Release v2.1.16

This is to fix this issue when starting the (frozen) application on Windows:

```
  File "C:\Python36-x64\lib\site-packages\jsonschema\validators.py", line 505, in <module>
  File "C:\Python36-x64\lib\site-packages\jsonschema\_utils.py", line 57, in load_schema
  File "C:\Python36-x64\lib\pkgutil.py", line 634, in get_data
OSError: [Errno 34] Result too large: 'jsonschema\\schemas\\draft6.json'
```

CHANGELOG

@@ -1,5 +1,51 @@
 # Change Log
 
+## 2.1.16 15/04/2019
+
+* Fix broken embedded console for Ethernet switch. Fixes #1574
+* Remove old unused argument option. Fixes #1569
+
+## 2.1.15 21/03/2019
+
+* Fix IOU symlink issue on remote servers.
+* Fix vcpus configuration for GNS3 VM on VMware. Ref #2738.
+
+## 2.1.14 27/02/2019
+
+* Fix issue when setting cpuid.corespersocket for the GNS3 VM. Fixes https://github.com/GNS3/gns3-gui/issues/2723
+* Bump ACPI Shutdown Timeout to 120 seconds. Ref #1536
+
+## 2.1.13 26/02/2019
+
+* Force jsonschema dependency to 2.6.0
+* Less aggressive connections to uBridge. Ref #1289
+* Fix topology images (Pictures) disappearing from projects. Fixes #1514.
+* Reset MAC addresses when duplicating a project. Fixes #1522
+* Fix API call to create a node from an appliance doesn't return the new node data. Fixes #1527
+* Detect invalid environment variable and send a warning when creating a Docker node. Ref #2683
+* Do not export/import symlinks for projects. Fixes #2699
+* Fix symlink not being created for duplicated IOU devices. Fixes https://github.com/GNS3/gns3-gui/issues/2699
+* Configure coresPerSocket value in VMX file for the GNS3 VM. Fixes https://github.com/GNS3/gns3-gui/issues/2688
+* Count logical CPUs to detect if the number of vCPUs is too high when configuring the GNS3 VM. Fixes #2688.
+* Add explicit error when trying to pull a Docker image from Docker Hub without Internet access. Fixes #1506.
+* Fixes double display output in GRUB in QEMU v3.1. Fixes #1516.
+
+## 2.1.12 23/01/2019
+
+* Tune how to get the size of SVG images. Ref https://github.com/GNS3/gns3-gui/issues/2674.
+* Automatically create a symbolic link to the IOU image in the IOU working directory. Fixes #1484
+* Fix link pause/filters only work for the first interface of Docker containers. Fixes #1482
+* Telnet console resize support for Docker VM.
+* Fix _fix_permissions() garbles permissions in Docker VM. Ref #1428
+* Fix "None is not of type 'integer'" when opening project containing a Qemu VM. Fixes #2610.
+* Only require Xtigervnc or Xvfb+x11vnc for Docker with vnc console. Ref #1438
+* Support tigervnc in Docker VM. Ref #1438
+* Update minimum VIX version requirements for VMware. Ref #1415.
+
+## 2.1.11 28/09/2018
+
+* Catch some exceptions.
+
 ## 2.1.10 15/09/2018
 
 * Include locale information and GNS3 VM version in crash reports.
@@ -677,7 +723,7 @@
 * Fix naming of IOU serial interfaces
 * Improve timeout management
 * When exporting debug information export GNS3 VM vmx content
-* /debug for exporting debug informations
+* /debug for exporting debug information
 * Raise error if using a non linked clone VM twice
 * Fix a possible deadlock at exit
 * Fix import of some old dynamips topologies

docs/api/v2/compute/server/debug.rst

@@ -5,7 +5,7 @@
 
 GET /v2/compute/debug
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-Return debug informations about the compute
+Return debug information about the compute
 
 Response status codes
 **********************

gns3server/compute/base_manager.py

@@ -276,7 +276,7 @@ class BaseManager:
         destination_dir = destination_node.working_dir
         try:
             shutil.rmtree(destination_dir)
-            shutil.copytree(source_node.working_dir, destination_dir)
+            shutil.copytree(source_node.working_dir, destination_dir, symlinks=True, ignore_dangling_symlinks=True)
         except OSError as e:
             raise aiohttp.web.HTTPConflict(text="Can't duplicate node data: {}".format(e))
 

gns3server/compute/docker/__init__.py

@@ -200,7 +200,10 @@ class Docker(BaseManager):
 
         if progress_callback:
             progress_callback("Pulling '{}' from docker hub".format(image))
-        response = yield from self.http_query("POST", "images/create", params={"fromImage": image}, timeout=None)
+        try:
+            response = yield from self.http_query("POST", "images/create", params={"fromImage": image}, timeout=None)
+        except DockerError as e:
+            raise DockerError("Could not pull the '{}' image from Docker Hub, please check your Internet connection (original error: {})".format(image, e))
         # The pull api will stream status via an HTTP JSON stream
         content = ""
         while True:

gns3server/compute/docker/docker_vm.py

@@ -24,6 +24,7 @@ import shutil
 import psutil
 import shlex
 import aiohttp
+import subprocess
 import os
 
 from gns3server.utils.asyncio.telnet_server import AsyncioTelnetServer
@@ -82,12 +83,14 @@ class DockerVM(BaseNode):
         self._ethernet_adapters = []
         self._temporary_directory = None
         self._telnet_servers = []
-        self._x11vnc_process = None
+        self._vnc_process = None
+        self._xvfb_process = None
         self._console_resolution = console_resolution
         self._console_http_path = console_http_path
         self._console_http_port = console_http_port
         self._console_websocket = None
         self._extra_hosts = extra_hosts
+        self._permissions_fixed = False
 
         self._volumes = []
         # Keep a list of created bridge
@@ -344,6 +347,9 @@ class DockerVM(BaseNode):
         if self._environment:
             for e in self._environment.strip().split("\n"):
                 e = e.strip()
+                if e.split("=")[0] == "":
+                    self.project.emit("log.warning", {"message": "{} has invalid environment variable: {}".format(self.name, e)})
+                    continue
                 if not e.startswith("GNS3_"):
                     formatted = self._format_env(variables, e)
                     params["Env"].append(formatted)
@@ -439,12 +445,14 @@ class DockerVM(BaseNode):
 
             if self.console_type == "telnet":
                 yield from self._start_console()
+
             elif self.console_type == "http" or self.console_type == "https":
                 yield from self._start_http()
 
             if self.allocate_aux:
                 yield from self._start_aux()
 
+        self._permissions_fixed = False
         self.status = "started"
         log.info("Docker container '{name}' [{image}] started listen for {console_type} on {console}".format(name=self._name,
                                                                                                              image=self._image,
@@ -460,11 +468,10 @@ class DockerVM(BaseNode):
         # We can not use the API because docker doesn't expose a websocket api for exec
         # https://github.com/GNS3/gns3-gui/issues/1039
         try:
-            process = yield from asyncio.subprocess.create_subprocess_exec(
-                "docker", "exec", "-i", self._cid, "/gns3/bin/busybox", "script", "-qfc", "while true; do TERM=vt100 /gns3/bin/busybox sh; done", "/dev/null",
-                stdout=asyncio.subprocess.PIPE,
-                stderr=asyncio.subprocess.STDOUT,
-                stdin=asyncio.subprocess.PIPE)
+            process = yield from asyncio.subprocess.create_subprocess_exec("docker", "exec", "-i", self._cid, "/gns3/bin/busybox", "script", "-qfc", "while true; do TERM=vt100 /gns3/bin/busybox sh; done", "/dev/null",
+                                                                           stdout=asyncio.subprocess.PIPE,
+                                                                           stderr=asyncio.subprocess.STDOUT,
+                                                                           stdin=asyncio.subprocess.PIPE)
         except OSError as e:
             raise DockerError("Could not start auxiliary console process: {}".format(e))
         server = AsyncioTelnetServer(reader=process.stdout, writer=process.stdin, binary=True, echo=True)
@@ -472,7 +479,7 @@ class DockerVM(BaseNode):
             self._telnet_servers.append((yield from asyncio.start_server(server.run, self._manager.port_manager.console_host, self.aux)))
         except OSError as e:
             raise DockerError("Could not start Telnet server on socket {}:{}: {}".format(self._manager.port_manager.console_host, self.aux, e))
-        log.debug("Docker container '%s' started listen for auxilary telnet on %d", self.name, self.aux)
+        log.debug("Docker container '%s' started listen for auxiliary telnet on %d", self.name, self.aux)
 
