- Devel - libvirt List Archives

[Libvir] [PATCH] Add virDomainGetConnect and virNetworkGetConnect
by Richard W.M. Jones 25 Jun '07

25 Jun '07

As described in this thread: https://www.redhat.com/archives/libvir-list/2007-June/thread.html#00352 * These functions don't change the reference count for virConnectPtr. It's be a pain in the neck for callers otherwise. * I haven't changed the Python bindings, because I'm not confident enough that I can do that correctly and keep it backwards compatible for all callers (and anyway what would happen if newer Python bindings were used together with an older libvirt?). However I did add a comment to the generator noting that this could be changed in future. * I tested this with the OCaml bindings with no apparent ill-effects. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 2

[Libvir] [PATCH] Cleanup - remove virDomainKernel structure
by Richard W.M. Jones 25 Jun '07

25 Jun '07

... unless anyone can remember what it was for. It still builds fine without it. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

[Libvir] OCaml bindings for libvirt
by Richard W.M. Jones 25 Jun '07

25 Jun '07

Weekend project: http://et.redhat.com/~rjones/ocaml-libvirt/ Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

3 2

[Libvir] Useful additional functions for libvirt
by Richard W.M. Jones 25 Jun '07

25 Jun '07

[I haven't implemented any of these yet, but if people think they're a good idea, or at least not an actively bad idea, then I'll post a patch.] virDomainGetConnection virNetworkGetConnection Purpose: Given a virDomainPtr or virNetworkPtr, obtain the virConnectPtr. Reason: All the language bindings to libvirt need to keep the virConnectPtr separately alongside the virDomain/NetworkPtr, in the main so that they can query if an error has happened from inside some deep call. This is wasteful since the connection pointer is already included in the virDomain or virNetwork structure, so we should just provide a call to get it. virConnectGetHostname virConnectGetTransport virConnectGetURI Purpose: Get the remote hostname, remote transport (tls, ssh, etc.), and URI. Reason: In virt-manager it would be nice to display the remote hostname. However doing this at the moment requires parsing the connection URI, which is duplicated code and also significantly complicated. Instead, allow the remote driver to just give us this information, and in non-remote cases default to something sensible. The case for the other two calls is weaker, but it might still be useful to know something about the security of the actual transport selected, and also to not have to keep the URI around with the connection (we might also canonicalise the transport for the user). virConnectPing Purpose: "Ping" the hypervisor to see if its up. Reason: Since we now support remote connections, there is a much more signficant chance that we will lose contact with the hypervisor, for example if the host goes down. This will do some very minimal operation to cheaply test whether the hypervisor can be contacted. Of course we could do something like 'virConnectNumOfDomains', but it's not clear to me that this operation would always be cheap (eg. if we had to implement it through xend). Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

3 2

[Libvir] [PATCH] Allow maxnames to be 0 in List{, Defined}{Domains, Networks}
by Richard W.M. Jones 25 Jun '07

25 Jun '07

It seems a bit pathological to check if maxNames <= 0 and throw an error if it is 0. In particular if you don't have any defined domains or networks (both fairly common scenarios) then this causes an error when you just pass the result of NumOfNetworks straight into ListNetworks. Backend drivers & QEMU seem to have no problem with maxNames being zero, so let's allow it. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

[Libvir] [PATCH] Fix virDomainGetMaxVcpus in remote case
by Richard W.M. Jones 25 Jun '07

25 Jun '07

There's a typo in the remote_internal.c driver which causes it to call the wrong back-end function for virDomainGetMaxVcpus. The simple fix is attached. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

[Libvir] [PATCH] Fix virDomainGetVcpus in the remote case
by Richard W.M. Jones 25 Jun '07

25 Jun '07

I wasn't copying enough data in the cpumaps parameter. This fixes that bug ... Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

[Libvir] can not use the monitor of Qemu-dm
by Saori Fukuta 25 Jun '07

25 Jun '07

Hi, I opened the new bugzilla: Summary: can not use the monitor of Qemu-dm URL: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=245284 And I attached the patch to fix this problem. Thanks, Saori Fukuta

4 5

[Libvir] [PATCH] Xen driver doesn't support autostart, so remove it from xen_unified
by Richard W.M. Jones 22 Jun '07

22 Jun '07

Our implementation of the Xen driver doesn't support autostart[*]. However xen_unified.c has a loop for get/setAutostart which causes it to return an error without setting virterror first. This patch removes that loop. Rich. [*] Although Xen supports it, by putting files in /etc/xen/autostart/ (at least that was the "old Xen" way before the current insanity of getting rid of config files was invented). -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

