Fix conmon attach socket buffer size

The conmon buffer size is 8192, however the attach socket needs two extra
bytes. The first byte of each message will be the STREAM type. The last
byte is a null byte. So when we want to read 8192 message bytes we need
to read 8193 bytes since the first one is special.
check https://github.com/containers/conmon/blob/1ef246896b4f6566964ed861b98cd32d0e7bf7a2/src/ctr_stdio.c#L101-L107

This problem can be seen in podman-remote run/exec when it prints output
with 8192 or more bytes. The output will miss the 8192 byte.

Fixes #11496

Signed-off-by: Paul Holzinger <[email protected]>

libpod/oci_conmon_linux.go

@@ -46,7 +46,9 @@ import (
 const (
 	// This is Conmon's STDIO_BUF_SIZE. I don't believe we have access to it
 	// directly from the Go code, so const it here
-	bufferSize = conmonConfig.BufSize
+	// Important: The conmon attach socket uses an extra byte at the beginning of each
+	// message to specify the STREAM so we have to increase the buffer size by one
+	bufferSize = conmonConfig.BufSize + 1
 )
 
 // ConmonOCIRuntime is an OCI runtime managed by Conmon.

test/system/075-exec.bats

@@ -101,4 +101,32 @@ load helpers
     run_podman rm $cid
 }
 
+# #11496: podman-remote loses output
+@test "podman exec/run - missing output" {
+    local bigfile=${PODMAN_TMPDIR}/bigfile
+    local newfile=${PODMAN_TMPDIR}/newfile
+    # create a big file, bigger than the 8K buffer size
+    base64 /dev/urandom | head -c 20K > $bigfile
+
+    run_podman run --rm -v $bigfile:/tmp/test:Z $IMAGE cat /tmp/test
+    printf "%s" "$output" > $newfile
+    # use cmp to compare the files, this is very helpful since it will
+    # tell us the first wrong byte in case this fails
+    run cmp $bigfile $newfile
+    is "$output" "" "run output is identical with the file"
+
+    run_podman run -d --stop-timeout 0 -v $bigfile:/tmp/test:Z $IMAGE sleep inf
+    cid="$output"
+
+    run_podman exec $cid cat /tmp/test
+    printf "%s" "$output" > $newfile
+    # use cmp to compare the files, this is very helpful since it will
+    # tell us the first wrong byte in case this fails
+    run cmp $bigfile $newfile
+    is "$output" "" "exec output is identical with the file"
+
+    # Clean up
+    run_podman rm -f $cid
+}
+
 # vim: filetype=sh