     @asyncio.coroutine
     def _fix_permissions(self):
@@ -482,6 +489,7 @@ class DockerVM(BaseNode):
         """
 
         state = yield from self._get_container_state()
+        log.info("Docker container '{name}' fix ownership, state = {state}".format(name=self._name, state=state))
         if state == "stopped" or state == "exited":
             # We need to restart it to fix permissions
             yield from self.manager.query("POST", "containers/{}/start".format(self._cid))
@@ -509,6 +517,7 @@ class DockerVM(BaseNode):
             except OSError as e:
                 raise DockerError("Could not fix permissions for {}: {}".format(volume, e))
             yield from process.wait()
+            self._permissions_fixed = True
 
     @asyncio.coroutine
     def _start_vnc(self):
@@ -517,19 +526,46 @@ class DockerVM(BaseNode):
         """
 
         self._display = self._get_free_display_port()
-        if shutil.which("Xvfb") is None or shutil.which("x11vnc") is None:
-            raise DockerError("Please install Xvfb and x11vnc before using VNC support")
-        self._xvfb_process = yield from asyncio.create_subprocess_exec("Xvfb", "-nolisten", "tcp", ":{}".format(self._display), "-screen", "0", self._console_resolution + "x16")
-        # We pass a port for TCPV6 due to a crash in X11VNC if not here: https://github.com/GNS3/gns3-server/issues/569
-        self._x11vnc_process = yield from asyncio.create_subprocess_exec("x11vnc", "-forever", "-nopw", "-shared", "-geometry", self._console_resolution, "-display", "WAIT:{}".format(self._display), "-rfbport", str(self.console), "-rfbportv6", str(self.console), "-noncache", "-listen", self._manager.port_manager.console_host)
+        if not (shutil.which("Xtigervnc") or shutil.which("Xvfb") and shutil.which("x11vnc")):
+            raise DockerError("Please install tigervnc-standalone-server (recommended) or Xvfb + x11vnc before using VNC support")
+
+        if shutil.which("Xtigervnc"):
+            with open(os.path.join(self.working_dir, "vnc.log"), "w") as fd:
+                self._vnc_process = yield from asyncio.create_subprocess_exec("Xtigervnc",
+                                                                              "-geometry", self._console_resolution,
+                                                                              "-depth", "16",
+                                                                              "-interface", self._manager.port_manager.console_host,
+                                                                              "-rfbport", str(self.console),
+                                                                              "-AlwaysShared",
+                                                                              "-SecurityTypes", "None",
+                                                                              ":{}".format(self._display),
+                                                                              stdout=fd, stderr=subprocess.STDOUT)
+        else:
+            self._xvfb_process = yield from asyncio.create_subprocess_exec("Xvfb",
+                                                                           "-nolisten",
+                                                                           "tcp", ":{}".format(self._display),
+                                                                           "-screen", "0",
+                                                                           self._console_resolution + "x16")
+
+            # We pass a port for TCPV6 due to a crash in X11VNC if not here: https://github.com/GNS3/gns3-server/issues/569
+            with open(os.path.join(self.working_dir, "vnc.log"), "w") as fd:
+                self._vnc_process = yield from asyncio.create_subprocess_exec("x11vnc",
+                                                                              "-forever",
+                                                                              "-nopw",
+                                                                              "-shared",
+                                                                              "-geometry", self._console_resolution,
+                                                                              "-display", "WAIT:{}".format(self._display),
+                                                                              "-rfbport", str(self.console),
+                                                                              "-rfbportv6", str(self.console),
+                                                                              "-noncache",
+                                                                              "-listen", self._manager.port_manager.console_host,
+                                                                              stdout=fd, stderr=subprocess.STDOUT)
 
         x11_socket = os.path.join("/tmp/.X11-unix/", "X{}".format(self._display))
         yield from wait_for_file_creation(x11_socket)
+        #monitor_process(self._vnc_process, self._vnc_callback)
 
-        #monitor_process(self._xvfb_process, self._xvfb_callback)
-        #monitor_process(self._x11vnc_process, self._x11vnc_callback)
-
-    def _xvfb_callback(self, returncode):
+    def _vnc_callback(self, returncode):
         """
         Called when the process has stopped.
 
@@ -537,17 +573,7 @@ class DockerVM(BaseNode):
         """
 
         if returncode != 0:
-            self.project.emit("log.error", {"message": "The Xvfb process has stopped, return code: {}.".format(returncode)})
-
-    def _x11vnc_callback(self, returncode):
-        """
-        Called when the process has stopped.
-
-        :param returncode: Process returncode
-        """
-
-        if returncode != 0:
-            self.project.emit("log.error", {"message": "The x11vnc process has stopped, return code: {}.".format(returncode)})
+            self.project.emit("log.error", {"message": "The vnc process has stopped, return code: {}.".format(returncode)})
 
     @asyncio.coroutine
     def _start_http(self):
@@ -570,6 +596,19 @@ class DockerVM(BaseNode):
         ])
         self._telnet_servers.append((yield from asyncio.start_server(server.run, self._manager.port_manager.console_host, self.console)))
 
+    @asyncio.coroutine
+    def _window_size_changed_callback(self, columns, rows):
+        """
+        Called when the console window size has been changed.
+        (when naws is enabled in the Telnet server)
+
+        :param columns: number of columns
+        :param rows: number of rows
+        """
+
+        # resize the container TTY.
+        yield from self._manager.query("POST", "containers/{}/resize?h={}&w={}".format(self._cid, rows, columns))
+
     @asyncio.coroutine
     def _start_console(self):
         """
@@ -592,8 +631,7 @@ class DockerVM(BaseNode):
 
         output_stream = asyncio.StreamReader()
         input_stream = InputStream()
-
-        telnet = AsyncioTelnetServer(reader=output_stream, writer=input_stream, echo=True)
+        telnet = AsyncioTelnetServer(reader=output_stream, writer=input_stream, echo=True, naws=True, window_size_changed_callback=self._window_size_changed_callback)
         try:
             self._telnet_servers.append((yield from asyncio.start_server(telnet.run, self._manager.port_manager.console_host, self.console)))
         except OSError as e:
@@ -603,7 +641,6 @@ class DockerVM(BaseNode):
         input_stream.ws = self._console_websocket
 
         output_stream.feed_data(self.name.encode() + b" console is now available... Press RETURN to get started.\r\n")
-
         asyncio_ensure_future(self._read_console_output(self._console_websocket, output_stream))
 
     @asyncio.coroutine
@@ -678,14 +715,15 @@ class DockerVM(BaseNode):
             if state == "paused":
                 yield from self.unpause()
 
