!s(*int64=0xc00009db08) %!s(*int64=0xc00009db10) rwm} {%!s(bool=true) c %!s(*int64=0xc00009db38) %!s(*int64=0xc00009db40) rwm} {%!s(bool=true) c %!s(*int64=0xc00009db68) %!s(*int64=0xc00009db70) rwm} {%!s(bool=true) c %!s(*int64=0xc00009dbf8) %!s(*int64=0xc00009dc00) rwm} {%!s(bool=true) c %!s(*int64=0xc0000443e8) %!s(*int64=0xc000044480) m} {%!s(bool=true) b %!s(*int64=0xc0000443e8) %!s(*int64=0xc000044480) m} {%!s(bool=true) c %!s(*int64=0xc000044488) %!s(*int64=0xc000044480) rwm} {%!s(bool=true) c %!s(*int64=0xc000044490) %!s(*int64=0xc000044498) rwm}] {%!s(int32=0) %!s(uint32=0)}} resource controllers" error="cgroups: unable to remove path \"/sys/fs/cgroup/vc\": unlinkat /sys/fs/cgroup/vc/misc.events: operation not permitted" name=containerd-shim-v2 pid=1078787 sandbox=test-kata source=virtcontainers subsystem=sandbox
time="2023-09-21T20:37:29.065607103+02:00" level=error msg="Get \"http://shim/agent-url\": dial unix /run/kata/kata-test/shim-monitor.sock: connect: no such file or directory" arch=amd64 command=exec name=kata-runtime pid=1078981 source=runtime
time="2023-09-21T20:37:35.626022306+02:00" level=error msg="Get \"http://shim/agent-url\": dial unix /run/kata/test-kata/shim-monitor.sock: connect: no such file or directory" arch=amd64 command=exec name=kata-runtime pid=1079008 source=runtime
time="2023-09-21T20:38:07.821424511+02:00" level=error msg="rpc error: code = DeadlineExceeded desc = timed out connecting to vsock 2928791664:1026" arch=amd64 command=exec name=kata-runtime pid=1079031 source=runtime
time="2023-09-21T20:43:19.191045601+02:00" level=error msg="trace called before context set" name=Delete pid=13689 sandbox=test-kata source=containerd-kata-shim-v2 type=bug
time="2023-09-21T20:43:19.191307355+02:00" level=error msg="trace called before context set" name=Shutdown pid=13689 sandbox=test-kata source=containerd-kata-shim-v2 type=bug
time="2023-09-21T20:43:53.319368669+02:00" level=error msg="trace called before context set" name=Delete pid=14095 sandbox=test-kata source=containerd-kata-shim-v2 type=bug
time="2023-09-21T20:43:53.319681166+02:00" level=error msg="trace called before context set" name=Shutdown pid=14095 sandbox=test-kata source=containerd-kata-shim-v2 type=bug
time="2023-09-21T20:44:24.909325356+02:00" level=warning msg="Could not add /dev/vhost-net to the devices cgroup" name=containerd-shim-v2 pid=14182 sandbox=test-kata source=cgroups
time="2023-09-21T20:44:24.90938723+02:00" level=warning msg="Could not add /dev/vfio/vfio to the devices cgroup" name=containerd-shim-v2 pid=14182 sandbox=test-kata source=cgroups
```
Container manager details
# Container manager details
---
Docker
## Docker
docker version
``` Client: Docker Engine - Community Version: 20.10.17 API version: 1.41 Go version: go1.17.11 Git commit: 100c701 Built: Mon Jun 6 23:03:23 2022 OS/Arch: linux/amd64 Context: default Experimental: true Server: Docker Engine - Community Engine: Version: 20.10.17 API version: 1.41 (minimum version 1.12) Go version: go1.17.11 Git commit: a89b842 Built: Mon Jun 6 23:01:15 2022 OS/Arch: linux/amd64 Experimental: false containerd: Version: 1.6.6 GitCommit: 10c12954828e7c7c9b6e0ea9b0c02b01407d3ae1 runc: Version: 1.1.2 GitCommit: v1.1.2-0-ga916309 docker-init: Version: 0.19.0 GitCommit: de40ad0 ```
