Add global flags -profile-{block,cpu,heap,mutex} and -trace which
enable collection of the specified profile for the entire duration of a
container execution. This provides a way to definitively start profiling
before that application starts, rather than attempting to race with an
out-of-band `runsc debug`.
Note that only the main boot process is profiled.
This exposed a bug in Task.traceExecEvent: a crash when tracing and
-race are enabled. traceExecEvent is called off of the task goroutine,
but uses the Task as a context, which is a violation of the Task
contract. Switching to the AsyncContext fixes the issue.
Fixes#220
Set stdio ownership based on the container's user to ensure the
user can open/read/write to/from stdios.
1. stdios in the host are changed to have the owner be the same
uid/gid of the process running the sandbox. This ensures that the
sandbox has full control over it.
2. stdios owner owner inside the sandbox is changed to match the
container's user to give access inside the container and make it
behave the same as runc.
Fixes#6180
PiperOrigin-RevId: 384347009
According to the OCI spec Mount.Type is an optional field and it
defaults to "bind" when any of "bind" or "rbind" is included in
Mount.Options.
Also fix the shim to remove bind/rbind from options when mount is
converted from bind to tmpfs inside the Sentry.
Fixes#2330Fixes#3274
PiperOrigin-RevId: 371996891
The syscall package has been deprecated in favor of golang.org/x/sys.
Note that syscall is still used in some places because the following don't seem
to have an equivalent in unix package:
- syscall.SysProcIDMap
- syscall.Credential
Updates #214
PiperOrigin-RevId: 361381490
Previously, loader.signalProcess was inconsitently using both root and
container's PID namespace to find the process. It used root namespace
for the exec'd process and container's PID namespace for other processes.
This fixes the code to use the root PID namespace across the board, which
is the same PID reported in `runsc ps` (or soon will after
https://github.com/google/gvisor/pull/5519).
PiperOrigin-RevId: 358836297
This allows to find all containers inside a sandbox more efficiently.
This operation is required every time a container starts and stops,
and previously required loading *all* container state files to check
whether the container belonged to the sandbox.
Apert from being inneficient, it has caused problems when state files
are stale or corrupt, causing inavalability to create any container.
Also adjust commands `list` and `debug` to skip over files that fail
to load.
Resolves#5052
PiperOrigin-RevId: 348050637
When OOM score adjustment needs to be set, all the containers need to be
loaded to find all containers that belong to the sandbox. However, each
load signals the container to ensure it is still alive. OOM score
adjustment is set during creation and deletion of every container, generating
a flood of signals to all containers. The fix removes the signal check
when it's not needed.
There is also a race fetching OOM score adjustment value from the parent when
the sandbox exits at the same time (the time it took to signal containers above
made this window quite large). The fix is to store the original value
in the sandbox state file and use it when the value needs to be restored.
Also add more logging and made the existing ones more consistent to help with
debugging.
PiperOrigin-RevId: 340940799
There were a few problems with cgroups:
- cleanup loop what breaking too early
- parse of /proc/[pid]/cgroups was skipping "name=systemd"
because "name=" was not being removed from name.
- When no limits are specified, fillFromAncestor was not being
called, causing a failure to set cpuset.mems
Updates #4536
PiperOrigin-RevId: 337947356
ptrace was always selected as a dimension before, but not
anymore. Some tests were specifying "overlay" expecting that
to be in addition to the default.
PiperOrigin-RevId: 332004111
The bazel server was being started as the wrong user, leading to issues
where the container would suddenly exit during a build.
We can also simplify the waiting logic by starting the container in two
separate steps: those that must complete first, then the asynchronous bit.
PiperOrigin-RevId: 323391161
Run vs. exec, VFS1 vs. VFS2 were executable lookup were
slightly different from each other. Combine them all
into the same logic.
PiperOrigin-RevId: 315426443
- Add /tmp handling
- Apply mount options
- Enable more container_test tests
- Forward signals to child process when test respaws process
to run as root inside namespace.
Updates #1487
PiperOrigin-RevId: 314263281
Linux 4.18 and later make reads and writes coherent between pre-copy-up and
post-copy-up FDs representing the same file on an overlay filesystem. However,
memory mappings remain incoherent:
- Documentation/filesystems/overlayfs.rst, "Non-standard behavior": "If a file
residing on a lower layer is opened for read-only and then memory mapped with
MAP_SHARED, then subsequent changes to the file are not reflected in the
memory mapping."
- fs/overlay/file.c:ovl_mmap() passes through to the underlying FD without any
management of coherence in the overlay.
- Experimentally on Linux 5.2:
```
$ cat mmap_cat_page.c
#include <err.h>
#include <fcntl.h>
#include <stdio.h>
#include <string.h>
#include <sys/mman.h>
#include <unistd.h>
int main(int argc, char **argv) {
if (argc < 2) {
errx(1, "syntax: %s [FILE]", argv[0]);
}
const int fd = open(argv[1], O_RDONLY);
if (fd < 0) {
err(1, "open(%s)", argv[1]);
}
const size_t page_size = sysconf(_SC_PAGE_SIZE);
void* page = mmap(NULL, page_size, PROT_READ, MAP_SHARED, fd, 0);
if (page == MAP_FAILED) {
err(1, "mmap");
}
for (;;) {
write(1, page, strnlen(page, page_size));
if (getc(stdin) == EOF) {
break;
}
}
return 0;
}
$ gcc -O2 -o mmap_cat_page mmap_cat_page.c
$ mkdir lowerdir upperdir workdir overlaydir
$ echo old > lowerdir/file
$ sudo mount -t overlay -o "lowerdir=lowerdir,upperdir=upperdir,workdir=workdir" none overlaydir
$ ./mmap_cat_page overlaydir/file
old
^Z
[1]+ Stopped ./mmap_cat_page overlaydir/file
$ echo new > overlaydir/file
$ cat overlaydir/file
new
$ fg
./mmap_cat_page overlaydir/file
old
```
Therefore, while the VFS1 gofer client's behavior of reopening read FDs is only
necessary pre-4.18, replacing existing memory mappings (in both sentry and
application address spaces) with mappings of the new FD is required regardless
of kernel version, and this latter behavior is common to both VFS1 and VFS2.