-            yield from self._fix_permissions()
+            if not self._permissions_fixed:
+                yield from self._fix_permissions()
+
             state = yield from self._get_container_state()
             if state != "stopped" or state != "exited":
                 # t=5 number of seconds to wait before killing the container
                 try:
                     yield from self.manager.query("POST", "containers/{}/stop".format(self._cid), params={"t": 5})
-                    log.info("Docker container '{name}' [{image}] stopped".format(
-                        name=self._name, image=self._image))
+                    log.info("Docker container '{name}' [{image}] stopped".format(name=self._name, image=self._image))
                 except DockerHttp304Error:
                     # Container is already stopped
                     pass
@@ -725,12 +763,13 @@ class DockerVM(BaseNode):
             if state == "paused" or state == "running":
                 yield from self.stop()
             if self.console_type == "vnc":
-                if self._x11vnc_process:
+                if self._vnc_process:
                     try:
-                        self._x11vnc_process.terminate()
-                        yield from self._x11vnc_process.wait()
+                        self._vnc_process.terminate()
+                        yield from self._vnc_process.wait()
                     except ProcessLookupError:
                         pass
+                if self._xvfb_process:
                     try:
                         self._xvfb_process.terminate()
                         yield from self._xvfb_process.wait()

gns3server/compute/iou/iou_vm.py

@@ -517,6 +517,17 @@ class IOUVM(BaseNode):
 
             if "IOURC" not in os.environ:
                 env["IOURC"] = iourc_path
+
+            # create a symbolic link to the image to avoid IOU error "failed code signing checks"
+            # on newer images, see https://github.com/GNS3/gns3-server/issues/1484
+            try:
+                symlink = os.path.join(self.working_dir, os.path.basename(self.path))
+                if os.path.islink(symlink):
+                    os.unlink(symlink)
+                os.symlink(self.path, symlink)
+            except OSError as e:
+                raise IOUError("Could not create symbolic link: {}".format(e))
+
             command = yield from self._build_command()
             try:
                 log.info("Starting IOU: {}".format(command))
@@ -651,6 +662,13 @@ class IOUVM(BaseNode):
                             pass
             self._iou_process = None
 
+        try:
+            symlink = os.path.join(self.working_dir, os.path.basename(self.path))
+            if os.path.islink(symlink):
+                os.unlink(symlink)
+        except OSError as e:
+            log.warning("Could not delete symbolic link: {}".format(e))
+
         self._started = False
         self.save_configs()
 

gns3server/compute/port_manager.py

@@ -23,7 +23,7 @@ import logging
 log = logging.getLogger(__name__)
 
 
-# This ports are disallowed by Chrome and Firefox to avoid issues, we skip them as well
+# These ports are disallowed by Chrome and Firefox to avoid issues, we skip them as well
 BANNED_PORTS = set((1, 7, 9, 11, 13, 15, 17, 19, 20, 21, 22, 23, 25, 37, 42, 43, 53, 77, 79, 87, 95, 101, 102, 103,
                     104, 109, 110, 111, 113, 115, 117, 119, 123, 135, 139, 143, 179, 389, 465, 512, 513, 514, 515, 526,
                     530, 531, 532, 540, 556, 563, 587, 601, 636, 993, 995, 2049, 3659, 4045, 6000, 6665, 6666, 6667,

gns3server/compute/project.py

@@ -415,7 +415,7 @@ class Project:
         """
 
         files = []
-        for dirpath, dirnames, filenames in os.walk(self.path):
+        for dirpath, dirnames, filenames in os.walk(self.path, followlinks=False):
             for filename in filenames:
                 if not filename.endswith(".ghost"):
                     path = os.path.relpath(dirpath, self.path)

gns3server/compute/qemu/__init__.py

@@ -186,7 +186,7 @@ class Qemu(BaseManager):
             return ""
         else:
             try:
-                output = yield from subprocess_check_output(qemu_path, "-version")
+                output = yield from subprocess_check_output(qemu_path, "-version", "-nographic")
                 match = re.search("version\s+([0-9a-z\-\.]+)", output)
                 if match:
                     version = match.group(1)

gns3server/compute/qemu/qemu_vm.py

@@ -974,7 +974,7 @@ class QemuVM(BaseNode):
                 try:
                     if self.acpi_shutdown:
                         yield from self._control_vm("system_powerdown")
-                        yield from gns3server.utils.asyncio.wait_for_process_termination(self._process, timeout=30)
+                        yield from gns3server.utils.asyncio.wait_for_process_termination(self._process, timeout=120)
                     else:
                         self._process.terminate()
                         yield from gns3server.utils.asyncio.wait_for_process_termination(self._process, timeout=3)
@@ -1601,18 +1601,19 @@ class QemuVM(BaseNode):
 
         return network_options
 
-    def _graphic(self):
+    @asyncio.coroutine
+    def _disable_graphics(self):
         """
-        Adds the correct graphic options depending of the OS
+        Disable graphics depending of the QEMU version
         """
 
-        if sys.platform.startswith("win"):
+        if any(opt in self._options for opt in ["-display", "-nographic", "-curses", "-sdl" "-spice", "-vnc"]):
             return []
-        if len(os.environ.get("DISPLAY", "")) > 0:
-            return []
-        if "-nographic" not in self._options:
+        version = yield from self.manager.get_qemu_version(self.qemu_path)
+        if version and parse_version(version) >= parse_version("3.0"):
+            return ["-display", "none"]
+        else:
             return ["-nographic"]
-        return []
 
     def _run_with_kvm(self, qemu_path, options):
         """
@@ -1678,7 +1679,8 @@ class QemuVM(BaseNode):
             raise QemuError("Console type {} is unknown".format(self._console_type))
         command.extend(self._monitor_options())
         command.extend((yield from self._network_options()))
-        command.extend(self._graphic())
+        if self._console_type == "telnet":
+            command.extend((yield from self._disable_graphics()))
         if additional_options:
             try:
                 command.extend(shlex.split(additional_options))

gns3server/compute/vmware/__init__.py

@@ -149,6 +149,7 @@ class VMware(BaseManager):
         VIX 1.13 was the release for Player 6.
         VIX 1.14 was the release for Player 7.
         VIX 1.15 was the release for Workstation Player 12.
+        VIX 1.17 was the release for Workstation Player 14.
 
         :param player_version: VMware Player major version.
         """
@@ -162,6 +163,8 @@ class VMware(BaseManager):
             yield from self.check_vmrun_version(minimum_required_version="1.14.0")
         elif player_version >= 12:
             yield from self.check_vmrun_version(minimum_required_version="1.15.0")
+        elif player_version >= 14:
+            yield from self.check_vmrun_version(minimum_required_version="1.17.0")
         self._host_type = "player"
 
     @asyncio.coroutine
@@ -172,6 +175,7 @@ class VMware(BaseManager):
         VIX 1.13 was the release for Workstation 10.
         VIX 1.14 was the release for Workstation 11.
         VIX 1.15 was the release for Workstation Pro 12.
+        VIX 1.17 was the release for Workstation Pro 14.
 
         :param ws_version: VMware Workstation major version.
         """
@@ -185,6 +189,8 @@ class VMware(BaseManager):
             yield from self.check_vmrun_version(minimum_required_version="1.14.0")
         elif ws_version >= 12:
             yield from self.check_vmrun_version(minimum_required_version="1.15.0")
+        elif ws_version >= 14:
+            yield from self.check_vmrun_version(minimum_required_version="1.17.0")
         self._host_type = "ws"
 