docker info
``` Client: Context: default Debug Mode: false Plugins: app: Docker App (Docker Inc., v0.9.1-beta3) buildx: Docker Buildx (Docker Inc., v0.8.2-docker) scan: Docker Scan (Docker Inc., v0.17.0) Server: Containers: 564 Running: 5 Paused: 0 Stopped: 559 Images: 979 Server Version: 20.10.17 Storage Driver: overlay2 Backing Filesystem: extfs Supports d_type: true Native Overlay Diff: true userxattr: false Logging Driver: json-file Cgroup Driver: systemd Cgroup Version: 2 Plugins: Volume: local Network: bridge host ipvlan macvlan null overlay Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog Swarm: inactive Runtimes: io.containerd.runc.v2 io.containerd.runtime.v1.linux runc Default Runtime: runc Init Binary: docker-init containerd version: 10c12954828e7c7c9b6e0ea9b0c02b01407d3ae1 runc version: v1.1.2-0-ga916309 init version: de40ad0 Security Options: apparmor seccomp Profile: default cgroupns Kernel Version: 6.2.0-32-generic Operating System: Ubuntu 22.04.3 LTS OSType: linux Architecture: x86_64 CPUs: 16 Total Memory: 13.5GiB Name: yoga ID: XCYU:FAYX:Y2SB:2XOQ:4OZD:I2ZW:DCQH:D5CU:2UVO:BDVF:J4YX:MN4E Docker Root Dir: /var/lib/docker Debug Mode: false Registry: https://index.docker.io/v1/ Labels: Experimental: false Insecure Registries: 127.0.0.0/8 Live Restore Enabled: false ```
systemctl show docker
``` Type=notify Restart=always NotifyAccess=main RestartUSec=2s TimeoutStartUSec=infinity TimeoutStopUSec=infinity TimeoutAbortUSec=infinity TimeoutStartFailureMode=terminate TimeoutStopFailureMode=terminate RuntimeMaxUSec=infinity WatchdogUSec=0 WatchdogTimestamp=n/a WatchdogTimestampMonotonic=0 RootDirectoryStartOnly=no RemainAfterExit=no GuessMainPID=yes MainPID=4809 ControlPID=0 FileDescriptorStoreMax=0 NFileDescriptorStore=0 StatusErrno=0 Result=success ReloadResult=success CleanResult=success UID=[not set] GID=[not set] NRestarts=0 OOMPolicy=continue ExecMainStartTimestamp=Thu 2023-09-21 20:40:56 CEST ExecMainStartTimestampMonotonic=96343693 ExecMainExitTimestamp=n/a ExecMainExitTimestampMonotonic=0 ExecMainPID=4809 ExecMainCode=0 ExecMainStatus=0 ExecStart={ path=/usr/bin/dockerd ; argv[]=/usr/bin/dockerd -H fd:// --containerd=/run/containerd/containerd.sock ; ignore_errors=no ; start_time=[Thu 2023-09-21 20:40:56 CEST] ; stop_time=[n/a] ; pid=4809 ; code=(null) ; status=0/0 } ExecStartEx={ path=/usr/bin/dockerd ; argv[]=/usr/bin/dockerd -H fd:// --containerd=/run/containerd/containerd.sock ; flags= ; start_time=[Thu 2023-09-21 20:40:56 CEST] ; stop_time=[n/a] ; pid=4809 ; code=(null) ; status=0/0 } ExecReload={ path=/bin/kill ; argv[]=/bin/kill -s HUP $MAINPID ; ignore_errors=no ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; code=(null) ; status=0/0 } ExecReloadEx={ path=/bin/kill ; argv[]=/bin/kill -s HUP $MAINPID ; flags= ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; code=(null) ; status=0/0 } Slice=system.slice ControlGroup=/system.slice/docker.service MemoryCurrent=362627072 MemoryAvailable=infinity CPUUsageNSec=4360292000 EffectiveCPUs=0-15 EffectiveMemoryNodes=0 TasksCurrent=108 IPIngressBytes=[no data] IPIngressPackets=[no data] IPEgressBytes=[no data] IPEgressPackets=[no data] IOReadBytes=18446744073709551615 IOReadOperations=18446744073709551615 