3 2

[Libvir] Avoid leak upon failed realloc
by Jim Meyering 22 Jun '07

22 Jun '07

This matters only if/when this realloc call fails. And then, only in the small way that with this patch, there's a slightly better chance of recovering from the low-memory condition. 2007-06-22 Jim Meyering <jim(a)meyering.net> * qemud/driver.c (qemudMonitorCommand): Avoid leak upon failed realloc. Index: qemud/driver.c =================================================================== RCS file: /data/cvs/libvirt/qemud/driver.c,v retrieving revision 1.21 diff -u -p -r1.21 driver.c --- qemud/driver.c 15 Jun 2007 13:44:19 -0000 1.21 +++ qemud/driver.c 22 Jun 2007 09:43:59 -0000 @@ -75,6 +75,7 @@ int qemudMonitorCommand(struct qemud_ser for (;;) { char data[1024]; int got = read(vm->monitor, data, sizeof(data)); + char *b; if (got == 0) { if (buf) @@ -91,8 +92,11 @@ int qemudMonitorCommand(struct qemud_ser free(buf); return -1; } - if (!(buf = realloc(buf, size+got+1))) + if (!(b = realloc(buf, size+got+1))) { + free(buf); return -1; + } + buf = b; memmove(buf+size, data, got); buf[size+got] = '\0'; size += got;

3 3

[Libvir] [PATCH] Add support for scheduler functions in remote
by Richard W.M. Jones 22 Jun '07

22 Jun '07

This is a rather tedious patch, but I've tested it against the Xen driver and it works for getting and setting the various parameters and scheduler type. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 3

[Libvir] [PATCH] Fix string handling in virDomain{Get, Set}SchedulerParameters
by Richard W.M. Jones 22 Jun '07

22 Jun '07

This small patch fixes some bugs in the handling of the field string in virDomainGetSchedulerParameters and makes a similar pre-emptive fix to virDomainSetSchedulerParameters. Also, please don't use !strcmp(a,b), because it confuses me. Better is to write strcmp(a,b) == 0 to mean "strings match" and strcmp(a,b) != 0 to mean "strings don't match". Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

4 10

[Libvir] [PATCH] Fix remote virDomainGetXMLDesc
by Richard W.M. Jones 21 Jun '07

21 Jun '07

I was wondering why dumpxml didn't work on remote connections. Turned out to be the same bug that Dan found last week: https://www.redhat.com/archives/libvir-list/2007-June/msg00186.html The attached patch includes that fix, plus it adds proper error handling into xend_internal.c:xenDaemonDomainDumpXML and related functions. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 2

[Libvir] [PATCH] Fix remote virConnectGetMaxVcpus
by Richard W.M. Jones 21 Jun '07

21 Jun '07

I've been going over the interface to libvirt, trying out every call locally and remotely with all possible combinations of parameters that I can think of. Not surprisingly one or two problems have cropped up. So here's a short series of patches to fix the problems I've found. virConnectGetMaxVcpus has a second "type" parameter. In practice this isn't used and can be passed as NULL. However that fails in the remote case. The fix is simple enough -- attached. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

[Libvir] [PATCH] Correct documentation for virConnectOpen*
by Richard W.M. Jones 20 Jun '07

20 Jun '07

Simple patch to fix the currently incorrect documentation for these two functions. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

[Libvir] RFC: replace "no support for hypervisor" error
by Richard W.M. Jones 20 Jun '07

20 Jun '07

The "no support for hypervisor" error is both very common and pretty annoying because it gives you nowhere to go to find out what you did wrong. $ virsh -c xen://localhost/ libvir: Xen Daemon error : no support for hypervisor virsh: error: failed to connect to the hypervisor (The actual problem is that my URI is wrong, which is a separate bug in the Xen driver. But the error message gives me no particular indication of what is wrong or how to correct the situation). I think there are several possible solutions to this - discuss, or suggest your own. (1) Document the URI formats fully. -- This is a feature which has been requested a few times and I will do it anyway if Daniel Veillard will do the magic to set up a "http://libvirt.org/uri.html" page. (2) Add a method to query available drivers and URI syntax. -- Something along the lines of char *virQueryDrivers (void); Note that it doesn't need an open connection. -- The remote case would have to be handled specially because you want to find out what's available locally and what's available remotely, and in the remote case you do need some sort of a connection. (3) Split VIR_ERR_NO_SUPPORT into two categories. Currently this category mixes up cases where we fail to open a connection, and cases where there is no driver support for a particular operation (even with an open connection). In the first case, go through all the places which return this error and add proper diagnostic information to the error messages. -- Again, I am prepared to do this if people think it's a good idea. (4) Add diagnostics to higher layers such as virsh and virt-manager. -- I would be less happy with this because it ends up repeating code, and the diagnostics could get out of date w.r.t. what libvirt can do. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

3 11

[Libvir] Re: URI documentation and xen:/// patch
by Richard W.M. Jones 19 Jun '07

