search

kata-containers / runtime

Kata Containers version 1.x runtime (for version 2.x see https://github.com/kata-containers/kata-containers).
https://katacontainers.io/
Apache License 2.0
2.1k stars 375 forks source link

Sandbox can't be launched when memory exceed 120Gi in centos #1244

Closed wenlxie closed 4 years ago

wenlxie commented 5 years ago

Description of problem

I am now integrating kata runtime with k8s. In the test environment, I found that my pod can't become running because sandbox can't be launched. When the container's request memory is 120Gi +, then we can see this issue. But when the memory is lower than 120Gi, then it works fine.

Expected result

The sandbox can launched

Actual result

It fails.

(replace this text with the output of the kata-collect-data.sh script, after you have reviewed its content to ensure it does not contain any private information).

Meta details

Running kata-collect-data.sh version 1.5.0 (commit 5f7fcd773089a615b776862f92217e987f06df0a) at 2019-02-15.19:36:43.520404111-0700.

Runtime is /usr/local/bin/kata-runtime.

kata-env

Output of "/usr/local/bin/kata-runtime kata-env":

[Meta]
  Version = "1.0.20"

[Runtime]
  Debug = false
  Trace = false
  DisableGuestSeccomp = false
  DisableNewNetNs = false
  Path = "/var/usrlocal/bin/kata-runtime"
  [Runtime.Version]
    Semver = "1.5.0"
    Commit = "5f7fcd773089a615b776862f92217e987f06df0a"
    OCI = "1.0.1-dev"
  [Runtime.Config]
    Path = "/etc/kata-containers/configuration.toml"

[Hypervisor]
  MachineType = "pc"
  Version = "QEMU emulator version 2.11.2(kata-static)\nCopyright (c) 2003-2017 Fabrice Bellard and the QEMU Project developers"
  Path = "/var/usrlocal/kata/opt/kata/bin/qemu-system-x86_64"
  BlockDeviceDriver = "virtio-scsi"
  EntropySource = "/dev/urandom"
  Msize9p = 8192
  MemorySlots = 10
  Debug = false
  UseVSock = false

[Image]
  Path = "/var/usrlocal/kata/opt/kata/share/kata-containers/kata-containers-centos.img"

[Kernel]
  Path = "/var/usrlocal/kata/opt/kata/share/kata-containers/vmlinuz-4.14.67-22"
  Parameters = "init=/usr/lib/systemd/systemd systemd.unit=kata-containers.target systemd.mask=systemd-networkd.service systemd.mask=systemd-networkd.socket"

[Initrd]
  Path = ""

[Proxy]
  Type = "kataProxy"
  Version = "kata-proxy version 1.5.0-9e77a0b1925d1ab3afda7d46ec256be8a9cff222"
  Path = "/usr/local/kata/opt/kata/libexec/kata-containers/kata-proxy"
  Debug = false

[Shim]
  Type = "kataShim"
  Version = "kata-shim version 1.5.0-efbf3bb25065ce89099630b753d218cdc678e758"
  Path = "/var/usrlocal/kata/opt/kata/libexec/kata-containers/kata-shim"
  Debug = false

[Agent]
  Type = "kata"

[Host]
  Kernel = "4.15.0-43.generic.x86_64"
  Architecture = "amd64"
  VMContainerCapable = true
  SupportVSocks = true
  [Host.Distro]
    Name = "CentOS Linux"
    Version = "7"
  [Host.CPU]
    Vendor = "GenuineIntel"
    Model = "Intel(R) Xeon(R) CPU E5-2683 v4 @ 2.10GHz"

[Netmon]
  Version = "kata-netmon version 1.5.0"
  Path = "/usr/local/kata/opt/kata/libexec/kata-containers/kata-netmon"
  Debug = false
  Enable = false

Runtime config files

Runtime default config files

/etc/kata-containers/configuration.toml
/usr/share/defaults/kata-containers/configuration.toml

Runtime config file contents

Output of "cat "/etc/kata-containers/configuration.toml"":

[hypervisor.qemu]
path = "/usr/local/kata/opt/kata/bin/qemu-system-x86_64"
kernel = "/usr/local/kata/opt/kata/share/kata-containers/vmlinuz.container"
image = "/usr/local/kata/opt/kata/share/kata-containers/kata-containers.img"
machine_type = "pc"
kernel_params = ""
firmware = ""
machine_accelerators=""
default_vcpus = 1
default_maxvcpus = 0
default_bridges = 1
#memory_slots = 10
disable_block_device_use = false

block_device_driver = "virtio-scsi"

enable_iothreads = false

#enable_mem_prealloc = true

#enable_hugepages = true

#enable_swap = true

#enable_debug = true

#disable_nesting_checks = true

#msize_9p = 8192

#use_vsock = true

#hotplug_vfio_on_root_bus = true

#disable_vhost_net = true

#entropy_source= "/dev/urandom"

#guest_hook_path = "/usr/share/oci/hooks"

#enable_template = true

[proxy.kata]
path = "/usr/local/kata/opt/kata/libexec/kata-containers/kata-proxy"

#enable_debug = true

[shim.kata]
path = "/usr/local/kata/opt/kata/libexec/kata-containers/kata-shim"

#enable_debug = true

[agent.kata]

[netmon]
#enable_netmon = true

# Specify the path to the netmon binary.
path = "/usr/local/kata/opt/kata/libexec/kata-containers/kata-netmon"

# If enabled, netmon messages will be sent to the system log
# (default: disabled)
#enable_debug = true

[runtime]
internetworking_model="macvtap"
#disable_new_netns = true

Output of "cat "/opt/kata/share/defaults/kata-containers/configuration.toml"":

# Copyright (c) 2017-2019 Intel Corporation
#
# SPDX-License-Identifier: Apache-2.0
#

# XXX: WARNING: this file is auto-generated.
# XXX:
# XXX: Source file: "cli/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/vmlinuz.container"
image = "/opt/kata/share/kata-containers/kata-containers.img"
machine_type = "pc"

# 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 = ""

# 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=""

# 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.
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

# 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

# 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,
# 9pfs is used instead to pass the rootfs.
disable_block_device_use = false

# 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"

# 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 swap of vm memory. Default false.
# The behaviour is undefined if mem_prealloc is also set to true
#enable_swap = true

# This option changes the default hypervisor and kernel parameters
# to enable debug output where available. This extra output is added
# to the proxy logs, but only when proxy debug is also enabled.
#
# 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 true and vsocks are supported, use vsocks to communicate directly
# with the agent and no proxy is started, otherwise use unix
# sockets and start a proxy to communicate with the agent.
# Default false
#use_vsock = 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. This value is valid for "pc" machine type.
# Default false
#hotplug_vfio_on_root_bus = true

# If host doesn't support vhost_net, set to true. Thus we won't create vhost fds for nics.
# Default false
#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"

# 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/osbuilder
#
# Hooks must be stored in a subdirectory of guest_hook_path according to their
# hook type, i.e. "guest_hook_path/{prestart,postart,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 will scanning for hooks,
# but it will not abort container execution.
#guest_hook_path = "/usr/share/oci/hooks"

[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

[proxy.kata]
path = "/opt/kata/libexec/kata-containers/kata-proxy"

# If enabled, proxy messages will be sent to the system log
# (default: disabled)
#enable_debug = true

[shim.kata]
path = "/opt/kata/libexec/kata-containers/kata-shim"

# If enabled, shim messages will be sent to the system log
# (default: disabled)
#enable_debug = true

# If enabled, the shim will create opentracing.io traces and spans.
# (See https://www.jaegertracing.io/docs/getting-started).
#
# Note: By default, the shim runs in a separate network namespace. Therefore,
# to allow it to send trace details to the Jaeger agent running on the host,
# it is necessary to set 'disable_new_netns=true' so that it runs in the host
# network namespace.
#
# (default: disabled)
#enable_tracing = true

[agent.kata]
# There is no field for this section. The goal is only to be able to
# specify which type of agent the user wants to use.