IOWriteBytes=18446744073709551615 IOWriteOperations=18446744073709551615 Delegate=yes DelegateControllers=cpu cpuacct cpuset io blkio memory devices pids bpf-firewall bpf-devices bpf-foreign bpf-socket-bind CPUAccounting=yes CPUWeight=[not set] StartupCPUWeight=[not set] CPUShares=[not set] StartupCPUShares=[not set] CPUQuotaPerSecUSec=infinity CPUQuotaPeriodUSec=infinity IOAccounting=no IOWeight=[not set] StartupIOWeight=[not set] BlockIOAccounting=no BlockIOWeight=[not set] StartupBlockIOWeight=[not set] MemoryAccounting=yes DefaultMemoryLow=0 DefaultMemoryMin=0 MemoryMin=0 MemoryLow=0 MemoryHigh=infinity MemoryMax=infinity MemorySwapMax=infinity MemoryLimit=infinity DevicePolicy=auto TasksAccounting=yes TasksMax=infinity IPAccounting=no ManagedOOMSwap=auto ManagedOOMMemoryPressure=auto ManagedOOMMemoryPressureLimit=0 ManagedOOMPreference=none Environment=USE_VOLATILE=1 UMask=0022 LimitCPU=infinity LimitCPUSoft=infinity LimitFSIZE=infinity LimitFSIZESoft=infinity LimitDATA=infinity LimitDATASoft=infinity LimitSTACK=infinity LimitSTACKSoft=8388608 LimitCORE=infinity LimitCORESoft=infinity LimitRSS=infinity LimitRSSSoft=infinity LimitNOFILE=infinity LimitNOFILESoft=infinity LimitAS=infinity LimitASSoft=infinity LimitNPROC=infinity LimitNPROCSoft=infinity LimitMEMLOCK=8388608 LimitMEMLOCKSoft=8388608 LimitLOCKS=infinity LimitLOCKSSoft=infinity LimitSIGPENDING=54796 LimitSIGPENDINGSoft=54796 LimitMSGQUEUE=819200 LimitMSGQUEUESoft=819200 LimitNICE=0 LimitNICESoft=0 LimitRTPRIO=0 LimitRTPRIOSoft=0 LimitRTTIME=infinity LimitRTTIMESoft=infinity OOMScoreAdjust=-500 CoredumpFilter=0x33 Nice=0 IOSchedulingClass=2 IOSchedulingPriority=4 CPUSchedulingPolicy=0 CPUSchedulingPriority=0 CPUAffinityFromNUMA=no NUMAPolicy=n/a TimerSlackNSec=50000 CPUSchedulingResetOnFork=no NonBlocking=no StandardInput=null StandardOutput=journal StandardError=inherit TTYReset=no TTYVHangup=no TTYVTDisallocate=no SyslogPriority=30 SyslogLevelPrefix=yes SyslogLevel=6 SyslogFacility=3 LogLevelMax=-1 LogRateLimitIntervalUSec=0 LogRateLimitBurst=0 SecureBits=0 CapabilityBoundingSet=cap_chown cap_dac_override cap_dac_read_search cap_fowner cap_fsetid cap_kill cap_setgid cap_setuid cap_setpcap cap_linux_immutable cap_net_bind_service cap_net_broadcast cap_net_admin cap_net_raw cap_ipc_lock cap_ipc_owner cap_sys_module cap_sys_rawio cap_sys_chroot cap_sys_ptrace cap_sys_pacct cap_sys_admin cap_sys_boot cap_sys_nice cap_sys_resource cap_sys_time cap_sys_tty_config cap_mknod cap_lease cap_audit_write cap_audit_control cap_setfcap cap_mac_override cap_mac_admin cap_syslog cap_wake_alarm cap_block_suspend cap_audit_read cap_perfmon cap_bpf cap_checkpoint_restore DynamicUser=no RemoveIPC=no PrivateTmp=no PrivateDevices=no ProtectClock=no ProtectKernelTunables=no ProtectKernelModules=no ProtectKernelLogs=no ProtectControlGroups=no PrivateNetwork=no PrivateUsers=no PrivateMounts=no PrivateIPC=no ProtectHome=no ProtectSystem=no SameProcessGroup=no UtmpMode=init IgnoreSIGPIPE=yes NoNewPrivileges=no SystemCallErrorNumber=2147483646 LockPersonality=no RuntimeDirectoryPreserve=no RuntimeDirectoryMode=0755 StateDirectoryMode=0755 CacheDirectoryMode=0755 LogsDirectoryMode=0755 ConfigurationDirectoryMode=0755 TimeoutCleanUSec=infinity MemoryDenyWriteExecute=no RestrictRealtime=no