     @asyncio.coroutine

gns3server/compute/vpcs/vpcs_vm.py

@@ -315,7 +315,7 @@ class VPCSVM(BaseNode):
         if sys.platform.startswith("win32"):
             try:
                 self._process.send_signal(signal.CTRL_BREAK_EVENT)
-            except OSError:
+            except (SystemError, OSError):
                 pass
         else:
             try:

gns3server/controller/compute.py

@@ -446,9 +446,13 @@ class Compute:
             self._capabilities = response.json
 
             if response.json["version"].split("-")[0] != __version__.split("-")[0]:
-                msg = "GNS3 controller version {} is not the same as compute server {} version {}".format(__version__,
-                                                                                                          self._name,
-                                                                                                          response.json["version"])
+                if self._name.startswith("GNS3 VM"):
+                    msg = "GNS3 version {} is not the same as the GNS3 VM version {}. Please upgrade the GNS3 VM.".format(__version__,
+                                                                                                                          response.json["version"])
+                else:
+                    msg = "GNS3 controller version {} is not the same as compute server {} version {}".format(__version__,
+                                                                                                              self._name,
+                                                                                                              response.json["version"])
                 if __version_info__[3] == 0:
                     # Stable release
                     log.error(msg)

gns3server/controller/drawing.py

@@ -55,7 +55,7 @@ class Drawing:
         return self._id
 
     @property
-    def ressource_filename(self):
+    def resource_filename(self):
         """
         If the svg content has been dump to an external file return is name otherwise None
         """

gns3server/controller/export_project.py

@@ -31,7 +31,7 @@ log = logging.getLogger(__name__)
 
 
 @asyncio.coroutine
-def export_project(project, temporary_dir, include_images=False, keep_compute_id=False, allow_all_nodes=False):
+def export_project(project, temporary_dir, include_images=False, keep_compute_id=False, allow_all_nodes=False, reset_mac_addresses=False):
     """
     Export a project to a zip file.
 
@@ -42,6 +42,7 @@ def export_project(project, temporary_dir, include_images=False, keep_compute_id
     :param include images: save OS images to the zip file
     :param keep_compute_id: If false replace all compute id by local (standard behavior for .gns3project to make it portable)
     :param allow_all_nodes: Allow all nodes type to be include in the zip even if not portable
+    :param reset_mac_addresses: Reset MAC addresses for every nodes.
 
     :returns: ZipStream object
     """
@@ -61,10 +62,10 @@ def export_project(project, temporary_dir, include_images=False, keep_compute_id
     # First we process the .gns3 in order to be sure we don't have an error
     for file in os.listdir(project._path):
         if file.endswith(".gns3"):
-            yield from _patch_project_file(project, os.path.join(project._path, file), zstream, include_images, keep_compute_id, allow_all_nodes, temporary_dir)
+            yield from _patch_project_file(project, os.path.join(project._path, file), zstream, include_images, keep_compute_id, allow_all_nodes, temporary_dir, reset_mac_addresses)
 
     # Export the local files
-    for root, dirs, files in os.walk(project._path, topdown=True):
+    for root, dirs, files in os.walk(project._path, topdown=True, followlinks=False):
         files = [f for f in files if _is_exportable(os.path.join(root, f))]
         for file in files:
             path = os.path.join(root, file)
@@ -125,6 +126,7 @@ def _patch_mtime(path):
         new_mtime = file_date.replace(year=1980).timestamp()
         os.utime(path, (st.st_atime, new_mtime))
 
+
 def _is_exportable(path):
     """
     :returns: True if file should not be included in the final archive
@@ -134,6 +136,10 @@ def _is_exportable(path):
     if path.endswith("snapshots"):
         return False
 
+    # do not export symlinks
+    if os.path.islink(path):
+        return False
+
     # do not export directories of snapshots
     if "{sep}snapshots{sep}".format(sep=os.path.sep) in path:
         return False
@@ -155,7 +161,7 @@ def _is_exportable(path):
 
 
 @asyncio.coroutine
-def _patch_project_file(project, path, zstream, include_images, keep_compute_id, allow_all_nodes, temporary_dir):
+def _patch_project_file(project, path, zstream, include_images, keep_compute_id, allow_all_nodes, temporary_dir, reset_mac_addresses):
     """
     Patch a project file (.gns3) to export a project.
     The .gns3 file is renamed to project.gns3
@@ -188,6 +194,10 @@ def _patch_project_file(project, path, zstream, include_images, keep_compute_id,
                 if "properties" in node and node["node_type"] != "docker":
                     for prop, value in node["properties"].items():
 
+                        # reset the MAC address
+                        if reset_mac_addresses and prop in ("mac_addr", "mac_address"):
+                            node["properties"][prop] = None
+
                         if node["node_type"] == "iou":
                             if not prop == "path":
                                 continue

gns3server/controller/gns3vm/vmware_gns3_vm.py

@@ -72,15 +72,19 @@ class VMwareGNS3VM(BaseGNS3VM):
         if ram % 4 != 0:
             raise GNS3VMError("Allocated memory {} for the GNS3 VM must be a multiple of 4".format(ram))
 
-        available_vcpus = psutil.cpu_count(logical=False)
+        available_vcpus = psutil.cpu_count(logical=True)
         if vcpus > available_vcpus:
             raise GNS3VMError("You have allocated too many vCPUs for the GNS3 VM! (max available is {} vCPUs)".format(available_vcpus))
 
         try:
             pairs = VMware.parse_vmware_file(self._vmx_path)
-            pairs["numvcpus"] = str(vcpus)
-            pairs["memsize"] = str(ram)
-            VMware.write_vmx_file(self._vmx_path, pairs)
+            if vcpus > 1:
+                pairs["numvcpus"] = str(vcpus)
+                cores_per_sockets = int(vcpus / psutil.cpu_count(logical=False))
+                if cores_per_sockets > 1:
+                    pairs["cpuid.corespersocket"] = str(cores_per_sockets)
+                pairs["memsize"] = str(ram)
+                VMware.write_vmx_file(self._vmx_path, pairs)
             log.info("GNS3 VM vCPU count set to {} and RAM amount set to {}".format(vcpus, ram))
         except OSError as e:
             raise GNS3VMError('Could not read/write VMware VMX file "{}": {}'.format(self._vmx_path, e))

gns3server/controller/import_project.py

@@ -184,9 +184,11 @@ def _move_files_to_compute(compute, project_id, directory, files_path):
 
     location = os.path.join(directory, files_path)
     if os.path.exists(location):
-        for (dirpath, dirnames, filenames) in os.walk(location):
+        for (dirpath, dirnames, filenames) in os.walk(location, followlinks=False):
             for filename in filenames:
                 path = os.path.join(dirpath, filename)
+                if os.path.islink(path):
+                    continue
                 dst = os.path.relpath(path, directory)
                 yield from _upload_file(compute, project_id, path, dst)
         yield from wait_run_in_executor(shutil.rmtree, os.path.join(directory, files_path))
@@ -213,9 +215,11 @@ def _import_images(controller, path):
 
     image_dir = controller.images_path()
     root = os.path.join(path, "images")
-    for (dirpath, dirnames, filenames) in os.walk(root):
+    for (dirpath, dirnames, filenames) in os.walk(root, followlinks=False):
         for filename in filenames:
             path = os.path.join(dirpath, filename)
+            if os.path.islink(path):
+                continue
             dst = os.path.join(image_dir, os.path.relpath(path, root))
             os.makedirs(os.path.dirname(dst), exist_ok=True)
             shutil.move(path, dst)

gns3server/controller/node.py

@@ -91,12 +91,12 @@ class Node:
 
         # Update node properties with additional elements
         for prop in kwargs:
-            if prop not in ignore_properties:
+            if prop and prop not in ignore_properties:
                 if hasattr(self, prop):
                     try:
                         setattr(self, prop, kwargs[prop])
                     except AttributeError as e:
-                        log.critical("Can't set attribute %s", prop)
+                        log.critical("Cannot set attribute '%s'".format(prop))
                         raise e
                 else:
                     if prop not in self.CONTROLLER_ONLY_PROPERTIES and kwargs[prop] is not None and kwargs[prop] != "":
@@ -254,7 +254,7 @@ class Node:
         try:
             self._width, self._height, filetype = self._project.controller.symbols.get_size(val)
         except (ValueError, OSError) as e:
-            log.error("Could not write symbol: {}".format(e))
+            log.error("Could not set symbol: {}".format(e))
             # If symbol is invalid we replace it by the default
             self.symbol = ":/symbols/computer.svg"
         if self._label is None:

gns3server/controller/project.py

@@ -146,7 +146,6 @@ class Project:
                     }
                 )
 
