date:20090115

Re: [libvirt] [PATCH] Improve xend_get error message

2009-01-15 Thread Daniel P. Berrange

On Wed, Jan 14, 2009 at 06:04:19PM -0800, john.le...@sun.com wrote:
 # HG changeset patch
 # User john.le...@sun.com
 # Date 1231982481 28800
 # Node ID 082e0f7d5de236e69bea177c8d4c4204350144c1
 # Parent  1a16d4b76232845091c1f800569ec462860cb1fb
 Improve xend_get error message
 
 Signed-off-by: John Levon john.le...@sun.com

ACK

Daniel
-- 
|: Red Hat, Engineering, London   -o-   http://people.redhat.com/berrange/ :|
|: http://libvirt.org  -o-  http://virt-manager.org  -o-  http://ovirt.org :|
|: http://autobuild.org   -o- http://search.cpan.org/~danberr/ :|
|: GnuPG: 7D3B9505  -o-  F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

--
Libvir-list mailing list
Libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] [PATCH] Solaris least privilege support

2009-01-15 Thread Daniel P. Berrange

On Wed, Jan 14, 2009 at 07:32:28PM -0800, john.le...@sun.com wrote:
 @@ -638,10 +657,32 @@ static int qemudInitPaths(struct qemud_s
  static int qemudInitPaths(struct qemud_server *server,
char *sockname,
char *roSockname,
 -  int maxlen) {
 +  int maxlen)
 +{
  uid_t uid = geteuid();
 -
 +#ifdef __sun
 +char *base = NULL;
 +
 +if (virAsprintf (base, %s/run/libvirt, LOCAL_STATE_DIR) == -1) {
 +VIR_ERROR0(_(Out of memory));
 +return -1;
 +}
 +if (mkdir (base, 0755)) {
 +if (errno != EEXIST) {
 +VIR_ERROR0(_(unable to create rundir));
 +free (base);
 +exit(-1);
 +}
 +}
 +
 +free (base);
 +#endif

This chunk is potentially usefull on Linux too, so can just remove
the #ifdef here. Also make sure to use VIR_FREE rather than free()
Finally, the exit(-1) should be a return -1;

 +#ifdef __sun
 +if (uid == 60) {
 +#else
  if (!uid) {
 +#endif

Rather than have this magic value in the code, lets put a #define at
the top of the file 

 #ifdef __sun
 #define SYSTEM_ACCOUNT 60
 #else
 #define SYSTEM_ACCOUNT 0
 #endif

And then just change all use of !uid to   'uid == SYSTEM_ACCOUNT'

 +#ifdef __sun
 +{
 +ucred_t *ucred = NULL;
 +const priv_set_t *privs;
 +
 +if (getpeerucred (fd, ucred) == -1 ||
 +(privs = ucred_getprivset (ucred, PRIV_EFFECTIVE)) == NULL) {
 +if (ucred != NULL)
 +ucred_free (ucred);
 +close (fd);
 +return -1;
 +}
 +
 +if (!priv_ismember (privs, PRIV_VIRT_MANAGE)) {
 +ucred_free (ucred);
 +close (fd);
 +return -1;
 +}
 +
 +ucred_free (ucred);

Can move the ucred_free up before priv_ismember() call and thus
avoid the need for the call in the cleanup path.

 +}
 +#endif /* __sun */
 +
  /* Disable Nagle.  Unix sockets will ignore this. */
  setsockopt (fd, IPPROTO_TCP, TCP_NODELAY, (void *)no_slow_start,
  sizeof no_slow_start);
 @@ -2140,6 +2204,10 @@ remoteReadConfigFile (struct qemud_serve
  if (auth_unix_rw == REMOTE_AUTH_POLKIT)
  unix_sock_rw_mask = 0777;
  #endif
 +#ifdef __sun
 +unix_sock_rw_mask = 0666;
 +#endif

Can we move this up to the declaration point - this function should
only be reacting to config file settings. FOr the global default we
can just declare it 

#ifdef __sun
static int unix_sock_rw_mask = 0666; /* Allow world */
static int unix_sock_ro_mask = 0777; /* Allow world */
#else
static int unix_sock_rw_mask = 0700; /* Allow user only */
static int unix_sock_ro_mask = 0777; /* Allow world */
#endif


 +#ifdef __sun
 +static void
 +qemudSetupPrivs (struct qemud_server *server)
 +{
 +chown (/var/run/libvirt, 60, 60);
 +chown (/var/run/libvirt/libvirt-sock, 60, 60);
 +chmod (/var/run/libvirt/libvirt-sock, 0666);
 +chown (server-logDir, 60, 60);

Again prefer to see the #define SYTEM_ACCOUNT instead of the magic
numbers.