[netmon]
# If enabled, the network monitoring process gets started when the
# sandbox is created. This allows for the detection of some additional
# network being added to the existing network namespace, after the
# sandbox has been created.
# (default: disabled)
#enable_netmon = true

# Specify the path to the netmon binary.
path = "/opt/kata/libexec/kata-containers/kata-netmon"

# If enabled, netmon messages will be sent to the system log
# (default: disabled)
#enable_debug = true

[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:
#
#   - bridged
#     Uses a linux bridge to interconnect the container interface to
#     the VM. Works for most cases except macvlan and ipvlan.
#
#   - 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="macvtap"

# 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

# If enabled, the runtime will create opentracing.io traces and spans.
# (See https://www.jaegertracing.io/docs/getting-started).
# (default: disabled)
#enable_tracing = true

# 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 `enable_netmon`
# `disable_new_netns` conflicts with `internetworking_model=bridged` 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.
# If you are using docker, `disable_new_netns` only works with `docker run --net=none`
# (default: false)
#disable_new_netns = true

Config file /usr/share/defaults/kata-containers/configuration.toml not found

KSM throttler

version

Output of "--version":

kata-collect-data.sh version 1.5.0 (commit 5f7fcd773089a615b776862f92217e987f06df0a)

systemd service

Image details

---
osbuilder:
  url: "https://github.com/kata-containers/osbuilder"
  version: "unknown"
rootfs-creation-time: "2019-02-11T14:55:18.794194275+0000Z"
description: "osbuilder rootfs"
file-format-version: "0.0.2"
architecture: "x86_64"
base-distro:
  name: "Centos"
  version: "7"
  packages:
    default:
      - "bridge-utils"
      - "iproute"
      - "iptables"
      - "net-tools"
      - "nfs-utils"
      - "systemd"
    extra:
      - "bash"
agent:
  url: "https://github.com/kata-containers/agent"
  name: "kata-agent"
  version: "1.5.0-rc2-91a87bc7176750eee2f032aa564e8b28f4ad0e73"
  agent-is-init-daemon: "no"

Initrd details

No initrd

Logfiles

Runtime logs

Recent runtime problems found in system journal:

time="2019-02-08T18:26:20.060071324-07:00" level=error msg="Container ID (72adba7d365c782b16a6fcf66e4becbaaf2606586537e8c6fa38e584003131fe) does not exist" arch=amd64 command=state container=72adba7d365c782b16a6fcf66e4becbaaf2606586537e8c6fa38e584003131fe name=kata-runtime pid=5107 source=runtime
time="2019-02-08T18:26:20.092439463-07:00" level=error msg="Container ID (72adba7d365c782b16a6fcf66e4becbaaf2606586537e8c6fa38e584003131fe) does not exist" arch=amd64 command=state container=72adba7d365c782b16a6fcf66e4becbaaf2606586537e8c6fa38e584003131fe name=kata-runtime pid=5114 source=runtime
time="2019-02-09T21:10:12.95855186-07:00" level=error msg="Container ID (8a6b705405d9b4053799057e51901fa17eb650c128fb22a743f150f6c36dfdea) does not exist" arch=amd64 command=state container=8a6b705405d9b4053799057e51901fa17eb650c128fb22a743f150f6c36dfdea name=kata-runtime pid=13928 source=runtime
time="2019-02-10T22:41:21.514906376-07:00" level=warning msg="fetch sandbox device failed" arch=amd64 command=create container=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df error="open /run/vc/sbs/e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df/devices.json: no such file or directory" name=kata-runtime pid=47233 sandbox=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df sandboxid=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df source=virtcontainers subsystem=sandbox
time="2019-02-10T22:41:22.51165651-07:00" level=warning msg="unsupported address" address="fe80::64b8:adff:febb:5465/64" arch=amd64 command=create container=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df name=kata-runtime pid=47233 source=virtcontainers subsystem=network unsupported-address-type=ipv6
time="2019-02-10T22:41:22.511801094-07:00" level=warning msg="unsupported route" arch=amd64 command=create container=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df destination="fe80::/64" name=kata-runtime pid=47233 source=virtcontainers subsystem=network unsupported-route-type=ipv6
time="2019-02-14T22:29:40.673292897-07:00" level=error msg="rpc error: code = Unavailable desc = transport is closing" arch=amd64 command=create container=8aeaa7c3818326e047e5a53ae94f58ac680b9f955c0c5dcd0c42d5e3a7cee52a name=kata-runtime pid=47389 source=runtime
time="2019-02-14T22:30:25.679861365-07:00" level=error msg="context deadline exceeded" arch=amd64 command=state container=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df name=kata-runtime pid=47741 source=runtime
time="2019-02-14T22:31:10.728727876-07:00" level=error msg="context deadline exceeded" arch=amd64 command=state container=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df name=kata-runtime pid=52828 source=runtime
time="2019-02-15T02:43:34.36673938-07:00" level=warning msg="fetch sandbox device failed" arch=amd64 command=create container=adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c error="open /run/vc/sbs/adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c/devices.json: no such file or directory" name=kata-runtime pid=62757 sandbox=adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c sandboxid=adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c source=virtcontainers subsystem=sandbox
time="2019-02-15T02:43:35.424885934-07:00" level=warning msg="unsupported address" address="fe80::eca0:aaff:fe41:bf8/64" arch=amd64 command=create container=adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c name=kata-runtime pid=62757 source=virtcontainers subsystem=network unsupported-address-type=ipv6
time="2019-02-15T02:43:35.425026329-07:00" level=warning msg="unsupported route" arch=amd64 command=create container=adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c destination="fe80::/64" name=kata-runtime pid=62757 source=virtcontainers subsystem=network unsupported-route-type=ipv6
time="2019-02-15T02:54:57.145504841-07:00" level=error msg="context deadline exceeded" arch=amd64 command=state container=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df name=kata-runtime pid=10914 source=runtime
time="2019-02-15T02:58:00.741052274-07:00" level=error msg="rpc error: code = Unavailable desc = transport is closing" arch=amd64 command=create container=b1b653828793790fa4a5f4de5b75e0cefc3b344f8eb04e2756a64b1471de25ae name=kata-runtime pid=62881 source=runtime
time="2019-02-15T03:03:32.993013847-07:00" level=warning msg="fetch sandbox device failed" arch=amd64 command=create container=0e4ca8780e1157d7be2cd7a72b2b4f3ddfbb7f05aeb6f946352c695d3345d10e error="open /run/vc/sbs/0e4ca8780e1157d7be2cd7a72b2b4f3ddfbb7f05aeb6f946352c695d3345d10e/devices.json: no such file or directory" name=kata-runtime pid=24568 sandbox=0e4ca8780e1157d7be2cd7a72b2b4f3ddfbb7f05aeb6f946352c695d3345d10e sandboxid=0e4ca8780e1157d7be2cd7a72b2b4f3ddfbb7f05aeb6f946352c695d3345d10e source=virtcontainers subsystem=sandbox
time="2019-02-15T03:03:34.097068278-07:00" level=warning msg="unsupported address" address="fe80::90a5:e3ff:fe30:50c1/64" arch=amd64 command=create container=0e4ca8780e1157d7be2cd7a72b2b4f3ddfbb7f05aeb6f946352c695d3345d10e name=kata-runtime pid=24568 source=virtcontainers subsystem=network unsupported-address-type=ipv6
time="2019-02-15T03:03:34.097187524-07:00" level=warning msg="unsupported route" arch=amd64 command=create container=0e4ca8780e1157d7be2cd7a72b2b4f3ddfbb7f05aeb6f946352c695d3345d10e destination="fe80::/64" name=kata-runtime pid=24568 source=virtcontainers subsystem=network unsupported-route-type=ipv6
time="2019-02-15T03:04:26.081288663-07:00" level=error msg="context deadline exceeded" arch=amd64 command=state container=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df name=kata-runtime pid=25182 source=runtime
time="2019-02-15T03:13:11.129955149-07:00" level=warning msg="fetch sandbox device failed" arch=amd64 command=create container=fa74c4c8a35f15d8b95dc8475e98dbb2241ebfb10114a258f96bf25fcde17cea error="open /run/vc/sbs/fa74c4c8a35f15d8b95dc8475e98dbb2241ebfb10114a258f96bf25fcde17cea/devices.json: no such file or directory" name=kata-runtime pid=31443 sandbox=fa74c4c8a35f15d8b95dc8475e98dbb2241ebfb10114a258f96bf25fcde17cea sandboxid=fa74c4c8a35f15d8b95dc8475e98dbb2241ebfb10114a258f96bf25fcde17cea source=virtcontainers subsystem=sandbox
time="2019-02-15T03:13:12.38998092-07:00" level=warning msg="unsupported address" address="fe80::784f:86ff:febd:e623/64" arch=amd64 command=create container=fa74c4c8a35f15d8b95dc8475e98dbb2241ebfb10114a258f96bf25fcde17cea name=kata-runtime pid=31443 source=virtcontainers subsystem=network unsupported-address-type=ipv6
time="2019-02-15T03:13:12.390084088-07:00" level=warning msg="unsupported route" arch=amd64 command=create container=fa74c4c8a35f15d8b95dc8475e98dbb2241ebfb10114a258f96bf25fcde17cea destination="fe80::/64" name=kata-runtime pid=31443 source=virtcontainers subsystem=network unsupported-route-type=ipv6