-
     def reset(self):
         """
         Called when open/close a project. Cleanup internal stuff
@@ -704,24 +703,26 @@ class Project:
             # We don't care if a compute is down at this step
             except (ComputeError, aiohttp.web.HTTPError, aiohttp.ClientResponseError, TimeoutError):
                 pass
-        self._cleanPictures()
+        self._clean_pictures()
         self._status = "closed"
         if not ignore_notification:
             self.controller.notification.emit("project.closed", self.__json__())
 
-    def _cleanPictures(self):
+    def _clean_pictures(self):
         """
-        Delete unused images
+        Delete unused pictures.
         """
 
-        # Project have been deleted
-        if not os.path.exists(self.path):
+        # Project have been deleted or is loading or is not opened
+        if not os.path.exists(self.path) or self._loading or self._status != "opened":
             return
         try:
             pictures = set(os.listdir(self.pictures_directory))
             for drawing in self._drawings.values():
                 try:
-                    pictures.remove(drawing.ressource_filename)
+                    resource_filename = drawing.resource_filename
+                    if resource_filename:
+                        pictures.remove(resource_filename)
                 except KeyError:
                     pass
 
@@ -733,11 +734,12 @@ class Project:
                 except KeyError:
                     pass
 
-
-            for pict in pictures:
-                os.remove(os.path.join(self.pictures_directory, pict))
+            for pic_filename in pictures:
+                path = os.path.join(self.pictures_directory, pic_filename)
+                log.info("Deleting unused picture '{}'".format(path))
+                os.remove(path)
         except OSError as e:
-            log.warning(str(e))
+            log.warning("Could not delete unused pictures: {}".format(e))
 
     @asyncio.coroutine
     def delete(self):
@@ -929,13 +931,13 @@ class Project:
         self.dump()
         try:
             with tempfile.TemporaryDirectory() as tmpdir:
-                zipstream = yield from export_project(self, tmpdir, keep_compute_id=True, allow_all_nodes=True)
+                zipstream = yield from export_project(self, tmpdir, keep_compute_id=True, allow_all_nodes=True, reset_mac_addresses=True)
                 project_path = os.path.join(tmpdir, "project.gns3p")
                 yield from wait_run_in_executor(self._create_duplicate_project_file, project_path, zipstream)
                 with open(project_path, "rb") as f:
                     project = yield from import_project(self._controller, str(uuid.uuid4()), f, location=location, name=name, keep_compute_id=True)
         except (ValueError, OSError, UnicodeEncodeError) as e:
-            raise aiohttp.web.HTTPConflict(text="Can not duplicate project: {}".format(str(e)))
+            raise aiohttp.web.HTTPConflict(text="Cannot duplicate project: {}".format(str(e)))
 
         if previous_status == "closed":
             yield from self.close()

gns3server/crash_report.py

@@ -57,7 +57,7 @@ class CrashReport:
     Report crash to a third party service
     """
 
-    DSN = "https://71472e2845b2422cab117bdfa8af55ec:da50c8f08b9f44d8b905356741c278df@sentry.io/38482"
+    DSN = "https://6536a51af3154084859112bacb4f2277:72822a229faa463ba6c9338126a6f280@sentry.io/38482"
     if hasattr(sys, "frozen"):
         cacert = get_resource("cacert.pem")
         if cacert is not None and os.path.isfile(cacert):

gns3server/handlers/api/compute/docker_handler.py

@@ -263,10 +263,11 @@ class DockerHandler:
 
         docker_manager = Docker.instance()
         container = docker_manager.get_node(request.match_info["node_id"], project_id=request.match_info["project_id"])
-        nio = container.ethernet_adapters[int(request.match_info["adapter_number"])].get_nio(0)
+        adapter_number = int(request.match_info["adapter_number"])
+        nio = container.ethernet_adapters[adapter_number].get_nio(0)
         if "filters" in request.json and nio:
             nio.filters = request.json["filters"]
-        yield from container.adapter_update_nio_binding(int(request.match_info["port_number"]), nio)
+        yield from container.adapter_update_nio_binding(adapter_number, nio)
         response.set_status(201)
         response.json(request.json)
 

gns3server/handlers/api/compute/iou_handler.py

@@ -301,12 +301,9 @@ class IOUHandler:
         nio = vm.adapters[adapter_number].get_nio(port_number)
         if "filters" in request.json and nio:
             nio.filters = request.json["filters"]
-        yield from vm.adapter_update_nio_binding(
-            adapter_number,
-            port_number,
-            nio)
+        yield from vm.adapter_update_nio_binding(adapter_number, port_number, nio)
         response.set_status(201)
-        response.json(nio)
+        response.json(request.json)
 
     @Route.delete(
         r"/projects/{project_id}/iou/nodes/{node_id}/adapters/{adapter_number:\d+}/ports/{port_number:\d+}/nio",

gns3server/handlers/api/compute/server_handler.py

@@ -42,9 +42,9 @@ class ServerHandler:
 
     @Route.get(
         r"/debug",
-        description="Return debug informations about the compute",
+        description="Return debug information about the compute",
         status_codes={
-            201: "Writed"
+            201: "Written"
         })
     def debug(request, response):
         response.content_type = "text/plain"

gns3server/handlers/api/compute/vmware_handler.py

@@ -279,12 +279,13 @@ class VMwareHandler:
 
         vmware_manager = VMware.instance()
         vm = vmware_manager.get_node(request.match_info["node_id"], project_id=request.match_info["project_id"])
-        nio = vm.ethernet_adapters[int(request.match_info["adapter_number"])]
+        adapter_number = int(request.match_info["adapter_number"])
+        nio = vm.ethernet_adapters[adapter_number]
         if "filters" in request.json and nio:
             nio.filters = request.json["filters"]
-            yield from vm.adapter_update_nio_binding(int(request.match_info["adapter_number"]), nio)
-            response.set_status(201)
-            response.json(request.json)
+        yield from vm.adapter_update_nio_binding(adapter_number, nio)
+        response.set_status(201)
+        response.json(request.json)
 
     @Route.delete(
         r"/projects/{project_id}/vmware/nodes/{node_id}/adapters/{adapter_number:\d+}/ports/{port_number:\d+}/nio",

gns3server/handlers/api/controller/appliance_handler.py

@@ -67,8 +67,9 @@ class ApplianceHandler:
 
         controller = Controller.instance()
         project = controller.get_project(request.match_info["project_id"])
-        yield from project.add_node_from_appliance(request.match_info["appliance_id"],
-                                                   x=request.json["x"],
-                                                   y=request.json["y"],
-                                                   compute_id=request.json.get("compute_id"))
+        node = yield from project.add_node_from_appliance(request.match_info["appliance_id"],
+                                                          x=request.json["x"],
+                                                          y=request.json["y"],
+                                                          compute_id=request.json.get("compute_id"))
         response.set_status(201)
+        response.json(node)

gns3server/handlers/api/controller/compute_handler.py

@@ -59,7 +59,7 @@ class ComputeHandler:
 
     @Route.put(
         r"/computes/{compute_id}",
-        description="Get a compute server information",
+        description="Update a compute server",
         status_codes={
             200: "Compute server updated",
             400: "Invalid request",

gns3server/handlers/api/controller/project_handler.py

@@ -287,9 +287,9 @@ class ProjectHandler:
 
         try:
             with tempfile.TemporaryDirectory() as tmp_dir:
-                datas = yield from export_project(
-                    project, tmp_dir,
-                    include_images=bool(int(request.query.get("include_images", "0"))))
+                stream = yield from export_project(project,
+                                                   tmp_dir,
+                                                   include_images=bool(int(request.query.get("include_images", "0"))))
                 # We need to do that now because export could failed and raise an HTTP error
                 # that why response start need to be the later possible
                 response.content_type = 'application/gns3project'
@@ -297,7 +297,7 @@ class ProjectHandler:
                 response.enable_chunked_encoding()
                 yield from response.prepare(request)
 