RestrictSUIDSGID=no RestrictNamespaces=no MountAPIVFS=no KeyringMode=private ProtectProc=default ProcSubset=all ProtectHostname=no KillMode=process KillSignal=15 RestartKillSignal=15 FinalKillSignal=9 SendSIGKILL=yes SendSIGHUP=no WatchdogSignal=6 Id=docker.service Names=docker.service Requires=docker.socket sysinit.target containerd.service system.slice Wants=network-online.target WantedBy=multi-user.target Conflicts=shutdown.target Before=multi-user.target shutdown.target After=sysinit.target containerd.service firewalld.service system.slice systemd-journald.socket network-online.target docker.socket basic.target TriggeredBy=docker.socket Documentation=https://docs.docker.com Description=Docker Application Container Engine LoadState=loaded ActiveState=active FreezerState=running SubState=running FragmentPath=/lib/systemd/system/docker.service UnitFileState=enabled UnitFilePreset=enabled StateChangeTimestamp=Thu 2023-09-21 20:41:03 CEST StateChangeTimestampMonotonic=103340536 InactiveExitTimestamp=Thu 2023-09-21 20:40:56 CEST InactiveExitTimestampMonotonic=96343905 ActiveEnterTimestamp=Thu 2023-09-21 20:41:03 CEST ActiveEnterTimestampMonotonic=103340536 ActiveExitTimestamp=n/a ActiveExitTimestampMonotonic=0 InactiveEnterTimestamp=n/a InactiveEnterTimestampMonotonic=0 CanStart=yes CanStop=yes CanReload=yes CanIsolate=no CanFreeze=yes StopWhenUnneeded=no RefuseManualStart=no RefuseManualStop=no AllowIsolate=no DefaultDependencies=yes OnSuccessJobMode=fail OnFailureJobMode=replace IgnoreOnIsolate=no NeedDaemonReload=no JobTimeoutUSec=infinity JobRunningTimeoutUSec=infinity JobTimeoutAction=none ConditionResult=yes AssertResult=yes ConditionTimestamp=Thu 2023-09-21 20:40:56 CEST ConditionTimestampMonotonic=96342203 AssertTimestamp=Thu 2023-09-21 20:40:56 CEST AssertTimestampMonotonic=96342212 Transient=no Perpetual=no StartLimitIntervalUSec=1min StartLimitBurst=3 StartLimitAction=none FailureAction=none SuccessAction=none InvocationID=18642f65b2e84eeeacd1151fa1609182 CollectMode=inactive ```
containerd
## containerd
containerd --version
``` containerd containerd.io 1.6.6 10c12954828e7c7c9b6e0ea9b0c02b01407d3ae1 ```
systemctl show containerd
``` Type=notify Restart=always NotifyAccess=main RestartUSec=5s TimeoutStartUSec=1min 30s TimeoutStopUSec=1min 30s TimeoutAbortUSec=1min 30s TimeoutStartFailureMode=terminate TimeoutStopFailureMode=terminate RuntimeMaxUSec=infinity WatchdogUSec=0 WatchdogTimestamp=n/a WatchdogTimestampMonotonic=0 RootDirectoryStartOnly=no RemainAfterExit=no GuessMainPID=yes MainPID=4474 ControlPID=0 FileDescriptorStoreMax=0 NFileDescriptorStore=0 StatusErrno=0 Result=success ReloadResult=success CleanResult=success UID=[not set] GID=[not set] NRestarts=0 OOMPolicy=continue ExecMainStartTimestamp=Thu 2023-09-21 20:40:56 CEST ExecMainStartTimestampMonotonic=96172434 ExecMainExitTimestamp=n/a ExecMainExitTimestampMonotonic=0 ExecMainPID=4474 ExecMainCode=0 ExecMainStatus=0 ExecStartPre={ path=/sbin/modprobe ; argv[]=/sbin/modprobe overlay ; ignore_errors=yes ; start_time=[Thu 2023-09-21 20:40:56 CEST] ; stop_time=[Thu 2023-09-21 20:40:56 CEST] ; pid=4389 ; code=exited ; status=0 } ExecStartPreEx={ path=/sbin/modprobe ; argv[]=/sbin/modprobe overlay ; flags=ignore-failure ; start_time=[Thu 2023-09-21 