Proxy logs

Recent proxy problems found in system journal:

time="2019-02-10T22:19:39.335534648-07:00" level=fatal msg="failed to handle exit signal" error="close unix @->/run/vc/vm/320650fc9dbdf75465b74bf53f01a1e58191c7f0fabefd81f636d2c39b1bb92d/kata.sock: use of closed network connection" name=kata-proxy pid=14990 sandbox=320650fc9dbdf75465b74bf53f01a1e58191c7f0fabefd81f636d2c39b1bb92d source=proxy
time="2019-02-14T22:29:40.33853425-07:00" level=warning msg="2019/02/14 22:29:40 [ERR] yamux: Failed to write header: write unix @->/run/vc/vm/e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df/kata.sock: write: broken pipe\n" component=yamux name=kata-proxy pid=47253 sandbox=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df source=proxy
time="2019-02-14T22:29:40.338731988-07:00" level=fatal msg="channel error" error="session shutdown" name=kata-proxy pid=47253 sandbox=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df source=proxy
time="2019-02-15T02:58:00.394448389-07:00" level=fatal msg="channel error" error="session shutdown" name=kata-proxy pid=62776 sandbox=adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c source=proxy

Shim logs

Recent shim problems found in system journal:

time="2019-02-14T22:29:40.341281038-07:00" level=error msg="failed waiting for process" container=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df error="rpc error: code = Unavailable desc = transport is closing" exec-id=e5f1ce65e7af56facf1289971d0fee1cd5b83d8b6e35015515e21d0052e583df name=kata-shim pid=1 source=shim
time="2019-02-15T03:00:20.396223073-07:00" level=error msg="failed waiting for process" container=adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c error="rpc error: code = Unavailable desc = all SubConns are in TransientFailure, latest connection error: connection error: desc = \"transport: Error while dialing rpc error: code = DeadlineExceeded desc = timed out connecting to unix socket ////run/vc/sbs/adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c/proxy.sock\"" exec-id=adc9b60a7ce858a6b5b570b947021465e9161167a905c70a8be942e660ef9b6c name=kata-shim pid=1 source=shim

Throttler logs

No recent throttler problems found in system journal.

Container manager details

Have docker

Docker

Output of "docker version":

Client:
 Version:           18.09.2
 API version:       1.39
 Go version:        go1.10.6
 Git commit:        6247962
 Built:             Sun Feb 10 04:13:27 2019
 OS/Arch:           linux/amd64
 Experimental:      false

Server: Docker Engine - Community
 Engine:
  Version:          18.09.2
  API version:      1.39 (minimum version 1.12)
  Go version:       go1.10.6
  Git commit:       6247962
  Built:            Sun Feb 10 03:47:25 2019
  OS/Arch:          linux/amd64
  Experimental:     false

Output of "docker info":

Containers: 0
 Running: 0
 Paused: 0
 Stopped: 0
Images: 0
Server Version: 18.09.2
Storage Driver: overlay2
 Backing Filesystem: xfs
 Supports d_type: true
 Native Overlay Diff: true
Logging Driver: json-file
Cgroup Driver: cgroupfs
Plugins:
 Volume: local
 Network: bridge host macvlan null overlay
 Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog
Swarm: inactive
Runtimes: runc
Default Runtime: runc
Init Binary: docker-init
containerd version: 9754871865f7fe2f4e74d43e2fc7ccd237edcbce
runc version: 09c8266bf2fcf9519a651b04ae54c967b9ab86ec
init version: fec3683
Security Options:
 seccomp
  Profile: default
Kernel Version: 4.15.0-43.generic.x86_64
Operating System: CentOS Linux 7 (Core)
OSType: linux
Architecture: x86_64
CPUs: 64
Total Memory: 377.5GiB
Name: xxxxxxxx
ID: XIQY:V6PV:P76V:GYXC:EVBX:O727:PX4H:BFTR:ILE6:J7XW:QXIF:UKA4
Docker Root Dir: /var/lib/docker
Debug Mode (client): false
Debug Mode (server): false
Registry: https://index.docker.io/v1/
Labels:
Experimental: false
Insecure Registries:
 127.0.0.0/8
Live Restore Enabled: true
Product License: Community Engine

WARNING: No swap limit support
WARNING: bridge-nf-call-iptables is disabled
WARNING: bridge-nf-call-ip6tables is disabled

Output of "systemctl show docker":

Type=notify
Restart=always
NotifyAccess=main
RestartUSec=2s
TimeoutStartUSec=0
TimeoutStopUSec=0
WatchdogUSec=0
WatchdogTimestamp=Fri 2019-02-15 03:10:24 -07
WatchdogTimestampMonotonic=19318162
StartLimitInterval=60000000
StartLimitBurst=3
StartLimitAction=none
FailureAction=none
PermissionsStartOnly=no
RootDirectoryStartOnly=no
RemainAfterExit=no
GuessMainPID=yes
MainPID=13697
ControlPID=0
FileDescriptorStoreMax=0
StatusErrno=0
Result=success
ExecMainStartTimestamp=Fri 2019-02-15 03:10:19 -07
ExecMainStartTimestampMonotonic=14344567
ExecMainExitTimestampMonotonic=0
ExecMainPID=13697
ExecMainCode=0
ExecMainStatus=0
ExecStart={ path=/usr/bin/dockerd ; argv[]=/usr/bin/dockerd -H fd:// ; ignore_errors=no ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; code=(null) ; status=0/0 }
ExecStartPost={ path=/usr/sbin/iptables ; argv[]=/usr/sbin/iptables -P FORWARD ACCEPT ; ignore_errors=no ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; 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 }
Slice=system.slice
ControlGroup=/system.slice/docker.service
MemoryCurrent=159477760
TasksCurrent=147
Delegate=yes
CPUAccounting=no
CPUShares=18446744073709551615
StartupCPUShares=18446744073709551615
CPUQuotaPerSecUSec=infinity
BlockIOAccounting=no
BlockIOWeight=18446744073709551615
StartupBlockIOWeight=18446744073709551615
MemoryAccounting=no
MemoryLimit=18446744073709551615
DevicePolicy=auto
TasksAccounting=no
TasksMax=18446744073709551615
Environment=PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin
UMask=0022
LimitCPU=18446744073709551615
LimitFSIZE=18446744073709551615
LimitDATA=18446744073709551615
LimitSTACK=18446744073709551615
LimitCORE=18446744073709551615
LimitRSS=18446744073709551615
LimitNOFILE=18446744073709551615
LimitAS=18446744073709551615
LimitNPROC=18446744073709551615
LimitMEMLOCK=65536
LimitLOCKS=18446744073709551615
LimitSIGPENDING=1546157
LimitMSGQUEUE=819200
LimitNICE=0
LimitRTPRIO=0
LimitRTTIME=18446744073709551615
OOMScoreAdjust=0
Nice=0
IOScheduling=0
CPUSchedulingPolicy=0
CPUSchedulingPriority=0
TimerSlackNSec=50000
CPUSchedulingResetOnFork=no
NonBlocking=no
StandardInput=null
StandardOutput=journal
StandardError=inherit
TTYReset=no
TTYVHangup=no
TTYVTDisallocate=no
SyslogPriority=30
SyslogLevelPrefix=yes
SecureBits=0
CapabilityBoundingSet=18446744073709551615
AmbientCapabilities=0
MountFlags=0
PrivateTmp=no
PrivateNetwork=no
PrivateDevices=no
ProtectHome=no
ProtectSystem=no
SameProcessGroup=no
IgnoreSIGPIPE=yes
NoNewPrivileges=no
SystemCallErrorNumber=0
RuntimeDirectoryMode=0755
KillMode=process
KillSignal=15
SendSIGKILL=yes
SendSIGHUP=no
Id=docker.service
Names=docker.service
Requires=basic.target docker.socket
Wants=network-online.target system.slice
WantedBy=multi-user.target
ConsistsOf=docker.socket
Conflicts=shutdown.target
Before=shutdown.target multi-user.target
After=firewalld.service docker.socket system.slice network-online.target systemd-journald.socket basic.target
TriggeredBy=docker.socket
Documentation=https://docs.docker.com
Description=Docker Application Container Engine
LoadState=loaded
ActiveState=active
SubState=running
FragmentPath=/usr/lib/systemd/system/docker.service
UnitFileState=enabled
UnitFilePreset=disabled
InactiveExitTimestamp=Fri 2019-02-15 03:10:19 -07
InactiveExitTimestampMonotonic=14344610
ActiveEnterTimestamp=Fri 2019-02-15 03:10:24 -07
ActiveEnterTimestampMonotonic=19321276
ActiveExitTimestampMonotonic=0
InactiveEnterTimestampMonotonic=0
CanStart=yes
CanStop=yes
CanReload=yes
CanIsolate=no
StopWhenUnneeded=no
RefuseManualStart=no
RefuseManualStop=no
AllowIsolate=no
DefaultDependencies=yes
OnFailureJobMode=replace
IgnoreOnIsolate=no
IgnoreOnSnapshot=no
NeedDaemonReload=no
JobTimeoutUSec=0
JobTimeoutAction=none
ConditionResult=yes
AssertResult=yes
ConditionTimestamp=Fri 2019-02-15 03:10:19 -07
ConditionTimestampMonotonic=14342957
AssertTimestamp=Fri 2019-02-15 03:10:19 -07
AssertTimestampMonotonic=14342958
Transient=no

Have kubectl

Kubernetes

Output of "kubectl version":

Client Version: version.Info{Major:"1", Minor:"12+", GitVersion:"v1.12.3-439+fabd6e72745b91", GitCommit:"fabd6e72745b91409cd03e44d783d26a101e9198", GitTreeState:"clean", BuildDate:"2019-01-29T05:50:07Z", GoVersion:"go1.10.4", Compiler:"gc", Platform:"linux/amd64"}
The connection to the server localhost:8080 was refused - did you specify the right host or port?