Is the chmod of the socket really required for solaris ? We already
set the umask before creating the socket, so it ought to get desired
permissions from moment it appears. We avoided an explicit chmod
because that leaves a gap between socket creation, and correct ownership
and permissions being configured.

Also, if this libvirtd is running as UID 60, so the chown really
needed ?  We also setgid before creating the socket so that it
gets desired group ownership at time of creation, rather than
having to change it post-create.

 +
 +if (__init_daemon_priv (PU_RESETGROUPS | PU_CLEARLIMITSET,
 +60, 60, PRIV_XVM_CONTROL, NULL)) {
 +fprintf (stderr, additional privileges are required\n);
 +exit (1);
 +}
 +
 +if (priv_set (PRIV_OFF, PRIV_ALLSETS, PRIV_FILE_LINK_ANY, PRIV_PROC_INFO,
 +PRIV_PROC_SESSION, PRIV_PROC_EXEC, PRIV_PROC_FORK, NULL)) {
 +fprintf (stderr, failed to set reduced privileges\n);
 +exit (1);
 +}
 +}
 +#else
 +#define qemudSetupPrivs(a)
 +#endif

  
  /* Print command-line usage. */
  static void
 @@ -2417,6 +2510,8 @@ int main(int argc, char **argv) {
  sig_action.sa_handler = SIG_IGN;
  sigaction(SIGPIPE, sig_action, NULL);
  
 +qemudSetupPrivs(server);
 +
  if (!(server = qemudInitialize(sigpipe[0]))) {


This would appear to make it try to change the socket ownership
and permissions, before we've actually created the socket, which
is much later in the main() method where we call NetworkInit


 diff --git a/qemud/remote.c b/qemud/remote.c
 --- a/qemud/remote.c
 +++ b/qemud/remote.c
 @@ -424,6 +424,15 @@ remoteDispatchOpen (struct qemud_server 
  flags = args-flags;
  if (client-readonly) flags |= VIR_CONNECT_RO;
  
 +#ifdef __sun
 +/*
 + * On Solaris, all clients are forced to go via virtd. As a result,
 + * virtd must indicate it

Re: [libvirt] [PATCH] Introduce virt-console

2009-01-15 Thread Richard W.M. Jones

On Wed, Jan 14, 2009 at 07:21:10AM -0800, john.le...@sun.com wrote:
 +#ifdef __sun

I'm really unhappy about adding #ifdef __sun everywhere in this file.
Configure should detect the features that are needed / available /
missing, and the #ifdefs should go on the result of that.

Rich.

-- 
Richard Jones, Emerging Technologies, Red Hat  http://et.redhat.com/~rjones
Read my OCaml programming blog: http://camltastic.blogspot.com/
Fedora now supports 68 OCaml packages (the OPEN alternative to F#)
http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora

--
Libvir-list mailing list
Libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] Compiling on Windows with MinGW/MSYS

2009-01-15 Thread Richard W.M. Jones

On Thu, Jan 15, 2009 at 11:21:51AM +0100, Brecht Sanders wrote:
Hi,
It's been a long time since I tried to compile libvirt on Windows using
the MinGW/MSYS environment.
I tried it again and I have good news and bad news.
The good news is thet when I disable support for most things
(--without-xen --without-qemu --without-openvz --without-uml
--without-test --without-libvirtd --with-remote-pid-file=none
--with-init-scripts=none --with-depends --without-sasl --without-polkit)
it now builds fine and even the DLL was built (libvirt-0.dll).
The bad news is that when I enable Xen support, it now depends on the
xenstore library.
Unfortunately I wasn't able to build this library on Windows.
Is this dependancy necessary?
Does anyone know of xenstore on Windows?

You're well outside the envelope of what we've tried or what we
support. On Windows we only support client connections through the
remote driver, so you have to disable everything else:

http://hg.et.redhat.com/cgi-bin/hg-misc.cgi/fedora-mingw--devel/file/61e54817b85c/libvirt/mingw32-libvirt.spec#l48

I'm confused actually as to what use libvirt + Xen would be on Windows
-- does Xen Enterprise now have the ability to use Windows as dom0?

Rich.

--
Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones
virt-df lists disk usage of guests without needing to install any
software inside the virtual machine. Supports Linux and Windows.
http://et.redhat.com/~rjones/virt-df/

81 matches

Mail list logo