20:40:56 CEST] ; stop_time=[Thu 2023-09-21 20:40:56 CEST] ; pid=4389 ; code=exited ; status=0 } ExecStart={ path=/usr/bin/containerd ; argv[]=/usr/bin/containerd ; ignore_errors=no ; start_time=[Thu 2023-09-21 20:40:56 CEST] ; stop_time=[n/a] ; pid=4474 ; code=(null) ; status=0/0 } ExecStartEx={ path=/usr/bin/containerd ; argv[]=/usr/bin/containerd ; flags= ; start_time=[Thu 2023-09-21 20:40:56 CEST] ; stop_time=[n/a] ; pid=4474 ; code=(null) ; status=0/0 } Slice=system.slice ControlGroup=/system.slice/containerd.service MemoryCurrent=176992256 MemoryAvailable=infinity CPUUsageNSec=1672419000 EffectiveCPUs=0-15 EffectiveMemoryNodes=0 TasksCurrent=77 IPIngressBytes=[no data] IPIngressPackets=[no data] IPEgressBytes=[no data] IPEgressPackets=[no data] IOReadBytes=18446744073709551615 IOReadOperations=18446744073709551615 IOWriteBytes=18446744073709551615 IOWriteOperations=18446744073709551615 Delegate=yes DelegateControllers=cpu cpuacct cpuset io blkio memory devices pids bpf-firewall bpf-devices bpf-foreign bpf-socket-bind CPUAccounting=yes CPUWeight=[not set] StartupCPUWeight=[not set] CPUShares=[not set] StartupCPUShares=[not set] CPUQuotaPerSecUSec=infinity CPUQuotaPeriodUSec=infinity IOAccounting=no IOWeight=[not set] StartupIOWeight=[not set] BlockIOAccounting=no BlockIOWeight=[not set] StartupBlockIOWeight=[not set] MemoryAccounting=yes DefaultMemoryLow=0 DefaultMemoryMin=0 MemoryMin=0 MemoryLow=0 MemoryHigh=infinity MemoryMax=infinity MemorySwapMax=infinity MemoryLimit=infinity DevicePolicy=auto TasksAccounting=yes TasksMax=infinity IPAccounting=no ManagedOOMSwap=auto ManagedOOMMemoryPressure=auto ManagedOOMMemoryPressureLimit=0 ManagedOOMPreference=none Environment=PATH=/opt/kata/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin UMask=0022 LimitCPU=infinity LimitCPUSoft=infinity LimitFSIZE=infinity LimitFSIZESoft=infinity LimitDATA=infinity LimitDATASoft=infinity LimitSTACK=infinity LimitSTACKSoft=8388608 LimitCORE=infinity LimitCORESoft=infinity LimitRSS=infinity LimitRSSSoft=infinity LimitNOFILE=infinity LimitNOFILESoft=infinity LimitAS=infinity LimitASSoft=infinity LimitNPROC=infinity LimitNPROCSoft=infinity LimitMEMLOCK=8388608 LimitMEMLOCKSoft=8388608 LimitLOCKS=infinity LimitLOCKSSoft=infinity LimitSIGPENDING=54796 LimitSIGPENDINGSoft=54796 LimitMSGQUEUE=819200 LimitMSGQUEUESoft=819200 LimitNICE=0 LimitNICESoft=0 LimitRTPRIO=0 LimitRTPRIOSoft=0 LimitRTTIME=infinity LimitRTTIMESoft=infinity OOMScoreAdjust=-999 CoredumpFilter=0x33 Nice=0 IOSchedulingClass=2 IOSchedulingPriority=4 CPUSchedulingPolicy=0 CPUSchedulingPriority=0 CPUAffinityFromNUMA=no NUMAPolicy=n/a TimerSlackNSec=50000 CPUSchedulingResetOnFork=no NonBlocking=no StandardInput=null StandardOutput=journal StandardError=inherit TTYReset=no TTYVHangup=no TTYVTDisallocate=no SyslogPriority=30 SyslogLevelPrefix=yes SyslogLevel=6 SyslogFacility=3 LogLevelMax=-1 LogRateLimitIntervalUSec=0 LogRateLimitBurst=0 SecureBits=0 CapabilityBoundingSet=cap_chown cap_dac_override cap_dac_read_search cap_fowner cap_fsetid cap_kill cap_setgid cap_setuid cap_setpcap cap_linux_immutable cap_net_bind_service cap_net_broadcast cap_net_admin cap_net_raw cap_ipc_lock cap_ipc_owner cap_sys_module cap_sys_rawio cap_sys_chroot cap_sys_ptrace cap_sys_pacct cap_sys_admin