-                for data in datas:
+                for data in stream:
                     response.write(data)
                     yield from response.drain()
 

gns3server/handlers/api/controller/server_handler.py

@@ -139,7 +139,7 @@ class ServerHandler:
         r"/debug",
         description="Dump debug information to disk (debug directory in config directory). Work only for local server",
         status_codes={
-            201: "Writed"
+            201: "Written"
         })
     def debug(request, response):
 
@@ -156,7 +156,7 @@ class ServerHandler:
                 f.write(ServerHandler._getDebugData())
         except Exception as e:
             # If something is wrong we log the info to the log and we hope the log will be include correctly to the debug export
-            log.error("Could not export debug informations {}".format(e), exc_info=1)
+            log.error("Could not export debug information {}".format(e), exc_info=1)
 
         try:
             if Controller.instance().gns3vm.engine == "vmware":

gns3server/handlers/api/controller/symbol_handler.py

@@ -60,7 +60,7 @@ class SymbolHandler:
         r"/symbols/{symbol_id:.+}/raw",
         description="Write the symbol file",
         status_codes={
-            200: "Symbol returned"
+            200: "Symbol written"
         },
         raw=True)
     def upload(request, response):
@@ -78,6 +78,7 @@ class SymbolHandler:
                     f.write(chunk)
         except (UnicodeEncodeError, OSError) as e:
             raise aiohttp.web.HTTPConflict(text="Could not write symbol file '{}': {}".format(path, e))
+
         # Reset the symbol list
         controller.symbols.list()
         response.set_status(204)

gns3server/run.py

@@ -91,7 +91,6 @@ def parse_arguments(argv):
     parser.add_argument("--host", help="run on the given host/IP address")
     parser.add_argument("--port", help="run on the given port", type=int)
     parser.add_argument("--ssl", action="store_true", help="run in SSL mode")
-    parser.add_argument("--no-ubridge", action="store_false", help="do not use ubridge to handle node connections")
     parser.add_argument("--config", help="Configuration file")
     parser.add_argument("--certfile", help="SSL cert file")
     parser.add_argument("--certkey", help="SSL key file")

gns3server/schemas/dynamips_vm.py

@@ -137,7 +137,7 @@ VM_CREATE_SCHEMA = {
         },
         "mac_addr": {
             "description": "Base MAC address",
-            "type": "string",
+            "type": ["null", "string"],
             "minLength": 1,
             "pattern": "^([0-9a-fA-F]{4}\\.){2}[0-9a-fA-F]{4}$"
         },
@@ -355,7 +355,7 @@ VM_UPDATE_SCHEMA = {
         },
         "mac_addr": {
             "description": "Base MAC address",
-            "type": "string",
+            "type": ["null", "string"],
             "minLength": 1,
             "pattern": "^([0-9a-fA-F]{4}\\.){2}[0-9a-fA-F]{4}$"
         },
@@ -595,7 +595,7 @@ VM_OBJECT_SCHEMA = {
         },
         "mac_addr": {
             "description": "Base MAC address",
-            "type": "string",
+            "type": ["null", "string"]
             #"minLength": 1,
             #"pattern": "^([0-9a-fA-F]{4}\\.){2}[0-9a-fA-F]{4}$"
         },

gns3server/schemas/qemu.py

@@ -537,7 +537,7 @@ QEMU_OBJECT_SCHEMA = {
             "description": "Console TCP port",
             "minimum": 1,
             "maximum": 65535,
-            "type": "integer"
+            "type": ["integer", "null"]
         },
         "console_type": {
             "description": "Console type",

gns3server/ubridge/ubridge_hypervisor.py

@@ -69,7 +69,7 @@ class UBridgeHypervisor:
         connection_success = False
         last_exception = None
         while time.time() - begin < timeout:
-            yield from asyncio.sleep(0.01)
+            yield from asyncio.sleep(0.1)
             try:
                 self._reader, self._writer = yield from asyncio.open_connection(host, self._port)
             except OSError as e:
@@ -84,6 +84,7 @@ class UBridgeHypervisor:
             log.info("Connected to uBridge hypervisor after {:.4f} seconds".format(time.time() - begin))
 
         try:
+            yield from asyncio.sleep(0.1)
             version = yield from self.send("hypervisor version")
             self._version = version[0].split("-", 1)[0]
         except IndexError:
@@ -237,7 +238,7 @@ class UBridgeHypervisor:
                                             .format(host=self._host, port=self._port, command=command, run=self.is_running()))
                     else:
                         retries += 1
-                        yield from asyncio.sleep(0.1)
+                        yield from asyncio.sleep(0.5)
                         continue
                 retries = 0
                 buf += chunk.decode("utf-8")

gns3server/utils/asyncio/embed_shell.py

@@ -198,8 +198,8 @@ class UnstoppableEventLoop(EventLoop):
 
 
 class ShellConnection(TelnetConnection):
-    def __init__(self, reader, writer, shell, loop):
-        super(ShellConnection, self).__init__(reader, writer)
+    def __init__(self, reader, writer, shell, window_size_changed_callback, loop):
+        super(ShellConnection, self).__init__(reader, writer, window_size_changed_callback)
         self._shell = shell
         self._loop = loop
         self._cli = None
@@ -239,9 +239,12 @@ class ShellConnection(TelnetConnection):
     def disconnected(self):
         pass
 
+    @asyncio.coroutine
     def window_size_changed(self, columns, rows):
         self._size = Size(rows=rows, columns=columns)
         self._cb.terminal_size_changed()
+        if self._window_size_changed_callback:
+            yield from self._window_size_changed_callback(columns, rows)
 
     @asyncio.coroutine
     def feed(self, data):
@@ -291,8 +294,8 @@ def create_telnet_shell(shell, loop=None):
     if loop is None:
         loop = asyncio.get_event_loop()
 
-    def factory(reader, writer):
-        return ShellConnection(reader, writer, shell, loop)
+    def factory(reader, writer, window_size_changed_callback):
+        return ShellConnection(reader, writer, shell, window_size_changed_callback, loop)
 
     return AsyncioTelnetServer(binary=True, echo=True, naws=True, connection_factory=factory)
 

gns3server/utils/asyncio/telnet_server.py

@@ -15,7 +15,7 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
-import re
+
 import asyncio
 import asyncio.subprocess
 import struct
@@ -62,10 +62,11 @@ READ_SIZE = 1024
 
 class TelnetConnection(object):
     """Default implementation of telnet connection which may but may not be used."""
-    def __init__(self, reader, writer):
+    def __init__(self, reader, writer, window_size_changed_callback=None):
         self.is_closing = False
         self._reader = reader
         self._writer = writer
+        self._window_size_changed_callback = window_size_changed_callback
 
     @property
     def reader(self):
@@ -85,10 +86,13 @@ class TelnetConnection(object):
         """Method called when client is disconnecting"""
         pass
 
+    @asyncio.coroutine
     def window_size_changed(self, columns, rows):
         """Method called when window size changed, only can occur when
          `naws` flag is enable in server configuration."""
-        pass
+
+        if self._window_size_changed_callback:
+            yield from self._window_size_changed_callback(columns, rows)
 
     @asyncio.coroutine
     def feed(self, data):
@@ -116,7 +120,7 @@ class TelnetConnection(object):
 class AsyncioTelnetServer:
     MAX_NEGOTIATION_READ = 10
 