19 Jun '07

Daniel Veillard wrote: > On Tue, Jun 19, 2007 at 05:15:40PM +0100, Richard W.M. Jones wrote: >> Daniel Veillard wrote: >> $ virsh -c ///var/lib/xend/xend-socket list >> libvir: error : no support for hypervisor >> virsh: error: failed to connect to the hypervisor > > Hum, I not sure I ever tried to get that working. Maybe > file:///var/lib/xend/xend-socket > would work better, though libxml2 URI parsing should be able to > parse ///var/lib/xend/xend-socket and make it more or less equivalent > to the file:/// one. I would strace both to check. Yeah, I tried all the obvious ones like that. I'll run it under gdb later & find out what's really going on. I was just pointing out that it didn't seem to work right now ... Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

1 0

[Libvir] [PATCH] hypervisor version
by Mark Johnson 19 Jun '07

19 Jun '07

This is another patch which may not be popular? Xen's extra version does not fit in libvirt's release field (since it's part of an int). Instead of printing out the wrong value, just display major.minor in virsh. Mark

4 12

[Libvir] [PATCH] Fix the check of <cpumap> in virsh vcpupin
by Masayuki Sunou 19 Jun '07

19 Jun '07

Hi Because virsh vcpupin does not check a form of <cpumap>, when non-numerical letters are set, it does not become an error. This patch fixes it so that it become an error when non-numerical letters are set. Signed-off-by: Masayuki Sunou <fj1826dm(a)aa.jp.fujitsu.com> Thanks, Masayuki Sunou. ---------------------------------------------------------------------- Index: src/virsh.c =================================================================== RCS file: /data/cvs/libvirt/src/virsh.c,v retrieving revision 1.84 diff -u -p -r1.84 virsh.c --- src/virsh.c 15 Jun 2007 15:24:20 -0000 1.84 +++ src/virsh.c 18 Jun 2007 07:02:31 -0000 @@ -1530,6 +1530,7 @@ cmdVcpupin(vshControl * ctl, vshCmd * cm int vcpufound = 0; unsigned char *cpumap; int cpumaplen; + int i; if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; @@ -1563,6 +1564,22 @@ cmdVcpupin(vshControl * ctl, vshCmd * cm return FALSE; } + for(i = 0; cpulist[i]; i++) { + if ((i == 0) || (i == strlen(cpulist)-1)) { + if (!isdigit(cpulist[i])) { + vshError(ctl, FALSE, _("Invalid format '%s'."), cpulist); + virDomainFree(dom); + return FALSE; + } + } else { + if(!isdigit(cpulist[i]) && cpulist[i] != ',') { + vshError(ctl, FALSE, _("Invalid format '%s'."), cpulist); + virDomainFree(dom); + return FALSE; + } + } + } + cpumaplen = VIR_CPU_MAPLEN(VIR_NODEINFO_MAXCPUS(nodeinfo)); cpumap = vshCalloc(ctl, 1, cpumaplen); ----------------------------------------------------------------------

3 7

[Libvir] [RFC] Device attach/detach on virsh
by Masayuki Sunou 19 Jun '07

19 Jun '07

I want to add I/F to do attach/detatch of VIF and VBD to virsh with virDomainAttachDevice()/virDomainDetachDevice(). And, I have two proposals about I/F for virsh to do attach/detach of VIF and VBD. proposal 1: Virsh catches MAC, bridge name, device name (physical,virtual), and another by the command option. ex) ------------------------------------------------------------------ # virsh help attach(detach)-vif(vbd) NAME attach(detach)-vif(vbd) - attach(detach) vif(vbd) SYNOPSIS * VIF attach(detach)-vif <domain> <MAC> <bridge> ... or * VBD attach(detach)-vbd <domain> <virt-dev> ... DESCRIPTION Attach(Detach) vif(vbd) device OPTIONS <domain> domain name, id or uuid * VIF <MAC> MAC address of vif <bridge> bridge name of vif ... * VBD <virt-dev> virtual device name of vbd <phy-dev> physical device name of vbd ... ------------------------------------------------------------------ <advantage> - I/F is easy to use than proposal 1. (Because the user does not have to make XML) <disadvantage> - I/F increases because I/F of VIF and VBD becomes separate. (So, the addition of I/F is necessary at any time for supporting devices other than VIF and VBD. ) - Handling of optional analysis, handling of XML making are necessary in virsh.c, and processing becomes complicated. proposal 2: virsh catches a definition of a device by XML ex) ------------------------------------------------------------------ # virsh help attach(detach)-device NAME attach(detach)-device - attach(detach) device from an XML file SYNOPSIS attach(detach)-device <domain> <file> DESCRIPTION Attach(Detach) device from an XML <file> OPTIONS <domain> domain name, id or uuid <file> XML file of device description ------------------------------------------------------------------ <advantage> - I/F is unified without affecting a device type. (I/F is simple) - Handling of virsh.c is simple. (Optional analysis is not necessary) <disadvantage> - The user has to describe XML.(It is troublesome) I think that specifications that a user is easy to use (proposal 1) are better. Please, give me an opinion which proposal is better. Thanks, Masayuki Sunou