cap_sys_boot cap_sys_nice cap_sys_resource cap_sys_time cap_sys_tty_config cap_mknod cap_lease cap_audit_write cap_audit_control cap_setfcap cap_mac_override cap_mac_admin cap_syslog cap_wake_alarm cap_block_suspend cap_audit_read cap_perfmon cap_bpf cap_checkpoint_restore DynamicUser=no RemoveIPC=no PrivateTmp=no PrivateDevices=no ProtectClock=no ProtectKernelTunables=no ProtectKernelModules=no ProtectKernelLogs=no ProtectControlGroups=no PrivateNetwork=no PrivateUsers=no PrivateMounts=no PrivateIPC=no ProtectHome=no ProtectSystem=no SameProcessGroup=no UtmpMode=init IgnoreSIGPIPE=yes NoNewPrivileges=no SystemCallErrorNumber=2147483646 LockPersonality=no RuntimeDirectoryPreserve=no RuntimeDirectoryMode=0755 StateDirectoryMode=0755 CacheDirectoryMode=0755 LogsDirectoryMode=0755 ConfigurationDirectoryMode=0755 TimeoutCleanUSec=infinity MemoryDenyWriteExecute=no RestrictRealtime=no RestrictSUIDSGID=no RestrictNamespaces=no MountAPIVFS=no KeyringMode=private ProtectProc=default ProcSubset=all ProtectHostname=no KillMode=process KillSignal=15 RestartKillSignal=15 FinalKillSignal=9 SendSIGKILL=yes SendSIGHUP=no WatchdogSignal=6 Id=containerd.service Names=containerd.service Requires=system.slice sysinit.target RequiredBy=docker.service WantedBy=multi-user.target Conflicts=shutdown.target Before=docker.service multi-user.target shutdown.target After=network.target sysinit.target basic.target system.slice systemd-journald.socket local-fs.target Documentation=https://containerd.io Description=containerd container runtime LoadState=loaded ActiveState=active FreezerState=running SubState=running FragmentPath=/lib/systemd/system/containerd.service UnitFileState=enabled UnitFilePreset=enabled StateChangeTimestamp=Thu 2023-09-21 20:40:56 CEST StateChangeTimestampMonotonic=96339464 InactiveExitTimestamp=Thu 2023-09-21 20:40:56 CEST InactiveExitTimestampMonotonic=96147299 ActiveEnterTimestamp=Thu 2023-09-21 20:40:56 CEST ActiveEnterTimestampMonotonic=96339464 ActiveExitTimestamp=n/a ActiveExitTimestampMonotonic=0 InactiveEnterTimestamp=n/a InactiveEnterTimestampMonotonic=0 CanStart=yes CanStop=yes CanReload=no CanIsolate=no CanFreeze=yes StopWhenUnneeded=no RefuseManualStart=no RefuseManualStop=no AllowIsolate=no DefaultDependencies=yes OnSuccessJobMode=fail OnFailureJobMode=replace IgnoreOnIsolate=no NeedDaemonReload=no JobTimeoutUSec=infinity JobRunningTimeoutUSec=infinity JobTimeoutAction=none ConditionResult=yes AssertResult=yes ConditionTimestamp=Thu 2023-09-21 20:40:56 CEST ConditionTimestampMonotonic=96145520 AssertTimestamp=Thu 2023-09-21 20:40:56 CEST AssertTimestampMonotonic=96145527 Transient=no Perpetual=no StartLimitIntervalUSec=10s StartLimitBurst=5 StartLimitAction=none FailureAction=none SuccessAction=none InvocationID=78d4e36c096d4cdc906853c7e47c2ca3 CollectMode=inactive ```
cat /etc/containerd/config.toml
```toml # Copyright 2018-2022 Docker Inc. # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # http://www.apache.org/licenses/LICENSE-2.0 # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. disabled_plugins = ["cri"] #root = "/var/lib/containerd" #state = "/run/containerd" #subreaper = true #oom_score = 0 [plugins] [plugins."io.containerd.grpc.v1.cri"] [plugins."io.containerd.grpc.v1.cri".containerd] default_runtime_name = "kata" [plugins."io.containerd.grpc.v1.cri".containerd.runtimes] [plugins."io.containerd.grpc.v1.cri".containerd.runtimes.kata] runtime_type = "io.containerd.kata.v2" #[grpc] # address = "/run/containerd/containerd.sock" # uid = 0 # gid = 0 #[debug] # address = "/run/containerd/debug.sock" # uid = 0 # gid = 0 # level = "info" ```