-    def __init__(self, reader=None, writer=None, binary=True, echo=False, naws=False, connection_factory=None):
+    def __init__(self, reader=None, writer=None, binary=True, echo=False, naws=False, window_size_changed_callback=None, connection_factory=None):
         """
         Initializes telnet server
         :param naws when True make a window size negotiation
@@ -131,6 +135,7 @@ class AsyncioTelnetServer:
         self._lock = asyncio.Lock()
         self._reader_process = None
         self._current_read = None
+        self._window_size_changed_callback = window_size_changed_callback
 
         self._binary = binary
         # If echo is true when the client send data
@@ -139,8 +144,8 @@ class AsyncioTelnetServer:
         self._echo = echo
         self._naws = naws
 
-        def default_connection_factory(reader, writer):
-            return TelnetConnection(reader, writer)
+        def default_connection_factory(reader, writer, window_size_changed_callback):
+            return TelnetConnection(reader, writer, window_size_changed_callback)
 
         if connection_factory is None:
             connection_factory = default_connection_factory
@@ -190,7 +195,7 @@ class AsyncioTelnetServer:
     @asyncio.coroutine
     def run(self, network_reader, network_writer):
         # Keep track of connected clients
-        connection = self._connection_factory(network_reader, network_writer)
+        connection = self._connection_factory(network_reader, network_writer, self._window_size_changed_callback)
         self._connections[network_writer] = connection
 
         try:
@@ -215,7 +220,7 @@ class AsyncioTelnetServer:
             try:
                 writer.write_eof()
                 yield from writer.drain()
-            except ConnectionError:
+            except (AttributeError, ConnectionError):
                 continue
 
     @asyncio.coroutine
@@ -307,6 +312,7 @@ class AsyncioTelnetServer:
             cmd.append(buffer[location])
             return op
 
+    @asyncio.coroutine
     def _negotiate(self, data, connection):
         """ Performs negotiation commands"""
 
@@ -314,7 +320,7 @@ class AsyncioTelnetServer:
         if command == NAWS:
             if len(payload) == 4:
                 columns, rows = struct.unpack(str('!HH'), bytes(payload))
-                connection.window_size_changed(columns, rows)
+                yield from connection.window_size_changed(columns, rows)
             else:
                 log.warning('Wrong number of NAWS bytes')
         else:
@@ -373,7 +379,7 @@ class AsyncioTelnetServer:
                         break
 
                 # SE command is followed by IAC, remove the last two operations from stack
-                self._negotiate(negotiation[0:-2], connection)
+                yield from self._negotiate(negotiation[0:-2], connection)
 
             # This must be a 3-byte TELNET command
             else:

gns3server/utils/picture.py

@@ -15,6 +15,7 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
+import re
 import io
 import struct
 from xml.etree.ElementTree import ElementTree, ParseError
@@ -103,13 +104,21 @@ def get_size(data, default_width=0, default_height=0):
         root = tree.getroot()
 
         try:
-            width_attr = root.attrib.get("width", "0")
-            height_attr = root.attrib.get("height", "0")
+            width_attr = root.attrib.get("width", "100%")
+            height_attr = root.attrib.get("height", "100%")
             if width_attr.endswith("%") or height_attr.endswith("%"):
                 # check to viewBox attribute if width or height value is a percentage
-                _, _, width_attr, height_attr = root.attrib.get("viewBox").split()
+                viewbox = root.attrib.get("viewBox")
+                if not viewbox:
+                    raise ValueError("Invalid SVG file: missing viewBox attribute")
+                _, _, viewbox_width, viewbox_height = re.split(r'[\s,]+', viewbox)
+            if width_attr.endswith("%"):
+                width = _svg_convert_size(viewbox_width, width_attr)
             else:
                 width = _svg_convert_size(width_attr)
+            if height_attr.endswith("%"):
+                height = _svg_convert_size(viewbox_height, height_attr)
+            else:
                 height = _svg_convert_size(height_attr)
         except (AttributeError, IndexError) as e:
             raise ValueError("Invalid SVG file: {}".format(e))
@@ -117,11 +126,12 @@ def get_size(data, default_width=0, default_height=0):
     return width, height, filetype
 
 
-def _svg_convert_size(size):
+def _svg_convert_size(size, percent=None):
     """
     Convert svg size to the px version
 
     :param size: String with the size
+    :param percent: String with the percentage, None = 100%
     """
 
     # https://www.w3.org/TR/SVG/coords.html#Units
@@ -133,8 +143,11 @@ def _svg_convert_size(size):
         "in": 90,
         "px": 1
     }
-    if len(size) > 3:
+    factor = 1.0
+    if len(size) >= 3:
         if size[-2:] in conversion_table:
-            return round(float(size[:-2]) * conversion_table[size[-2:]])
-
-    return round(float(size))
+            factor = conversion_table[size[-2:]]
+            size = size[:-2]
+    if percent:
+        factor *= float(percent.rstrip("%")) / 100.0
+    return round(float(size) * factor)

gns3server/version.py

@@ -23,8 +23,8 @@
 # or negative for a release candidate or beta (after the base version
 # number has been incremented)
 
-__version__ = "2.1.10"
-__version_info__ = (2, 1, 10, 0)
+__version__ = "2.1.16"
+__version_info__ = (2, 1, 16, 0)
 
 # If it's a git checkout try to add the commit
 if "dev" in __version__:

requirements.txt

@@ -1,4 +1,4 @@
-jsonschema>=2.4.0
+jsonschema==2.6.0
 aiohttp>=2.3.3,<2.4.0 # pyup: ignore
 aiohttp-cors>=0.5.3,<0.6.0 # pyup: ignore
 yarl>=0.11
@@ -8,4 +8,4 @@ psutil>=3.0.0
 zipstream>=1.1.4
 typing>=3.5.3.0 # Otherwise yarl fails with python 3.4
 prompt-toolkit==1.0.15
-async-timeout<3.0.0 # pyup: ignore; 3.0 drops support for python 3.4
+async-timeout<3.0.0 # pyup: ignore; 3.0 drops support for python 3.4

tests/compute/docker/test_docker_vm.py

@@ -971,13 +971,12 @@ def test_mount_binds(vm, tmpdir):
 
 def test_start_vnc(vm, loop):
     vm.console_resolution = "1280x1024"
-    with patch("shutil.which", return_value="/bin/x"):
+    with patch("shutil.which", return_value="/bin/Xtigervnc"):
         with asyncio_patch("gns3server.compute.docker.docker_vm.wait_for_file_creation") as mock_wait:
             with asyncio_patch("asyncio.create_subprocess_exec") as mock_exec:
                 loop.run_until_complete(asyncio.ensure_future(vm._start_vnc()))
     assert vm._display is not None
-    mock_exec.assert_any_call("Xvfb", "-nolisten", "tcp", ":{}".format(vm._display), "-screen", "0", "1280x1024x16")
-    mock_exec.assert_any_call("x11vnc", "-forever", "-nopw", "-shared", "-geometry", "1280x1024", "-display", "WAIT:{}".format(vm._display), "-rfbport", str(vm.console), "-rfbportv6", str(vm.console), "-noncache", "-listen", "127.0.0.1")
+    assert mock_exec.call_args[0] == ("Xtigervnc", "-geometry", vm.console_resolution, "-depth", "16", "-interface", "127.0.0.1", "-rfbport", str(vm.console), "-AlwaysShared", "-SecurityTypes", "None", ":{}".format(vm._display))
     mock_wait.assert_called_with("/tmp/.X11-unix/X{}".format(vm._display))
 