6 28

[Libvir] PATCH 0/3: Refactor QEMU daemon
by Daniel P. Berrange 18 Jun '07

18 Jun '07

The 3 patches which follow are work-in-progress to re-factor the QEMU daemon / driver to eventually adhere to the main libvirt internal driver API. Once this work is complete, there will only need to be a single daemon running which can provide both remote & QEMU capabilities at once with no QEMU specific code in it. This is following some of the ideas I set out here http://www.redhat.com/archives/libvir-list/2007-May/msg00083.html 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 36

[Libvir] Register libvirtd ports with IANA ?
by Daniel P. Berrange 18 Jun '07

18 Jun '07

For the libvirtd we currently use two ports 16509 - TCP unencrypted stream 16514 - TLS encrypted stream My first thought is that we should really use consequetive port numbers eg 16510 and 16511. Second, we should register our port numbers with IANA http://www.iana.org/protocols/forms.htm Thoughts ? 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 7

[Libvir] PATCH: Fix marshalling of ID on wire with remote
by Daniel P. Berrange 18 Jun '07

18 Jun '07

THe code which marshalls virDomainptr objects onto the wire in the remote driver forgets to include the 'id' value. We get away with this for Xen and QEMU because they do most things based off name or UUID. The test driver however wants the ID number. The attached patch fixes this 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] [PATCH] setmem checks
by Mark Johnson 18 Jun '07

18 Jun '07

This one may be a little contentious... This patch does a couple of things... It won't let you set memory if the value is greater that maxmem. It also will lower memory if max memory is set to less than memory. I think it's a good argument that this checking should be in the hypervisor control daemon. IMO, it should be both here and in the daemon. I have no idea why the patch diff is so big, Below is the diff after it is applied. A little easier to read. Mark [root@fedora libvirt]# diff src/virsh.orig src/virsh.c 1664a1665 > virDomainInfo info; 1673a1675,1677 > if (virDomainGetInfo(dom, &info) != 0) { > info.maxMem = 0; > } 1675c1679 < if (kilobytes <= 0) { --- > if ((kilobytes <= 0) || (kilobytes > info.maxMem)) { 1708a1713 > virDomainInfo info; 1723a1729,1737 > if (virDomainGetInfo(dom, &info) != 0) { > info.memory = 0x7fffffff; > } > > if (kilobytes < info.memory) { > if (virDomainSetMemory(dom, kilobytes) != 0) { > return FALSE; > } > } [root@fedora libvirt]#

5 12

[Libvir] [PATCH] header file changes for Solaris
by Mark Johnson 16 Jun '07

16 Jun '07

This patch has the includes need to build on Solaris. I've been using ifdef linux & ifndef linux to distinguish between solaris and linux at this point. I'm not sure what the preference is... MRJ

5 20

[Libvir] virt-manager bug
by Mark Johnson 15 Jun '07

15 Jun '07

Not sure if this is the right alias... This fixes virt-manager -c test:///default : p5b-vm[1]#; diff /tmp/connection.orig /usr/share/virt-manager/virtManager/connection.py 194,195d193 > else: > return self.get_type() + ":" + self.get_hostname() def get_name(self): if self.get_type()[0:3] == "Xen": return "Xen: " + self.get_hostname() elif self.get_type() == "QEMU": if self.uri == "qemu:///session": return "QEMU session: " + self.get_hostname() else: return "QEMU system: " + self.get_hostname() else: return self.get_type() + ":" + self.get_hostname()

2 1

[Libvir] patch status
by Mark Johnson 15 Jun '07

15 Jun '07

Ok, just a quick check so I can get my head on straight... :-) This patch was withdrawn.. [PATCH] return 0 networks if Network driver not implemented These patches have been applied [PATCH] Spelling fix [PATCH] null pointer fix on debug build [PATCH] setvcpus message [PATCH] vcpuinfo message [PATCH] _N() has a conflict in Solaris This patch looks like it's about to be applied (with some addition changes from Rich) [PATCH] header file changes for Solaris And these three are still being hashed out. [PATCH] setmem checks [PATCH] hypervisor version [PATCH] Solaris dom0 support Mark

2 1

[Libvir] [PATCH] Solaris dom0 support
by Mark Johnson 15 Jun '07

15 Jun '07

This is a little bigger of a patch. It has a couple of things in it. First, in xen_internal.c and xs_internal.c have general dom0 support for Solaris. Again, using #ifdef/ifndef __linux__ to separate the logic. In xml.c and xend_internal.c, the code which requires kernel to be defined if ifdef __linux__'d. For our setup, If bootloader, kernel, and ramdisk aren't defined, pygrub will got looking for the right bits. There also code which puts in the the vncpasswd in xend_internal.c. This is the last patch I'll send out for today... Mark