Packages
# Packages Have `dpkg`
No `rpm`
---
dpkg -l|egrep "(cc-oci-runtime|cc-runtime|runv|kata-runtime|kata-ksm-throttler|kata-containers-image|linux-container|qemu-)"
``` ii ipxe-qemu-256k-compat-efi-roms 1.0.0+git-20150424.a25a16d-0ubuntu4 all PXE boot firmware - Compat EFI ROM images for qemu ii qemu-block-extra 1:6.2+dfsg-2ubuntu6.12 amd64 extra block backend modules for qemu-system and qemu-utils ii qemu-system-common 1:6.2+dfsg-2ubuntu6.12 amd64 QEMU full system emulation binaries (common files) ii qemu-system-data 1:6.2+dfsg-2ubuntu6.12 all QEMU full system emulation (data files) ii qemu-system-gui 1:6.2+dfsg-2ubuntu6.12 amd64 QEMU full system emulation binaries (user interface and audio support) ii qemu-system-x86 1:6.2+dfsg-2ubuntu6.12 amd64 QEMU full system emulation binaries (x86) ii qemu-utils 1:6.2+dfsg-2ubuntu6.12 amd64 QEMU utilities ```
Kata Monitor
Kata Monitor `kata-monitor`.
---
kata-monitor --version
``` kata-monitor Version: 0.3.0 Go version: go1.19.3 Git commit: ee57732fe08504773b1b5474f2248834ae1fbd66 OS/Arch: linux/amd64 ```
virtiofsd version is 1.3.0
Expected result
OOMs should not cause host mounts to become inaccessible.
Actual result
After any OOM inside the container, all attempts to access the host mount fail with EBADF error.
Adding
–cache never
to virtiofsd's flags did not make a difference.Show virtiofsd logs
Note: these logs are not from the same ctr invocation as above. ``` # Part 1: before OOM, inode=8 works fine Jun 15 21:44:53 hostname virtiofsd[2171030]: Received request: opcode=Lookup (1), inode=3, unique=2586, pid=128 Jun 15 21:44:53 hostname virtiofsd[2171030]: Replying OK, header: OutHeader { len: 144, error: 0, unique: 2586 } Jun 15 21:44:53 hostname virtiofsd[2171030]: Received request: opcode=Getattr (3), inode=8, unique=2588, pid=128 Jun 15 21:44:53 hostname virtiofsd[2171030]: Replying OK, header: OutHeader { len: 120, error: 0, unique: 2588 } Jun 15 21:44:53 hostname virtiofsd[2171030]: Received request: opcode=Opendir (27), inode=8, unique=2590, pid=128 Jun 15 21:44:53 hostname virtiofsd[2171030]: Received request: opcode=Readdirplus (44), inode=8, unique=2592, pid=128 # Part 2: during OOM, inode=8 is requested to be forgotten Jun 15 21:46:36 hostname virtiofsd[2171030]: HIPRIO_QUEUE_EVENT Jun 15 21:46:36 hostname virtiofsd[2171030]: Received request: opcode=Forget (2), inode=8, unique=3048, pid=0 # Part 3: after OOM, inode=8 returns EBADF Jun 15 21:47:59 hostname virtiofsd[2171030]: Received request: opcode=Getattr (3), inode=8, unique=103244, pid=136 Jun 15 21:47:59 hostname virtiofsd[2171030]: Replying ERROR, header: OutHeader { error: -9 (Bad file descriptor), unique: 10 ```Further information
My colleague @kmjohansen has proposed a solution to this problem in this patch series for the kernel's FUSE subsystem. We are, however, unsure if that's the right/best solution. Any thoughts from the kata devs on this?