 

tests/compute/qemu/test_qemu_vm.py

@@ -116,6 +116,7 @@ def test_is_running(vm, running_subprocess_mock):
 
 
 def test_start(loop, vm, running_subprocess_mock):
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="3.1.0")
     with asyncio_patch("gns3server.compute.qemu.QemuVM.start_wrap_console"):
         with asyncio_patch("asyncio.create_subprocess_exec", return_value=running_subprocess_mock) as mock:
             loop.run_until_complete(asyncio.ensure_future(vm.start()))
@@ -130,6 +131,7 @@ def test_stop(loop, vm, running_subprocess_mock):
     future = asyncio.Future()
     future.set_result(True)
     process.wait.return_value = future
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="3.1.0")
 
     with asyncio_patch("gns3server.compute.qemu.QemuVM.start_wrap_console"):
         with asyncio_patch("asyncio.create_subprocess_exec", return_value=process):
@@ -213,6 +215,7 @@ def test_port_remove_nio_binding(vm, loop):
 
 
 def test_close(vm, port_manager, loop):
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="3.1.0")
     with asyncio_patch("gns3server.compute.qemu.QemuVM.start_wrap_console"):
         with asyncio_patch("asyncio.create_subprocess_exec", return_value=MagicMock()):
             loop.run_until_complete(asyncio.ensure_future(vm.start()))
@@ -338,6 +341,7 @@ def test_disk_options(vm, tmpdir, loop, fake_qemu_img_binary):
 
 def test_cdrom_option(vm, tmpdir, loop, fake_qemu_img_binary):
 
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="3.1.0")
     vm._cdrom_image = str(tmpdir / "test.iso")
     open(vm._cdrom_image, "w+").close()
 
@@ -348,6 +352,7 @@ def test_cdrom_option(vm, tmpdir, loop, fake_qemu_img_binary):
 
 def test_bios_option(vm, tmpdir, loop, fake_qemu_img_binary):
 
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="3.1.0")
     vm._bios_image = str(tmpdir / "test.img")
     open(vm._bios_image, "w+").close()
 
@@ -454,6 +459,7 @@ def test_control_vm_expect_text(vm, loop, running_subprocess_mock):
 
 def test_build_command(vm, loop, fake_qemu_binary, port_manager):
 
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="3.1.0")
     os.environ["DISPLAY"] = "0:0"
     with asyncio_patch("asyncio.create_subprocess_exec", return_value=MagicMock()) as process:
         cmd = loop.run_until_complete(asyncio.ensure_future(vm._build_command()))
@@ -477,7 +483,9 @@ def test_build_command(vm, loop, fake_qemu_binary, port_manager):
             "-device",
             "e1000,mac={},netdev=gns3-0".format(vm._mac_address),
             "-netdev",
-            "socket,id=gns3-0,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio.rport, nio.lport)
+            "socket,id=gns3-0,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio.rport, nio.lport),
+            "-display",
+            "none"
         ]
 
 
@@ -486,6 +494,7 @@ def test_build_command_manual_uuid(vm, loop, fake_qemu_binary, port_manager):
     If user has set a uuid we keep it
     """
 
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="3.1.0")
     vm.options = "-uuid e1c307a4-896f-11e6-81a5-3c07547807cc"
     os.environ["DISPLAY"] = "0:0"
     with asyncio_patch("asyncio.create_subprocess_exec", return_value=MagicMock()) as process:
@@ -524,7 +533,8 @@ def test_build_command_kvm(linux_platform, vm, loop, fake_qemu_binary, port_mana
             "-device",
             "e1000,mac={},netdev=gns3-0".format(vm._mac_address),
             "-netdev",
-            "socket,id=gns3-0,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio.rport, nio.lport)
+            "socket,id=gns3-0,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio.rport, nio.lport),
+            "-nographic"
         ]
 
 
@@ -560,13 +570,52 @@ def test_build_command_kvm_2_4(linux_platform, vm, loop, fake_qemu_binary, port_
             "-device",
             "e1000,mac={},netdev=gns3-0".format(vm._mac_address),
             "-netdev",
-            "socket,id=gns3-0,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio.rport, nio.lport)
+            "socket,id=gns3-0,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio.rport, nio.lport),
+            "-nographic"
         ]
 
 
+def test_build_command_kvm_3_1(linux_platform, vm, loop, fake_qemu_binary, port_manager):
+    """
+    Qemu 2.4 introduce an issue with KVM
+    """
+    vm._run_with_kvm = MagicMock(return_value=True)
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="3.1.0")
+    os.environ["DISPLAY"] = "0:0"
+    with asyncio_patch("asyncio.create_subprocess_exec", return_value=MagicMock()) as process:
+        cmd = loop.run_until_complete(asyncio.ensure_future(vm._build_command()))
+        nio = vm._local_udp_tunnels[0][0]
+        assert cmd == [
+            fake_qemu_binary,
+            "-name",
+            "test",
+            "-m",
+            "256M",
+            "-smp",
+            "cpus=1",
+            "-enable-kvm",
+            "-machine",
+            "smm=off",
+            "-boot",
+            "order=c",
+            "-uuid",
+            vm.id,
+            "-serial",
+            "telnet:127.0.0.1:{},server,nowait".format(vm._internal_console_port),
+            "-net",
+            "none",
+            "-device",
+            "e1000,mac={},netdev=gns3-0".format(vm._mac_address),
+            "-netdev",
+            "socket,id=gns3-0,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio.rport, nio.lport),
+            "-display",
+            "none"
+        ]
+
 @pytest.mark.skipif(sys.platform.startswith("win"), reason="Not supported on Windows")
 def test_build_command_without_display(vm, loop, fake_qemu_binary):
 
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="2.5.0")
     os.environ["DISPLAY"] = ""
     with asyncio_patch("asyncio.create_subprocess_exec", return_value=MagicMock()) as process:
         cmd = loop.run_until_complete(asyncio.ensure_future(vm._build_command()))
@@ -575,6 +624,7 @@ def test_build_command_without_display(vm, loop, fake_qemu_binary):
 
 def test_build_command_two_adapters(vm, loop, fake_qemu_binary, port_manager):
 
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="2.5.0")
     os.environ["DISPLAY"] = "0:0"
     vm.adapters = 2
     with asyncio_patch("asyncio.create_subprocess_exec", return_value=MagicMock()) as process:
@@ -604,7 +654,8 @@ def test_build_command_two_adapters(vm, loop, fake_qemu_binary, port_manager):
             "-device",
             "e1000,mac={},netdev=gns3-1".format(int_to_macaddress(macaddress_to_int(vm._mac_address) + 1)),
             "-netdev",
-            "socket,id=gns3-1,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio2.rport, nio2.lport)
+            "socket,id=gns3-1,udp=127.0.0.1:{},localaddr=127.0.0.1:{}".format(nio2.rport, nio2.lport),
+            "-nographic"
         ]
 
 
@@ -613,6 +664,7 @@ def test_build_command_two_adapters_mac_address(vm, loop, fake_qemu_binary, port
     Should support multiple base vmac address
     """
 
+    vm.manager.get_qemu_version = AsyncioMagicMock(return_value="2.5.0")
     vm.adapters = 2
     vm.mac_address = "00:00:ab:0e:0f:09"
     mac_0 = vm._mac_address

tests/handlers/api/controller/test_appliance.py

@@ -75,14 +75,13 @@ def test_create_node_from_appliance(http_controller, controller, project, comput
         "name": "test",
         "symbol": "guest.svg",
         "default_name_format": "{name}-{0}",
-        "server": "example.com"
+        "compute_id": "example.com"
     })}
-    with asyncio_patch("gns3server.controller.project.Project.add_node_from_appliance") as mock:
+    with asyncio_patch("gns3server.controller.project.Project.add_node_from_appliance", return_value={"name": "test", "node_type": "qemu", "compute_id": "example.com"}) as mock:
         response = http_controller.post("/projects/{}/appliances/{}".format(project.id, id), {
             "x": 42,
             "y": 12
         })
     mock.assert_called_with(id, x=42, y=12, compute_id=None)
-    print(response.body)
     assert response.route == "/projects/{project_id}/appliances/{appliance_id}"
     assert response.status == 201