5 15

[Libvir] libvirt ./configure
by Mark Johnson 15 Jun '07

15 Jun '07

Just thought I'd throw a note to you.. I notice that if I do a configure on fedora7 as a user I get the following... checking If XenD UNIX socket /var/run/xend/xmlrpc.sock is accessible... no Which works when you do a configure as root. Not sure what your using that for so not sure if it's a problem.. Mark

2 1

[Libvir] Patch to get rid of index(3) and rindex(3)
by Richard W.M. Jones 15 Jun '07

15 Jun '07

According to index(3): [index and rindex appeared in] 4.3BSD; marked as LEGACY in POSIX.1-2001. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

3 3

[Libvir] Storage manager initial requirements and thoughts
by Hugh Brock 15 Jun '07

15 Jun '07

Hi all. At the Red Hat virtualization team meeting last week we spent some time talking about the problem of remote storage management, which is a requirement for at the very least creating remote guests. Remote storage management also delivers a lot of benefits for managing guest storage, provisioning, and a host of other issues. Note that we don't (I think) believe that storage management should be part of libvirt. However we might want to be able to grab storage information with libvirt when creating a guest, for example. Here's the tentative requirements we came up with, along with several questions that came up. We have two important use cases for remote storage management: --Create a new guest against an existing physical device --Create a new guest against a file on an existing physical device (SELinux context must match up on both) Requirements: 2 key requirements: enumerate devices, create files * Enumerate devices (unallocated disks) * storage pool * Unallocated space * Allocated volumes -- free -- in use * Read only / Read-write * Host availability * Volume -- Global unique name -- Local device name -- Usage -- Mounted locally -- Assigned to guest read only shared exclusive -- Inactive guest * Create a backing store for a guest * need to know what storage is available * Can be plain file * Can be a physical partition (local scsi/IDE, SAN) * Can be a logical volume * Can be network - iscsi/nbd Cases: Physical device --> create partitions Volume group --> create logical volumes Directory --> create files Todos: Investigate gparted, one of the partition management tools we already have (apis? remote accessibility?) (I believe Jim Meyering volunteered to take a look at this?) Identify what other scenarios we need to address. Please comment; I'm hoping we can get these into a more definite form sooner rather than later. -- Red Hat Virtualization Group http://redhat.com/virtualization Hugh Brock | virt-manager http://virt-manager.org hbrock(a)redhat.com | virtualization library http://libvirt.org

10 43

[Libvir] [PATCH] _N() has a conflict in Solaris
by Mark Johnson 15 Jun '07

15 Jun '07

There is a conflict with _N() in Solaris. Mark

3 4

[Libvir] [PATCH] setvcpus message
by Mark Johnson 15 Jun '07

15 Jun '07

This patch is probably not correct. I wasn't sure how to handle I18N in libvirt. Had a request to add an error message for setvcpus <domain> 0. This patch prints an error message for that case. Mark

3 2

[Libvir] [PATCH] return 0 networks if Network driver not implemented
by Mark Johnson 15 Jun '07

15 Jun '07

This handles net-list, etc. a little more gracefully for Xen. Mark

3 3

[Libvir] Solaris patches applied
by Richard W.M. Jones 15 Jun '07

15 Jun '07

I applied the easy ones: * src/internal.h, src/virsh.c: Replace _N with N_ so that Solaris builds work (Mark Johnson). * src/virsh.c: Add a couple of missing error messages (Mark Johnson). * python/types.c: Fix NULL pointer deref on DEBUG build (Mark Johnson). * src/virsh.c: Spelling fix (Mark Johnson). I can't test if it works on Solaris, but it still builds on Linux. Now to look at the last patch you sent ... Thanks for your contribution to libvirt. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

1 0

[Libvir] [PATCH] vcpuinfo message
by Mark Johnson 15 Jun '07

15 Jun '07

Another requested error message. This prints out a message if the vcpuinfo failed because the domain was shutdown. same comment as previous patch about I18N. Mark

2 1

[Libvir] [PATCH] Spelling fix
by Mark Johnson 15 Jun '07

15 Jun '07

Hi all, This is the first of 10 or 11 patches I'll be submitting which gets libvirt compiling/working on a Solaris dom0. Also some minor generic bug fixing.. Sorry it took me so long from the Xen Summit.. Starting with a very simple one. :-) As I test them on the cvs tree on Fedora, I'll submit them.. So they'll trickle in over the next couple of days. A couple of them will probably require some discussion and changes. I'm applying the patches with the following # patch -t -s -p1 -i ../patches/spelling on todays CVS gate and tested on a fedora 7 dom0 (LD_PRELOAD=src/.libs/libvirt.so src/.libs/virsh) Thanks, Mark

