- Devel - libvirt List Archives

[Libvir] man page for virsh
by andrew puch 11 Apr '06

11 Apr '06

Enclosed is the man and the html page that were generated from POD Plain Old Doc from perl. pod2man virsh.pod.1 > virsh.1 pod2html virsh.pod.1 > virsh.html Behold the power of perl :)

3 3

[Libvir] autogen.sh error
by Ronald Aigner 10 Apr '06

10 Apr '06

Hi all, I downloaded libvirt from CVS and tried to run autogen.sh which stops with the following error message: configure: error: Xen store library not found What else do I need? How do I implement support for another hypervisor? Where do I have to turn which screws? Thanks, Ron.

2 4

[Libvir] Release of libvirt-0.1.0
by Daniel Veillard 10 Apr '06

10 Apr '06

I mostly finished the big internal code revamp over the week-end, this with the addition of the latest contributions makes it a good time for a new release, I also bumped up the medium release numbers, with the cleanup in the code internals I feel better now having people relying on libvirt, and while we are still far away from an 1.0.0 release, it's important to indicate the change :-) Available at the usual place i.e. ftp://libvirt.org/libvirt/ * building fixes: - --with-xen-distdir option (Ronald Aigner) - out of tree build and pkginfo cflag fix (Daniel Berrange) * enhancement and fixes of the XML description format (David Lutterkort and Jim Fehlig) * new APIs: - for Node information - Reboot * internal code cleanup: - refactoring internals into a driver model - more error handling - structure sharing - thread safety and ref counting * bug fixes: - error message (Jim Meyering) - error allocation in virsh (Jim Meyering) - virDomainLookupByID (Jim Fehlig) * documentation: - updates on architecture, and formatr - typo fix (Jim Meyering) * bindings: - exception handling in examples (Jim Meyering) - perl ones out of tree (Daniel Berrange) * virsh: - more options, create, nodeinfo (Karel Zak) - renaming of some options (Karel Zak) - use stderr only for errors (Karel Zak) - man page (Andrew Puch) Thanks to everybody who contributed to this release ! Daniel -- Daniel Veillard | Red Hat http://redhat.com/ veillard(a)redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/

1 0

[Libvir] Re: virsh outputs to stderr(?!) why not stdout
by Karel Zak 06 Apr '06

06 Apr '06

On Thu, Apr 06, 2006 at 08:40:17AM -0400, Daniel Veillard wrote: > On Thu, Apr 06, 2006 at 02:37:37PM +0200, Karel Zak wrote: > > On Thu, Apr 06, 2006 at 11:46:57AM +0200, Jim Meyering wrote: > > > So, now I can use something like this to get a list of domain names: > > > > > > virsh list | tail -n +4 | awk '{print $2}' > > > > > > Is there some easier way? > > > > I think we can add some options to the list command: > > > > --noheader > > --nodom0 > > --cols=name,id,state,maxmem,usemem,.... > > > > without these options it will use default (current) format. I've > > already thought about this solution for the others commands (e.g. > > dominfo). I think this is a way how make it useful in scripts. > > Hum, why not discussing this on the libvir list ? Others may have > soem ideas too and this may promote virsh usage a bit ;-) Fixed :-) -- Karel Zak <kzak(a)redhat.com>

2 3

Re: [Libvir] BUG: virDomainLookupByID fails after XenStore path discovery
by Jim Fehlig 05 Apr '06

05 Apr '06

>>> "Bryan D. Payne" <bryan(a)thepaynes.cc> 04/04/06 6:23 PM >>> >> I'm just not seeing it included in path returned by >> xs_get_domain_path(), only see /local/domain/<num> > For what it's worth, I was trying to put a patch together for this, >but I kept getting memory corruption errors. I was trying to access >the name through the xenstore as shown below... > > /* lookup is easier with the Xen store so try it first */ > if (conn->xshandle != NULL) { > char *namepath = NULL; > path = xs_get_domain_path(conn->xshandle, (unsigned int) id); > namepath = strdup(path); > strcat(namepath, "/name"); > name = xs_read(conn->xshandle, NULL, namepath, NULL); > if (namepath) free(namepath); > } My brain is not working today :-). Don't know why I didn't consider xs_read(). I'll fix up your patch and test/submit it tomorrow. I think the only problem is concatenating "/name" on the end of namepath, which has only been allocated enough memory to hold path via strdup(). Thanks, Jim