Output of "kubectl config view":

apiVersion: v1
clusters: []
contexts: []
current-context: ""
kind: Config
preferences: {}
users: []

Output of "systemctl show kubelet":

Type=simple
Restart=always
NotifyAccess=none
RestartUSec=2s
TimeoutStartUSec=1min 30s
TimeoutStopUSec=1min 30s
WatchdogUSec=0
WatchdogTimestamp=Fri 2019-02-15 03:10:33 -07
WatchdogTimestampMonotonic=27557835
StartLimitInterval=0
StartLimitBurst=5
StartLimitAction=none
FailureAction=none
PermissionsStartOnly=no
RootDirectoryStartOnly=no
RemainAfterExit=no
GuessMainPID=yes
MainPID=21040
ControlPID=0
FileDescriptorStoreMax=0
StatusErrno=0
Result=success
ExecMainStartTimestamp=Fri 2019-02-15 03:10:33 -07
ExecMainStartTimestampMonotonic=27557802
ExecMainExitTimestampMonotonic=0
ExecMainPID=21040
ExecMainCode=0
ExecMainStatus=0
ExecStart={ path=/usr/local/bin/kubelet ; argv[]=/usr/local/bin/kubelet $DAEMON_ARGS ; ignore_errors=no ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; code=(null) ; status=0/0 }
Slice=system.slice
ControlGroup=/system.slice/kubelet.service
MemoryCurrent=48115712
TasksCurrent=0
Delegate=no
CPUAccounting=no
CPUShares=18446744073709551615
StartupCPUShares=18446744073709551615
CPUQuotaPerSecUSec=infinity
BlockIOAccounting=no
BlockIOWeight=18446744073709551615
StartupBlockIOWeight=18446744073709551615
MemoryAccounting=no
MemoryLimit=18446744073709551615
DevicePolicy=auto
TasksAccounting=no
TasksMax=18446744073709551615
EnvironmentFile=/etc/sysconfig/kubelet (ignore_errors=no)
UMask=0022
LimitCPU=18446744073709551615
LimitFSIZE=18446744073709551615
LimitDATA=18446744073709551615
LimitSTACK=18446744073709551615
LimitCORE=18446744073709551615
LimitRSS=18446744073709551615
LimitNOFILE=4096
LimitAS=18446744073709551615
LimitNPROC=1546157
LimitMEMLOCK=65536
LimitLOCKS=18446744073709551615
LimitSIGPENDING=1546157
LimitMSGQUEUE=819200
LimitNICE=0
LimitRTPRIO=0
LimitRTTIME=18446744073709551615
OOMScoreAdjust=0
Nice=0
IOScheduling=0
CPUSchedulingPolicy=0
CPUSchedulingPriority=0
TimerSlackNSec=50000
CPUSchedulingResetOnFork=no
NonBlocking=no
StandardInput=null
StandardOutput=journal
StandardError=inherit
TTYReset=no
TTYVHangup=no
TTYVTDisallocate=no
SyslogPriority=30
SyslogLevelPrefix=yes
SecureBits=0
CapabilityBoundingSet=18446744073709551615
AmbientCapabilities=0
MountFlags=0
PrivateTmp=no
PrivateNetwork=no
PrivateDevices=no
ProtectHome=no
ProtectSystem=no
SameProcessGroup=no
IgnoreSIGPIPE=yes
NoNewPrivileges=no
SystemCallErrorNumber=0
RuntimeDirectoryMode=0755
KillMode=process
KillSignal=15
SendSIGKILL=yes
SendSIGHUP=no
Id=kubelet.service
Names=kubelet.service
Requires=criproxy.service basic.target
Wants=criproxy.service system.slice
WantedBy=multi-user.target
Conflicts=shutdown.target
Before=shutdown.target multi-user.target
After=criproxy.service systemd-journald.socket basic.target system.slice
Documentation=https://github.com/kubernetes/kubernetes
Description=Kubernetes Kubelet Server
LoadState=loaded
ActiveState=active
SubState=running
FragmentPath=/etc/systemd/system/kubelet.service
UnitFileState=enabled
UnitFilePreset=disabled
InactiveExitTimestamp=Fri 2019-02-15 03:10:33 -07
InactiveExitTimestampMonotonic=27557850
ActiveEnterTimestamp=Fri 2019-02-15 03:10:33 -07
ActiveEnterTimestampMonotonic=27557850
ActiveExitTimestampMonotonic=0
InactiveEnterTimestampMonotonic=0
CanStart=yes
CanStop=yes
CanReload=no
CanIsolate=no
StopWhenUnneeded=no
RefuseManualStart=no
RefuseManualStop=no
AllowIsolate=no
DefaultDependencies=yes
OnFailureJobMode=replace
IgnoreOnIsolate=no
IgnoreOnSnapshot=no
NeedDaemonReload=no
JobTimeoutUSec=0
JobTimeoutAction=none
ConditionResult=yes
AssertResult=yes
ConditionTimestamp=Fri 2019-02-15 03:10:33 -07
ConditionTimestampMonotonic=27555739
AssertTimestamp=Fri 2019-02-15 03:10:33 -07
AssertTimestampMonotonic=27555739
Transient=no

No crio

Packages

No dpkg Have rpm Output of "rpm -qa|egrep "(cc-oci-runtimecc-runtimerunv|kata-proxy|kata-runtime|kata-shim|kata-ksm-throttler|kata-containers-image|linux-container|qemu-)"":

wenlxie commented 5 years ago

We install the packages via get the 1.5.0 release tar and then install it via binary.

wenlxie commented 5 years ago

If I enable the console log, then from the console, I can see the error:

[   61.832516] acpi PNP0C80:01: add_memory failed
[   61.834407] acpi PNP0C80:01: acpi_memory_enable_device() error
[   61.838092] Out of memory: Kill process 133 (kata-agent) score 4 or sacrifice child
[   61.841639] Killed process 204 (pause) total-vm:3824kB, anon-rss:1000kB, file-rss:4kB, shmem-rss:0kB
time="2019-02-15T05:42:36.692647012Z" level=info msg="ignoring unexpected signal" name=kata-agent pid=133 sandbox=66168c006afb63ba156a51f496607dff06e0560ade0d1d1e30f78bff92486b45 signal="child exited" source=agent
[   61.887771] Out of memory: Kill process 133 (kata-agent) score 4 or sacrifice child
[   61.890555] Killed process 133 (kata-agent) total-vm:463460kB, anon-rss:8760kB, file-rss:4kB, shmem-rss:0kB
[   61.937938] systemd-udevd[92]: worker [1189] did not accept message -111 (Connection refused), kill it
[   61.941428] systemd-udevd[92]: worker [1194] did not accept message -111 (Connection refused), kill it
[   61.944994] systemd-udevd[92]: worker [1196] did not accept message -111 (Connection refused), kill it
[   61.948558] systemd-udevd[92]: worker [1197] did not accept message -111 (Connection refused), kill it
[   61.994515] reboot: Power down
wenlxie commented 5 years ago

The image [Image] Path = "/var/usrlocal/kata/opt/kata/share/kata-containers/kata-containers-centos.img" is the image we build via osbuilder with debug log enabled. If we change to the default image, the issue still exists.