3 2

[Libvir] [PATCH] null pointer fix on debug build
by Mark Johnson 15 Jun '07

15 Jun '07

this fixes a null pointer dereference in printf on a DEBUG build. Mark

2 1

[Libvir] [PATCH] fix the Japanese messages
by Saori Fukuta 14 Jun '07

14 Jun '07

Hi, We are making a review of the Japanese messages. And we fix Japanese help messages first. Signed-off-by: Saori Fukuta <fukuta.saori(a)jp.fujitsu.com>, Minoru Takata <takata.minoru(a)jp.fujitsu.com> Best regards, Saori Fukuta Minoru Takata

4 5

[Libvir] no vfb for paravirt guest on RHEL 5 and strange args
by Glen Deem 13 Jun '07

13 Jun '07

Hi all, I'm creating a fairy simple paravirt machine under RHEL 5 (Xen 3.0.3, pygrub 0.6). A machine gets created but I see two problems: 1. no graphics console gets enabled (so RedHat VMM does not open it) so I cannot connect to it with vncviewer (am I naive to use vfb fpr a paravirt domU? - libvirt uses it, right?); 2. when I check what config has been used to create the domain (in /var/log/xen/xend.log) I see arguments ("args" or "extra") picked up from a virtual machine created a while age, my present "extra" parameter is ignored. I did regenerate xenstored/tdb and rebooted, still those args appear, drives my nuts :-). I'm actually after the graphics console, but I think both problems are related. Could anybody advise why I get no proper vfb setup (I'd expect a vnc-port vaue appearing in the xen store when it setup properly) and why I get those old args? Thanks in advance, Deem P.S. xm config ============ name = "para-2" memory = "256" disk = [ 'file:/var/lib/para-2.img,hda,w' ] vif = [ 'bridge=xenbr0' ] vfb = ['type=vnc, vncunused=1'] bootloader='/usr/bin/pygrub' vcpus=1 on_reboot = 'restart' on_crash = 'restart' extra = 'ignored' from /var/log/xen/xend.log: ================== [2007-06-12 17:13:56 xend.XendDomainInfo 5445] DEBUG (XendDomainInfo:940) XendDomainInfo.handleShutdownWatch [2007-06-12 17:41:57 xend.XendDomainInfo 5445] DEBUG (XendDomainInfo:190) XendDomainInfo.create(['vm', ['name', 'para-2'], ['memory', '256'], ['on_reboot', 'restart'], ['on_crash', 'restart'], ['vcpus', 1], ['bootloader', '/usr/bin/pygrub'], ['image', ['linux', ['ramdisk', '/var/lib/xen/initrd.NMeE1x'], ['kernel', '/var/lib/xen/vmlinuz.7QoaTW'], ['args', 'nosplash quiet rw acpi=on XXHOST=localhost.localdomain XXLDAP=127.0.0.1 resume2=swap:/dev/mapper/XxxxxxVG-swap vga=791 root=/dev/XxxxxxxVG/root']]], ['device', ['vbd', ['uname', 'file:/var/lib/para-2.img'], ['dev', 'hda'], ['mode', 'w']]], ['device', ['vif', ['bridge', 'xenbr0']]], ['device', ['vkbd']], ['device', ['vfb', ['vncunused', '1'], ['type', 'vnc'], ['display', ':0.0'], ['xauthority', '/root/.xauthfLRNzX']]]])

1 0

[Libvir] New vs old-style hypercalls
by Richard W.M. Jones 12 Jun '07

12 Jun '07

In xen_internal.c we've got code which claims to sniff out whether we are using new-style hypercall ioctls or old-style hypercall ioctls. hc.op = __HYPERVISOR_xen_version; hc.arg[0] = (unsigned long) XENVER_version; hc.arg[1] = 0; cmd = IOCTL_PRIVCMD_HYPERCALL; ret = ioctl(fd, cmd, (unsigned long) &hc); if ((ret != -1) && (ret != 0)) { #ifdef DEBUG fprintf(stderr, "Using new hypervisor call: %X\n", ret); #endif hv_version = ret; xen_ioctl_hypercall_cmd = cmd; goto detect_v2; } /* * check if the old hypercall are actually working */ v0_hc.op = __HYPERVISOR_xen_version; v0_hc.arg[0] = (unsigned long) XENVER_version; v0_hc.arg[1] = 0; cmd = _IOC(_IOC_NONE, 'P', 0, sizeof(v0_hypercall_t)); ret = ioctl(fd, cmd, (unsigned long) &v0_hc); But on 64 bit platforms (not 32 bit) cmd will be identical, so we'll just be trying the same ioctl twice. From <xen/linux/privcmd.h>: typedef struct privcmd_hypercall { __u64 op; __u64 arg[5]; } privcmd_hypercall_t; #define IOCTL_PRIVCMD_HYPERCALL \ _IOC(_IOC_NONE, 'P', 0, sizeof(privcmd_hypercall_t)) From src/xen_internal.c: typedef struct v0_hypercall_struct { unsigned long op; unsigned long arg[5]; } v0_hypercall_t; cmd = _IOC(_IOC_NONE, 'P', 0, sizeof(v0_hypercall_t)); On 64 bit platforms, both structures will be the same size. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