2 1

[Libvir] BUG: virDomainLookupByID fails after XenStore path discovery
by Bryan D. Payne 05 Apr '06

05 Apr '06

I just updated to the latest CVS version of libvir. It looks like there's a bug in virDomainLookupByID. The behavior that I'm seeing is that the path variable is set and so the code never enters the block to find the value for name. Therefore, at the end of the function, the condition (ret->name == NULL) is true and it jumps into the error code and returns NULL. If I comment out the code that sets the path variable, things seem to work normally. Although, I presume that the idea was to extract the name from the path information... -bryan - Bryan D. Payne Graduate Student, Computer Science Georgia Tech Information Security Center http://www.bryanpayne.org

3 5

[Libvir] Change of server
by Daniel Veillard 03 Apr '06

03 Apr '06

So far libvirt.org was hosted on my home server (IP 82.67.66.12), but I really needed to fix this as my home ADSL connectivity isn't that great. I moved all services to a new server (colocated on the same box as xmlsoft.org in a public research institute here in France), hopefully the transition to this new server (IP 194.199.20.115) will be transparent as DNS changes propagates, but some of you mays still access the old services for a little while. Tell me if you experience troubles with the new server, and if you don't see updates from libvirt.org CVS that may be related still using the old server. Hopefully bandwidth and relibaility should be quite higher now, thanks, Daniel -- Daniel Veillard | Red Hat http://redhat.com/ veillard(a)redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/

1 0

[Libvir] Patch to make reading of HTTP response more efficient
by Daniel P. Berrange 30 Mar '06

30 Mar '06

When processing the header of HTTP responses from Xend, libvirt currently reads data a single byte at a time. Since its using lowlevel UNIX read() function there is no buffering, which results in a large number of system calls. The attached patch re-factors the code so that it reads data in 4k chunks whenever possible. Since most rsponses are < 4k in size processing the response now only takes two read() syscalls, rather than several hundred. Of course most (90%) of the CPU overhead is still on the Xend/XenStored end of the channel, but that's out of our control until the XML-RPC API comes along. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|

2 1

[Libvir] unchecked `malloc' calls
by Jim Meyering 30 Mar '06

30 Mar '06

Hello, In browsing through the code, I noticed that there are many places in which the code doesn't check for malloc returning NULL. E.g., ./virsh.c: ids = malloc(sizeof(int) * maxid); ./virsh.c- virConnectListDomains(ctl->conn, &ids[0], maxid); ./virsh.c: *res = malloc(sz + 1); ./virsh.c- memcpy(*res, tkstr, sz); ./virsh.c: res = malloc(strlen(name) + 3); ./virsh.c- sprintf(res, "--%s", name); ./xmlrpc.c: buf = malloc(sizeof(*buf)); ./xmlrpc.c- buf->size = 1024; ./xmlrpc.c: xmlRpcValueDictElementPtr ret = malloc(sizeof(*ret)); ./xmlrpc.c- xmlNodePtr cur; ./xmlrpc.c- ./xmlrpc.c- memset(ret, 0, sizeof(*ret)); At first, seeing so many, I thought there must be a wrapper that handles the case of a NULL return value. But there are many places where the code *does* check for NULL. This may be a clue. Maybe it's planned... ./libvirt.c: * TODO: ... ./libvirt.c: * - memory wrappers for malloc/free ? Jim

3 3

[Libvir] RFC: requesting APIs for host physical resource discovery
by Daniel P. Berrange 29 Mar '06

29 Mar '06

The current libvirt APIs allow the host's physical resources to be split up and allocated to guest domains, however, there is no way to discover what the available host resources actually are. Thus I would like to suggest the inclusion of new APIs to enable host resource discovery. As a starting point I'd like to be able to query the following information: * Number of physical CPUs - ability to enumerate the CPUs in the host, both those currently present, and theoretical maximum (to take account of hotplug). * Amount of RAM - actual physical RAM present, and that available for guest usage (eg discounting that reserved by a hypervisor or equiv) * CPU relationship - ie ability to distinguish between CPUs which are hyperthread siblings, on same core, or on separate sockets Alonside these basic queries it would be desirable to add a further resource resource management API to allow for setting of a guest domain's CPU affinity. ie ability control what CPUs the VMM is allowed to schedule a domain on. Once this first basic set of caapbilties for resource discovery are provided for, then I believe it will be neccessary to provide some more advanced queries, in particular: * NUMA topology - ability to enumerate NUMA nodes, the CPUs associated with each node & the RAM range mapped to that node Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|

3 3