The console log:

SeaBIOS (version rel-1.11.0-0-g63451fca13-prebuilt.qemu-project.org)
Machine UUID 5487246d-5c9a-4cb7-a1e7-7e685743d05a
Booting from ROM..
bash-4.2# time="2019-02-15T08:00:33.765152406Z" level=info msg=announce device-handlers="mmioblk,blk,scsi,nvdimm" name=kata-agent pid=132 source=agent storage-handlers="9p,blk,mmioblk,sc3
time="2019-02-15T08:00:33.771298882Z" level=info msg="Started listening for uevents" name=kata-agent pid=132 source=agent subsystem=udevlistener
time="2019-02-15T08:00:33.775000008Z" level=info msg="agent grpc server starts" name=kata-agent pid=132 source=agent
time="2019-02-15T08:00:33.781045952Z" level=info msg="Getting interface from MAC address" interface-name=eth0 mac-address="ba:9a:da:d5:35:2b" name=kata-agent pci-address= pid=132 source=t
time="2019-02-15T08:00:33.78471215Z" level=info msg="Link found" interface-name=eth0 link="&{LinkAttrs:{Index:2 MTU:1500 TxQLen:1000 Name:eth0 HardwareAddr:ba:9a:da:d5:35:2b Flags:broadct
time="2019-02-15T08:00:33.799826234Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/di
time="2019-02-15T08:00:33.874638155Z" level=info msg="ignoring unexpected signal" name=kata-agent pid=132 sandbox=bd351b19f7fb770d9cdc6fafce913842e461259a10f9728089052326a14d4044 signal=t
time="2019-02-15T08:00:33.877933761Z" level=info msg="ignoring unexpected signal" name=kata-agent pid=132 sandbox=bd351b19f7fb770d9cdc6fafce913842e461259a10f9728089052326a14d4044 signal=t
time="2019-02-15T08:00:34.166432874Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.212933357Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.245957149Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.257942384Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.271811607Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.283766778Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.296702003Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.308916502Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.32091496Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/deu
time="2019-02-15T08:00:34.332937527Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.34592971Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/deu
time="2019-02-15T08:00:34.357923155Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.369918853Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.382504483Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.394920612Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.40691389Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/deu
time="2019-02-15T08:00:34.418923452Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.43182574Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/deu
time="2019-02-15T08:00:34.443916412Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:34.456540331Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/du
time="2019-02-15T08:00:36.662756147Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname=cpu/1/msr uevent-dr
time="2019-02-15T08:00:36.670462073Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname=cpu/1/cpuid ueventd
time="2019-02-15T08:00:36.670545353Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/dl
time="2019-02-15T08:00:36.670901527Z" level=info msg="Received add uevent" name=kata-agent pid=132 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/dy
[  184.754738] Out of memory: Kill process 132 (kata-agent) score 4 or sacrifice child
[  184.758297] Killed process 197 (pause) total-vm:3824kB, anon-rss:996kB, file-rss:4kB, shmem-rss:0kB
[  184.771424] Out of memory: Kill process 132 (kata-agent) score 4 or sacrifice child
[  184.775237] Killed process 132 (kata-agent) total-vm:529060kB, anon-rss:8732kB, file-rss:4kB, shmem-rss:0kB
[  184.869191] Out of memory: Kill process 1431 (systemd-udevd) score 1 or sacrifice child
[  184.872790] Killed process 1431 (systemd-udevd) total-vm:43516kB, anon-rss:2292kB, file-rss:0kB, shmem-rss:0kB
[  194.877789] reboot: Power down
wenlxie commented 5 years ago
  1. The qemu start parameter -m 2048M,slots=20,maxmem=387630M 2. From the kata runtime log, both object-add and device-add has no errors during memory hotplug
wenlxie commented 5 years ago

Is there any memory hotplug limitation per dimm? All of the memory will be add in the same dimm no matter how big the memory size is.

wenlxie commented 5 years ago

@grahamwhaley @egernst @jcvenegas @gnawux @caoruidong Please help to have a check.

wenlxie commented 5 years ago

Sometimes at the boot log, there are some trace, but not exist everytime:

time="2019-02-18T05:36:28.552089315Z" level=info msg="Received add uevent" name=kata-agent pid=138 source=agent subsystem=udevlistener uevent-action=add uevent-devname= uevent-devpath=/devices/system/memory/memory36 uevent-seqnum=1593 uev
ent-subsystem=memory
[    7.172271] kworker/u128:2: page allocation failure: order:9, mode:0x14084c0(GFP_KERNEL|__GFP_RETRY_MAYFAIL|__GFP_ZERO), nodemask=(null)
[    7.177327] kworker/u128:2 cpuset=/ mems_allowed=0
[    7.179504] CPU: 0 PID: 95 Comm: kworker/u128:2 Not tainted 4.14.67 #1
[    7.182229] Workqueue: kacpi_hotplug acpi_hotplug_work_fn
[    7.184455] Call Trace:
[    7.185465]  dump_stack+0x5c/0x78
[    7.186821]  warn_alloc.cold.32+0x93/0x125
[    7.188785]  ? compaction_zonelist_suitable+0xc2/0x130
[    7.190984]  __alloc_pages_nodemask+0xcc5/0xda0
[    7.192983]  vmemmap_alloc_block+0x35/0x73
[    7.194810]  __vmemmap_alloc_block_buf+0xfc/0x108
[    7.196758]  vmemmap_populate+0xee/0x227
[    7.198453]  sparse_mem_map_populate+0x22/0x32
[    7.200538]  sparse_add_one_section+0x4a/0x157
[    7.202417]  __add_pages+0x82/0x1b0
[    7.203842]  add_pages+0x10/0x60
[    7.205262]  add_memory_resource+0xe5/0x150
[    7.206981]  add_memory+0x7b/0xc0
[    7.208471]  ? acpi_memory_get_resource+0xe0/0xe0
[    7.210582]  acpi_memory_device_add+0x120/0x290
[    7.212459]  acpi_bus_attach+0xc5/0x1c0
[    7.214084]  acpi_bus_scan+0x2e/0x60
[    7.215469]  acpi_device_hotplug+0x248/0x420
[    7.217354]  acpi_hotplug_work_fn+0x15/0x20
[    7.219175]  process_one_work+0x1a1/0x300
[    7.220838]  worker_thread+0x26/0x3d0
[    7.222366]  ? process_one_work+0x300/0x300
[    7.224266]  kthread+0x10f/0x130
[    7.225734]  ? kthread_create_on_node+0x40/0x40
[    7.227670]  ret_from_fork+0x35/0x40
[    7.229298] Mem-Info:
[    7.230355] active_anon:4983 inactive_anon:2102 isolated_anon:0
[    7.230355]  active_file:23 inactive_file:22 isolated_file:0
[    7.230355]  unevictable:0 dirty:18 writeback:0 unstable:0
[    7.230355]  slab_reclaimable:589 slab_unreclaimable:3504
[    7.230355]  mapped:677 shmem:2110 pagetables:381 bounce:0
[    7.230355]  free:4288 free_pcp:232 free_cma:0
[    7.244002] Node 0 active_anon:19932kB inactive_anon:8408kB active_file:92kB inactive_file:88kB unevictable:0kB isolated(anon):0kB isolated(file):0kB mapped:2708kB dirty:72kB writeback:0kB shmem:8440kB writeback_tmp:0kB unstable:0kB al
l_unreclaimable? no
[    7.253727] DMA free:9420kB min:44kB low:56kB high:68kB active_anon:8kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB writepending:0kB present:15992kB managed:15908kB mlocked:0kB kernel_stack:0kB pagetables:0kB bo
unce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB
[    7.263882] lowmem_reserve[]: 0 1964 1964 1964
[    7.265738] DMA32 free:7628kB min:5648kB low:7656kB high:9664kB active_anon:19932kB inactive_anon:8408kB active_file:92kB inactive_file:88kB unevictable:0kB writepending:0kB present:2080636kB managed:2016036kB mlocked:0kB kernel_stack:
2192kB pagetables:1524kB bounce:0kB free_pcp:928kB local_pcp:732kB free_cma:0kB

then there will be errors:

fatal error: runtime: cannot allocate memory
mrlhansen commented 4 years ago

I experienced the exact same issue on our system today. Has any progress been made since this issue was opened?

devimc commented 4 years ago

@mrlhansen looks like this bug was fixed here https://github.com/kata-containers/packaging/pull/335 what version of kata containers are you using?

mrlhansen commented 4 years ago

@devimc I am using 1.11.0-rc0, specifically the packages from this repository. I checked that our kernel (v5.2.20) is compiled with the flag mentioned in the other issue.