[Libvir] about qemu IDE BMDMAState
by Atsushi SAKAI 12 Jun '07

12 Jun '07

Hi, Rich and Dan Following patch will include on Fedora7 updates? http://lists.xensource.com/archives/html/xen-devel/2007-05/msg00564.html This issue also effects Windows FV. Thanks Atsushi SAKAI

3 3

[Libvir] Merging remote patch
by Richard W.M. Jones 11 Jun '07

11 Jun '07

I'm merging the remote patch into libvirt CVS. Things may break, although I'm hoping to keep any brokenness to a minimum. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 3

[Libvir] Suggested list of files to rename
by Richard W.M. Jones 11 Jun '07

11 Jun '07

I've accumulated a collection of files in libvirt which I think should be renamed. This file is in CVS, called 'RENAMES', and attached below. What do you think? Rich. # Suggested list of file renames. # # File renames don't normally go into patches because they make # the patches much harder to read, so list them here instead. # # $Id: RENAMES,v 1.1 2007/06/11 12:25:06 rjones Exp $ # Clearer naming scheme after Xen-unified patch went in. src/xen_internal.c src/xen_internal_hv.c src/xen_internal.h src/xen_internal_hv.h src/proxy_internal.c src/xen_internal_proxy.c src/proxy_internal.h src/xen_internal_proxy.h src/xend_internal.c src/xen_internal_xend.c src/xend_internal.h src/xen_internal_xend.h src/xm_internal.c src/xen_internal_inactive.c src/xm_internal.h src/xen_internal_inactive.h src/xs_internal.c src/xen_internal_xenstore.c src/xs_internal.h src/xen_internal_xenstore.h src/xen_unified.c src/xen_internal.c src/xen_unified.h src/xen_internal.h # Test driver should really be called test_internal. src/test.c src/test_internal.c src/test.h src/test_internal.h # This would be better: src/*_internal*.c src/*_driver*.c src/*_internal*.h src/*_driver*.h # Qemud is now the qemud + remote driver. qemud/protocol.x qemud/qemud_protocol.x qemud/* remote/* -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

Réf. : [Libvir] Release of libvirt-0.2.3
by jean-paul.pigache＠bull.net 11 Jun '07

11 Jun '07

Daniel, Good to see all these enhencements in this version. Is there any chance to have this libvirt 0.2.3 in RHEL5.1 ? Jean-Paul Daniel Veillard <veillard(a)redhat.com> Envoyé par : libvir-list-bounces(a)redhat.com 08/06/2007 13:42 Veuillez répondre à veillard Pour : libvir-list(a)redhat.com cc : Objet : [Libvir] Release of libvirt-0.2.3 Nearly 2 months without a release, it was clearly time, a lot of changes had accumulated in CVS. Available from ftp://libvirt.org/libvirt/ * Documentation: - documentation for upcoming remote access (Richard Jones) - virConnectNumOfDefinedDomains doc (Jan Michael) - virsh help messages for dumpxml and net-dumpxml (Chris Wright) * Bug fixes: - RelaxNG schemas regexp fix (Robin Green) - RelaxNG arch bug (Mark McLoughlin) - large buffers bug fixes (Shigeki Sakamoto) - error on out of memory condition (Shigeki Sakamoto) - virshStrdup fix - non-root driver when using Xen bug (Richard Jones) - use --strict-order when running dnsmasq (Daniel Berrange) - virbr0 weirdness on restart (Mark McLoughlin) - keep connection error messages (Richard Jones) - increase QEmu read buffer on help (Daniel Berrange) - rpm dependance on dnsmasq (Daniel Berrange) - fix XML boot device syntax (Daniel Berrange) - QEmu memory bug (Daniel Berrange) - memory leak fix (Masayuki Sunou) - fix compiler flags (Richard Jones) - remove type ioemu on recent Xen HVM for paravirt drivers (Saori Fukuta) - uninitialized string bug (Masayuki Sunou) - allow init even if the daemon is not running - XML to config fix (Daniel Berrange) * Improvements: - add a special error class for the test module (Richard Jones) - virConnectGetCapabilities on proxy (Richard Jones) - allow network driver to decline usage (Richard Jones) - extend error messages for upcoming remote access (Richard Jones) - on_reboot support for QEmu (Daniel Berrange) - save daemon output in a log file (Daniel Berrange) - xenXMDomainDefineXML can override guest config (Hugh Brock) - add attach-device and detach-device commands to virsh (Masayuki Sunou and Mark McLoughlin and Richard Jones) - make virGetVersion case insensitive and Python bindings (Richard Jones) - new scheduler API (Atsushi SAKAI) - localizations updates - add logging option for virsh (Nobuhiro Itou) - allow arguments to be passed to bootloader (Hugh Brock) - increase the test suite (Daniel Berrange and Hugh Brock) * Cleanups: - Remove VIR_DRV_OPEN_QUIET (Richard Jones) - disable xm_internal.c for Xen > 3.0.3 (Daniel Berrange) - unused fields in _virDomain (Richard Jones) - export __virGetDomain and __virGetNetwork for libvirtd only (Richard Jones) - ignore old VNC config for HVM on recent Xen (Daniel Berrange) - various code cleanups - -Werror cleanup (Hugh Brock) thanks to everybody who helped for this release with bug reports or patches ! Daniel -- Red Hat Virtualization group http://redhat.com/virtualization/ Daniel Veillard | virtualization library http://libvirt.org/ veillard(a)redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/ -- Libvir-list mailing list Libvir-list(a)redhat.com https://www.redhat.com/mailman/listinfo/libvir-list