Re-document accordingly, and change the runsc flag to enabled by default.
New test:
- Before this CL: https://source.cloud.google.com/results/invocations/5b222d2c-e918-4bae-afc4-407f5bac509b
- After this CL: https://source.cloud.google.com/results/invocations/f28c747e-d89c-4d8c-a461-602b33e71aab
PiperOrigin-RevId: 311361267
Several tests are passing after getting TestAppExitStatus (run /bin/true)
changes. Make versions that run via VFS2 so that we know what is and isn't
working.
In addition, fix bug in VFSFile ReadFull. For the TestExePath test in
container_test.go, the case "unmasked" will return 0 bytes read with no
EOF err, causing the ReadFull call to spin.
PiperOrigin-RevId: 308428126
This change adds a layer of abstraction around the internal Docker APIs,
and eliminates all direct dependencies on Dockerfiles in the infrastructure.
A subsequent change will automated the generation of local images (with
efficient caching). Note that this change drops the use of bazel container
rules, as that experiment does not seem to be viable.
PiperOrigin-RevId: 308095430
Included:
- loader_test.go RunTest and TestStartSignal VFS2
- container_test.go TestAppExitStatus on VFS2
- experimental flag added to runsc to turn on VFS2
Note: shared mounts are not yet supported.
PiperOrigin-RevId: 307070753
When the sandbox runs in attached more, e.g. runsc do, runsc run, the
sandbox lifetime is controlled by the parent process. This wasn't working
in all cases because PR_GET_PDEATHSIG doesn't propagate through execve
when the process changes uid/gid. So it was getting dropped when the
sandbox execve's to change to user nobody.
PiperOrigin-RevId: 300601247
TestMultiContainerKillAll timed out under --race. Without logging,
we cannot tell if the process list is still increasing, but slowly,
or is stuck.
PiperOrigin-RevId: 297158834
Sometimes we get this error under TSAN:
"""
error getting process data from container: connecting to control server at PID
XXXX: connection refused
"""
The theory is that the top "sleep 20" was too short for TSAN, and the container
already exited, so we get connected refused. This commit changes the test to
let container signaling it's running by touching a file repeatedly forever
during the test.
PiperOrigin-RevId: 293710957
* Rename syncutil to sync.
* Add aliases to sync types.
* Replace existing usage of standard library sync package.
This will make it easier to swap out synchronization primitives. For example,
this will allow us to use primitives from github.com/sasha-s/go-deadlock to
check for lock ordering violations.
Updates #1472
PiperOrigin-RevId: 289033387
runsc debug --ps list all processes with all threads. This option is added to
the debug command but not to the ps command, because it is going to be used for
debug purposes and we want to add any useful information without thinking about
backward compatibility.
This will help to investigate syzkaller issues.
PiperOrigin-RevId: 285013668
Threadgroups already know their TTY (if they have one), which now contains the
TTY Index, and is returned in the Processes() call.
PiperOrigin-RevId: 284263850
container.startContainers() cannot be called twice in a test
(e.g. TestMultiContainerLoadSandbox) because the cleanup
function deletes the rootDir, together with information from
all other containers that may exist.
PiperOrigin-RevId: 276591806
Linux kernel before 4.19 doesn't implement a feature that updates
open FD after a file is open for write (and is copied to the upper
layer). Already open FD will continue to read the old file content
until they are reopened. This is especially problematic for gVisor
because it caches open files.
Flag was added to force readonly files to be reopenned when the
same file is open for write. This is only needed if using kernels
prior to 4.19.
Closes#1006
It's difficult to really test this because we never run on tests
on older kernels. I'm adding a test in GKE which uses kernels
with the overlayfs problem for 1.14 and lower.
PiperOrigin-RevId: 275115289
'docker exec' was getting CAP_NET_RAW even when --net-raw=false
because it was not filtered out from when copying container's
capabilities.
PiperOrigin-RevId: 272260451
The simple test script has gotten out of control. Shard this script into
different pieces and attempt to impose order on overall test structure. This
change helps lay some of the foundations for future improvements.
* The runsc/test directories are moved into just test/.
* The runsc/test/testutil package is split into logical pieces.
* The scripts/ directory contains new top-level targets.
* Each test is now responsible for building targets it requires.
* The install functionality is moved into `runsc` itself for simplicity.
* The existing kokoro run_tests.sh file now just calls all (can be split).
After this change is merged, I will create multiple distinct workflows for
Kokoro, one for each of the scripts currently targeted by `run_tests.sh` today,
which should dramatically reduce the time-to-run for the Kokoro tests, and
provides a better foundation for further improvements to the infrastructure.
PiperOrigin-RevId: 267081397
Michael Pratt
Chong Cai
Fabricio Voznika
Rahat Mahmood
Ayush Ranjan
Ting-Yu Wang
Howard Zhang
Adin Scannell
Jamie Liu
gVisor bot
moricho
Zach Koopmans
Andrei Vagin
Ian Gudger
Nicolas Lacasse