[root@host ~]# grep CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE /boot/config-$(uname -r)
CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE=y
devimc commented 4 years ago

I checked that our kernel (v5.2.20) is compiled with the flag mentioned in the other issue. [root@host ~]# grep CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE /boot/config-$(uname -r) CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE=y

v5.2.20? this is the host kernel, right? CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE is required in the guest, not in the host, if you are using a custom guest kernel, make sure it includes this feature.

mrlhansen commented 4 years ago

@devimc This is the host kernel yes. I did not realize the flag should be in the guest kernel - but that actually makes a lot more sense. Thanks a lot, I will try it out.

devimc commented 4 years ago

closing issue, since it was fixed here kata-containers/packaging#335, feel free to reopen if needed

mrlhansen commented 4 years ago

I updated to the latest version of Kata Containers (v1.11.1) and I am using the default kernel and configuration that ships with the packages - and I am still experiencing the same problem. When I specify 120Gi it works, but anything above that fails. According to the kernel config file in the kata-linux-container-debug package, this kernel does have the CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE variable correctly set.

devimc commented 4 years ago

@mrlhansen please run sudo kata-collect-data.sh and paste it here please

mrlhansen commented 4 years ago

@devimc The output is below. I am running everything via Kubernetes using the CRI plugin. For completeness, this is the yaml file I use to deploy the container.

apiVersion: apps/v1
kind: Deployment
metadata:
  name: alpine-kata
  namespace: default
  annotations:
    io.kubernetes.cri.untrusted-workload: "true"
spec:
  selector:
    matchLabels:
      workload.user.cattle.io/workloadselector: deployment-default-alpine-kata
  template:
    metadata:
      annotations:
        io.kubernetes.cri.untrusted-workload: "true"
      labels:
        workload.user.cattle.io/workloadselector: deployment-default-alpine-kata
    spec:
      containers:
      - args:
        - tail
        - -f
        - /dev/null
        image: alpine:latest
        imagePullPolicy: Always
        name: alpine-kata
        resources:
          limits:
            cpu: "32"
            memory: 120Gi
          requests:
            cpu: "32"
            memory: 120Gi
        securityContext:
          allowPrivilegeEscalation: false
          capabilities: {}
          privileged: false
          readOnlyRootFilesystem: false
          runAsNonRoot: false
Show kata-collect-data.sh details

# Meta details Running `kata-collect-data.sh` version `1.11.1 (commit 7b6f3b64873bbeed63f2f7937d83f1ed6ccd258a)` at `2020-06-18.08:21:58.538117677+0200`. --- Runtime is `/bin/kata-runtime`. # `kata-env` Output of "`/bin/kata-runtime kata-env`": ```toml [Meta] Version = "1.0.24" [Runtime] Debug = false Trace = false DisableGuestSeccomp = true DisableNewNetNs = false SandboxCgroupOnly = false Path = "/usr/bin/kata-runtime" [Runtime.Version] OCI = "1.0.1-dev" [Runtime.Version.Version] Semver = "1.11.1" Major = 1 Minor = 11 Patch = 1 Commit = "7b6f3b64873bbeed63f2f7937d83f1ed6ccd258a" [Runtime.Config] Path = "/usr/share/defaults/kata-containers/configuration.toml" [Hypervisor] MachineType = "pc" Version = "QEMU emulator version 4.1.1\nCopyright (c) 2003-2019 Fabrice Bellard and the QEMU Project developers" Path = "/usr/bin/qemu-vanilla-system-x86_64" BlockDeviceDriver = "virtio-scsi" EntropySource = "/dev/urandom" SharedFS = "virtio-9p" VirtioFSDaemon = "/usr/bin/virtiofsd" Msize9p = 8192 MemorySlots = 10 PCIeRootPort = 0 HotplugVFIOOnRootBus = false Debug = false UseVSock = false [Image] Path = "/usr/share/kata-containers/kata-containers-image_clearlinux_1.11.1_agent_34a000d0cf.img" [Kernel] Path = "/usr/share/kata-containers/vmlinuz-5.4.32.73-62.2.container" Parameters = "systemd.unit=kata-containers.target systemd.mask=systemd-networkd.service systemd.mask=systemd-networkd.socket scsi_mod.scan=none" [Initrd] Path = "" [Proxy] Type = "kataProxy" Path = "/usr/libexec/kata-containers/kata-proxy" Debug = false [Proxy.Version] Semver = "1.11.1-f61beff" Major = 1 Minor = 11 Patch = 1 Commit = "<>" [Shim] Type = "kataShim" Path = "/usr/libexec/kata-containers/kata-shim" Debug = false [Shim.Version] Semver = "1.11.1-9938269" Major = 1 Minor = 11 Patch = 1 Commit = "<>" [Agent] Type = "kata" Debug = false Trace = false TraceMode = "" TraceType = "" [Host] Kernel = "5.2.20-2.el7.cp" Architecture = "amd64" VMContainerCapable = true SupportVSocks = true [Host.Distro] Name = "CentOS Linux" Version = "7" [Host.CPU] Vendor = "GenuineIntel" Model = "Intel(R) Xeon(R) Gold 6130 CPU @ 2.10GHz" [Netmon] Path = "/usr/libexec/kata-containers/kata-netmon" Debug = false Enable = false [Netmon.Version] Semver = "1.11.1" Major = 1 Minor = 11 Patch = 1 Commit = "<>" ``` --- # Runtime config files ## Runtime default config files ``` /etc/kata-containers/configuration.toml /usr/share/defaults/kata-containers/configuration.toml ``` ## Runtime config file contents Config file `/etc/kata-containers/configuration.toml` not found Output of "`cat "/usr/share/defaults/kata-containers/configuration.toml"`": ```toml # Copyright (c) 2017-2019 Intel Corporation # # SPDX-License-Identifier: Apache-2.0 # # XXX: WARNING: this file is auto-generated. # XXX: # XXX: Source file: "cli/config/configuration-qemu.toml.in" # XXX: Project: # XXX: Name: Kata Containers # XXX: Type: kata [hypervisor.qemu] path = "/usr/bin/qemu-vanilla-system-x86_64" kernel = "/usr/share/kata-containers/vmlinuz.container" image = "/usr/share/kata-containers/kata-containers.img" machine_type = "pc" # 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 = "" # 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="" # 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. 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 # 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, # 9pfs is used instead to pass the rootfs. disable_block_device_use = false # Shared file system type: # - virtio-9p (default) # - virtio-fs shared_fs = "virtio-9p" # Path to vhost-user-fs daemon. virtio_fs_daemon = "/usr/bin/virtiofsd" # Default size of DAX cache in MiB virtio_fs_cache_size = 1024 # Extra args for virtiofsd daemon # # Format example: # ["-o", "arg1=xxx,arg2", "-o", "hello world", "--arg3=yyy"] # # see `virtiofsd -h` for possible options. virtio_fs_extra_args = [] # Cache mode: # # - none # 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 = "always" # 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" # 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 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 = "" # Enable swap of vm memory. Default false. # The behaviour is undefined if mem_prealloc is also set to true #enable_swap = true # This option changes the default hypervisor and kernel parameters # to enable debug output where available. This extra output is added # to the proxy logs, but only when proxy debug is also enabled. # # 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 true and vsocks are supported, use vsocks to communicate directly # with the agent and no proxy is started, otherwise use unix # sockets and start a proxy to communicate with the agent. # Default false #use_vsock = true # If false and nvdimm is supported, use nvdimm device to plug guest image. # Otherwise virtio-block device is used. # 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. This value is valid for "pc" machine type. # 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" # 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/osbuilder # # Hooks must be stored in a subdirectory of guest_hook_path according to their # hook type, i.e. "guest_hook_path/{prestart,postart,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 will scanning for hooks, # but it will not abort container execution. #guest_hook_path = "/usr/share/oci/hooks" [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" [proxy.kata] path = "/usr/libexec/kata-containers/kata-proxy" # If enabled, proxy messages will be sent to the system log # (default: disabled) #enable_debug = true [shim.kata] path = "/usr/libexec/kata-containers/kata-shim" # If enabled, shim messages will be sent to the system log # (default: disabled) #enable_debug = true # If enabled, the shim will create opentracing.io traces and spans. # (See https://www.jaegertracing.io/docs/getting-started). # # Note: By default, the shim runs in a separate network namespace. Therefore, # to allow it to send trace details to the Jaeger agent running on the host, # it is necessary to set 'disable_new_netns=true' so that it runs in the host # network namespace. # # (default: disabled) #enable_tracing = true [agent.kata] # If enabled, make the agent display debug-level messages. # (default: disabled) #enable_debug = true # Enable agent tracing. # # If enabled, the default trace mode is "dynamic" and the # default trace type is "isolated". The trace mode and type are set # explicity with the `trace_type=` and `trace_mode=` options. # # Notes: # # - Tracing is ONLY enabled when `enable_tracing` is set: explicitly # setting `trace_mode=` and/or `trace_type=` without setting `enable_tracing` # will NOT activate agent tracing. # # - See https://github.com/kata-containers/agent/blob/master/TRACING.md for # full details. # # (default: disabled) #enable_tracing = true # #trace_mode = "dynamic" #trace_type = "isolated" # 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=[] [netmon] # If enabled, the network monitoring process gets started when the # sandbox is created. This allows for the detection of some additional # network being added to the existing network namespace, after the # sandbox has been created. # (default: disabled) #enable_netmon = true # Specify the path to the netmon binary. path = "/usr/libexec/kata-containers/kata-netmon" # If enabled, netmon messages will be sent to the system log # (default: disabled) #enable_debug = true [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 # If enabled, the runtime will create opentracing.io traces and spans. # (See https://www.jaegertracing.io/docs/getting-started). # (default: disabled) #enable_tracing = true # 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 `enable_netmon` # `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. # If you are using docker, `disable_new_netns` only works with `docker run --net=none` # (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://godoc.org/github.com/kata-containers/runtime/virtcontainers#ContainerType sandbox_cgroup_only=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=[] ``` --- # KSM throttler ## version Output of "`/usr/libexec/kata-ksm-throttler/kata-ksm-throttler --version`": ``` kata-ksm-throttler version 1.11.1-60526f8 ``` Output of "`/usr/lib/systemd/system/kata-ksm-throttler.service --version`": ``` /bin/kata-collect-data.sh: line 178: /usr/lib/systemd/system/kata-ksm-throttler.service: Permission denied ``` ## systemd service # Image details ```yaml --- osbuilder: url: "https://github.com/kata-containers/osbuilder" version: "unknown" rootfs-creation-time: "2020-06-09T05:36:02.623014927+0000Z" description: "osbuilder rootfs" file-format-version: "0.0.2" architecture: "x86_64" base-distro: name: "Clear" version: "33320" packages: default: - "chrony" - "iptables-bin" - "kmod-bin" - "libudev0-shim" - "systemd" - "util-linux-bin" extra: agent: url: "https://github.com/kata-containers/agent" name: "kata-agent" version: "1.11.1-34a000d0cf10c2e6bf1b46cd2bbfaa911de62679" agent-is-init-daemon: "no" ``` --- # Initrd details No initrd --- # Logfiles ## Runtime logs No recent runtime problems found in system journal. ## Proxy logs No recent proxy problems found in system journal. ## Shim logs No recent shim problems found in system journal. ## Throttler logs No recent throttler problems found in system journal. --- # Container manager details Have `docker` ## Docker Output of "`docker version`": ``` Client: Version: 18.06.3-ce API version: 1.38 Go version: go1.10.3 Git commit: d7080c1 Built: Wed Feb 20 02:26:51 2019 OS/Arch: linux/amd64 Experimental: false Server: Engine: Version: 18.06.3-ce API version: 1.38 (minimum version 1.12) Go version: go1.10.3 Git commit: d7080c1 Built: Wed Feb 20 02:28:17 2019 OS/Arch: linux/amd64 Experimental: false ``` Output of "`docker info`": ``` Containers: 10 Running: 7 Paused: 0 Stopped: 3 Images: 15 Server Version: 18.06.3-ce Storage Driver: overlay2 Backing Filesystem: xfs Supports d_type: true Native Overlay Diff: true Logging Driver: json-file Cgroup Driver: cgroupfs Plugins: Volume: local Network: bridge host macvlan null overlay Log: awslogs fluentd gcplogs gelf journald json-file logentries splunk syslog Swarm: inactive Runtimes: runc Default Runtime: runc Init Binary: docker-init containerd version: 468a545b9edcd5932818eb9de8e72413e616e86e runc version: a592beb5bc4c4092b1b1bac971afed27687340c5 init version: fec3683 Security Options: seccomp Profile: default Kernel Version: 5.2.20-2.el7.cp Operating System: CentOS Linux 7 (Core) OSType: linux Architecture: x86_64 CPUs: 64 Total Memory: 376.5GiB Name: host.domain.com ID: QDFG:3RFO:J2T7:MQ5G:6HSB:F6A6:BWLH:AZKV:6SD4:XE7L:FMJD:ZFUS Docker Root Dir: /mnt/storage/docker Debug Mode (client): false Debug Mode (server): false Registry: https://index.docker.io/v1/ Labels: Experimental: false Insecure Registries: 127.0.0.0/8 Live Restore Enabled: true WARNING: bridge-nf-call-ip6tables is disabled ``` Output of "`systemctl show docker`": ``` Type=notify Restart=on-failure NotifyAccess=main RestartUSec=100ms TimeoutStartUSec=0 TimeoutStopUSec=1min 30s WatchdogUSec=0 WatchdogTimestamp=Wed 2020-06-17 10:07:00 CEST WatchdogTimestampMonotonic=47640512 StartLimitInterval=60000000 StartLimitBurst=3 StartLimitAction=none FailureAction=none PermissionsStartOnly=no RootDirectoryStartOnly=no RemainAfterExit=no GuessMainPID=yes MainPID=5457 ControlPID=0 FileDescriptorStoreMax=0 StatusErrno=0 Result=success ExecMainStartTimestamp=Wed 2020-06-17 10:06:45 CEST ExecMainStartTimestampMonotonic=33376547 ExecMainExitTimestampMonotonic=0 ExecMainPID=5457 ExecMainCode=0 ExecMainStatus=0 ExecStart={ path=/usr/bin/dockerd ; argv[]=/usr/bin/dockerd ; ignore_errors=no ; start_time=[Wed 2020-06-17 10:06:45 CEST] ; stop_time=[n/a] ; pid=5457 ; 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 } Slice=system.slice ControlGroup=/system.slice/docker.service MemoryCurrent=732512256 TasksCurrent=218 Delegate=yes CPUAccounting=no CPUShares=18446744073709551615 StartupCPUShares=18446744073709551615 CPUQuotaPerSecUSec=infinity BlockIOAccounting=no BlockIOWeight=18446744073709551615 StartupBlockIOWeight=18446744073709551615 MemoryAccounting=no MemoryLimit=18446744073709551615 DevicePolicy=auto TasksAccounting=no TasksMax=18446744073709551615 UMask=0022 LimitCPU=18446744073709551615 LimitFSIZE=18446744073709551615 LimitDATA=18446744073709551615 LimitSTACK=18446744073709551615 LimitCORE=18446744073709551615 LimitRSS=18446744073709551615 LimitNOFILE=18446744073709551615 LimitAS=18446744073709551615 LimitNPROC=18446744073709551615 LimitMEMLOCK=65536 LimitLOCKS=18446744073709551615 LimitSIGPENDING=1542224 LimitMSGQUEUE=819200 LimitNICE=0 LimitRTPRIO=0 LimitRTTIME=18446744073709551615 OOMScoreAdjust=0 Nice=0 IOScheduling=0 CPUSchedulingPolicy=0 CPUSchedulingPriority=0 TimerSlackNSec=50000 CPUSchedulingResetOnFork=no NonBlocking=no StandardInput=null StandardOutput=journal StandardError=inherit TTYReset=no TTYVHangup=no TTYVTDisallocate=no SyslogPriority=30 SyslogLevelPrefix=yes SecureBits=0 CapabilityBoundingSet=18446744073709551615 AmbientCapabilities=0 MountFlags=0 PrivateTmp=no PrivateNetwork=no PrivateDevices=no ProtectHome=no ProtectSystem=no SameProcessGroup=no IgnoreSIGPIPE=yes NoNewPrivileges=no SystemCallErrorNumber=0 RuntimeDirectoryMode=0755 KillMode=process KillSignal=15 SendSIGKILL=yes SendSIGHUP=no Id=docker.service Names=docker.service Requires=-.mount basic.target mnt-storage.mount system.slice Wants=network-online.target WantedBy=multi-user.target Conflicts=shutdown.target Before=multi-user.target shutdown.target After=firewalld.service systemd-journald.socket basic.target mnt-storage.mount -.mount network-online.target system.slice RequiresMountsFor=/mnt/storage Documentation=https://docs.docker.com Description=Docker Application Container Engine LoadState=loaded ActiveState=active SubState=running FragmentPath=/usr/lib/systemd/system/docker.service UnitFileState=enabled UnitFilePreset=disabled InactiveExitTimestamp=Wed 2020-06-17 10:06:45 CEST InactiveExitTimestampMonotonic=33376620 ActiveEnterTimestamp=Wed 2020-06-17 10:07:00 CEST ActiveEnterTimestampMonotonic=47640803 ActiveExitTimestampMonotonic=0 InactiveEnterTimestampMonotonic=0 CanStart=yes CanStop=yes CanReload=yes CanIsolate=no StopWhenUnneeded=no RefuseManualStart=no RefuseManualStop=no AllowIsolate=no DefaultDependencies=yes OnFailureJobMode=replace IgnoreOnIsolate=no IgnoreOnSnapshot=no NeedDaemonReload=no JobTimeoutUSec=0 JobTimeoutAction=none ConditionResult=yes AssertResult=yes ConditionTimestamp=Wed 2020-06-17 10:06:45 CEST ConditionTimestampMonotonic=33374329 AssertTimestamp=Wed 2020-06-17 10:06:45 CEST AssertTimestampMonotonic=33374330 Transient=no ``` No `kubectl` No `crio` No `containerd` --- # Packages No `dpkg` Have `rpm` Output of "`rpm -qa|egrep "(cc-oci-runtimecc-runtimerunv|kata-proxy|kata-runtime|kata-shim|kata-ksm-throttler|kata-containers-image|linux-container|qemu-)"`": ``` kata-proxy-1.11.1-5.1.x86_64 kata-shim-1.11.1-5.1.x86_64 kata-linux-container-5.4.32.73-62.2.x86_64 qemu-vanilla-4.1.1+git.99c5874a9b-50.2.x86_64 qemu-vanilla-data-4.1.1+git.99c5874a9b-50.2.x86_64 kata-proxy-bin-1.11.1-5.1.x86_64 kata-ksm-throttler-1.11.1-5.1.x86_64 kata-shim-bin-1.11.1-5.1.x86_64 kata-runtime-1.11.1-5.1.x86_64 qemu-vanilla-bin-4.1.1+git.99c5874a9b-50.2.x86_64 kata-containers-image-1.11.1-5.1.x86_64 kata-linux-container-debug-5.4.32.73-5.1.x86_64 ``` ---

devimc commented 4 years ago

@mrlhansen thanks, unfortunately logs section is empty, would you mind enabling full debug and try again ? thanks

mrlhansen commented 4 years ago

@devimc For some reason, even after enabling debugging, the collect script did not have any information in the log files section. Instead I manually extracted some information from the system log on the machine. If you look in the attached file, starting on line 1085, you see that the guest kernel crashes.

failed.txt

devimc commented 4 years ago

thanks @mrlhansen - could you apply the following patch to the runtime and try again?

diff --git a/virtcontainers/sandbox.go b/virtcontainers/sandbox.go
index 4db46657..29b043b0 100644
--- a/virtcontainers/sandbox.go
+++ b/virtcontainers/sandbox.go
@@ -1824,6 +1824,21 @@ func (s *Sandbox) updateResources() error {
    // Add default / rsvd memory for sandbox.
    sandboxMemoryByte += int64(s.hypervisor.hypervisorConfig().MemorySize) << utils.MibToBytesShift

+   // Update Memory
+   s.Logger().WithField("memory-sandbox-size-byte", sandboxMemoryByte).Debugf("Request to hypervisor to update memory")
+   newMemory, updatedMemoryDevice, err := s.hypervisor.resizeMemory(uint32(sandboxMemoryByte>>utils.MibToBytesShift), s.state.GuestMemoryBlockSizeMB, s.state.GuestMemoryHotplugProbe)
+   if err != nil {
+       return err
+   }
+   s.Logger().Debugf("Sandbox memory size: %d MB", newMemory)
+   if s.state.GuestMemoryHotplugProbe && updatedMemoryDevice.addr != 0 {
+       // notify the guest kernel about memory hot-add event, before onlining them
+       s.Logger().Debugf("notify guest kernel memory hot-add event via probe interface, memory device located at 0x%x", updatedMemoryDevice.addr)
+       if err := s.agent.memHotplugByProbe(updatedMemoryDevice.addr, uint32(updatedMemoryDevice.sizeMB), s.state.GuestMemoryBlockSizeMB); err != nil {
+           return err
+       }
+   }
+
    // Update VCPUs
    s.Logger().WithField("cpus-sandbox", sandboxVCPUs).Debugf("Request to hypervisor to update vCPUs")
    oldCPUs, newCPUs, err := s.hypervisor.resizeVCPUs(sandboxVCPUs)
@@ -1840,20 +1855,6 @@ func (s *Sandbox) updateResources() error {
    }
    s.Logger().Debugf("Sandbox CPUs: %d", newCPUs)

-   // Update Memory
-   s.Logger().WithField("memory-sandbox-size-byte", sandboxMemoryByte).Debugf("Request to hypervisor to update memory")
-   newMemory, updatedMemoryDevice, err := s.hypervisor.resizeMemory(uint32(sandboxMemoryByte>>utils.MibToBytesShift), s.state.GuestMemoryBlockSizeMB, s.state.GuestMemoryHotplugProbe)
-   if err != nil {
-       return err
-   }
-   s.Logger().Debugf("Sandbox memory size: %d MB", newMemory)
-   if s.state.GuestMemoryHotplugProbe && updatedMemoryDevice.addr != 0 {
-       // notify the guest kernel about memory hot-add event, before onlining them
-       s.Logger().Debugf("notify guest kernel memory hot-add event via probe interface, memory device located at 0x%x", updatedMemoryDevice.addr)
-       if err := s.agent.memHotplugByProbe(updatedMemoryDevice.addr, uint32(updatedMemoryDevice.sizeMB), s.state.GuestMemoryBlockSizeMB); err != nil {
-           return err
-       }
-   }
    if err := s.agent.onlineCPUMem(0, false); err != nil {
        return err
    }
mrlhansen commented 4 years ago

@devimc Thanks a lot for looking into this. I tried recompiling the runtime and replacing the kata-runtime executable. However, the error is still the same. In the system log it now looks like the memory is enabled before the cores, so I believe the patch is correctly included. In the new log file attached here, the kernel starts crashing on line 740.

failed2.txt

devimc commented 4 years ago

@mrlhansen thanks for testing this patch. I thinking that this bug is in QEMU 4.1, could you try with QEMU 5?

cc @stefanha @dagrh

mrlhansen commented 4 years ago

@devimc I can try with QEMU 5. Do you have a fork I should use, or just the vanilla version from the QEMU website? Do I need some special configure options?

devimc commented 4 years ago

@mrlhansen please use the vanilla version, but use the following script to configure it https://raw.githubusercontent.com/kata-containers/packaging/master/scripts/configure-hypervisor.sh

bash -f configure-hypervisor.sh qemu | xargs ./configure

dagrh commented 4 years ago

I don't think there's anything magic in qemu at the 128 G line (128G=37bits); upstream normally hits some fun at 1T and on desktop CPUs you hit problems at 512G (39b phys address) - but you're on an E5 so that shouldn't be a problem. That page allocation failure is fun - you're running out of memory trying to hot add memory :-)

davidhildenbrand commented 4 years ago

So, one theoretical issue is, if you're hotplugging fairly huge DIMMs, you might not have enough free memory to even hotplug the first DIMM properly.

E.g., the memmap needs 1.56% of memory. In addition to that, you need pagetables. Assume you have 1G of free memory, the biggest DIMM you can plug is in the range of <65G. So if you have <2G you can never hotplug a single DIMM bigger than 130G. I guess this is what you see here.

To workaround that, you can hotplug smaller, individual DIMMs (which you should online immediately automatically within the guest), or increase the initial memory size.

Also note, the bigger your first DIMM, the more DMA memory you might lose ...

mrlhansen commented 4 years ago

@davidhildenbrand @dagrh @devimc

Increasing the initial memory from 2G to 4G solved the problem. Thanks you all so much! Now that I know the reason and solution, it's so simple and obvious.

Just for future reference. Is is possible to configure kata to hotplug memory in smaller chunks spread over more DIMMs? Keeping in mind that this is running via Kubernetes.

devimc commented 4 years ago

@mrlhansen great! good to hear you found a solution

Just for future reference. Is is possible to configure kata to hotplug memory in smaller chunks spread over more DIMMs? Keeping in mind that this is running via Kubernetes.

yes, that's possible, the problem is that there is a limit in the number of memory slots, so if you have 10 slots and want to hot-add 120G, then the size for the chucks could be 12G, but this means you cannot hot-add more memory later because there are no more slots free.

davidhildenbrand commented 4 years ago

@mrlhansen great! good to hear you found a solution

Just for future reference. Is is possible to configure kata to hotplug memory in smaller chunks spread over more DIMMs? Keeping in mind that this is running via Kubernetes.

yes, that's possible, the problem is that there is a limit in the number of memory slots, so if you have 10 slots and want to hot-add 120G, then the size for the chucks could be 12G, but this means you cannot hot-add more memory later because there are no more slots free.

Just a note (especially, because what I'll be mentioning is still under heavy, early development), the problems you are running/mentioning in here (DIMM too big for initial memory, limited number of slots, not onlining memory can result in issues in the guest) are some of the things virtio-mem tries to solve.

mrlhansen commented 4 years ago

Thanks again for the help. Feel free to close the issue again.

devimc commented 4 years ago

closing issue.. thanks @mrlhansen