2 1

[Libvir] Release of libvirt-0.2.3
by Daniel Veillard 11 Jun '07

11 Jun '07

Nearly 2 months without a release, it was clearly time, a lot of changes had accumulated in CVS. Available from ftp://libvirt.org/libvirt/ * Documentation: - documentation for upcoming remote access (Richard Jones) - virConnectNumOfDefinedDomains doc (Jan Michael) - virsh help messages for dumpxml and net-dumpxml (Chris Wright) * Bug fixes: - RelaxNG schemas regexp fix (Robin Green) - RelaxNG arch bug (Mark McLoughlin) - large buffers bug fixes (Shigeki Sakamoto) - error on out of memory condition (Shigeki Sakamoto) - virshStrdup fix - non-root driver when using Xen bug (Richard Jones) - use --strict-order when running dnsmasq (Daniel Berrange) - virbr0 weirdness on restart (Mark McLoughlin) - keep connection error messages (Richard Jones) - increase QEmu read buffer on help (Daniel Berrange) - rpm dependance on dnsmasq (Daniel Berrange) - fix XML boot device syntax (Daniel Berrange) - QEmu memory bug (Daniel Berrange) - memory leak fix (Masayuki Sunou) - fix compiler flags (Richard Jones) - remove type ioemu on recent Xen HVM for paravirt drivers (Saori Fukuta) - uninitialized string bug (Masayuki Sunou) - allow init even if the daemon is not running - XML to config fix (Daniel Berrange) * Improvements: - add a special error class for the test module (Richard Jones) - virConnectGetCapabilities on proxy (Richard Jones) - allow network driver to decline usage (Richard Jones) - extend error messages for upcoming remote access (Richard Jones) - on_reboot support for QEmu (Daniel Berrange) - save daemon output in a log file (Daniel Berrange) - xenXMDomainDefineXML can override guest config (Hugh Brock) - add attach-device and detach-device commands to virsh (Masayuki Sunou and Mark McLoughlin and Richard Jones) - make virGetVersion case insensitive and Python bindings (Richard Jones) - new scheduler API (Atsushi SAKAI) - localizations updates - add logging option for virsh (Nobuhiro Itou) - allow arguments to be passed to bootloader (Hugh Brock) - increase the test suite (Daniel Berrange and Hugh Brock) * Cleanups: - Remove VIR_DRV_OPEN_QUIET (Richard Jones) - disable xm_internal.c for Xen > 3.0.3 (Daniel Berrange) - unused fields in _virDomain (Richard Jones) - export __virGetDomain and __virGetNetwork for libvirtd only (Richard Jones) - ignore old VNC config for HVM on recent Xen (Daniel Berrange) - various code cleanups - -Werror cleanup (Hugh Brock) thanks to everybody who helped for this release with bug reports or patches ! Daniel -- Red Hat Virtualization group http://redhat.com/virtualization/ Daniel Veillard | virtualization library http://libvirt.org/ veillard(a)redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/

2 2

[Libvir] PATCH: Fix bridge device handling with xm config files
by Daniel P. Berrange 07 Jun '07

07 Jun '07

The code for parsing xm config files & generating XML forgot to add in the <source bridge='xenbr0'> elements if an explicit bridge is requested. The reverse direction was already implemeneted. This patch fixes the missing handling of bridge devices & adds a test case 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] [PATCH] add bootloader_args to libvirt xml to support bootloaders (like pypxeboot for example) that require arguments
by Hugh Brock 07 Jun '07

07 Jun '07

This patch extends libvirt's XML to include a paramater Xen already supports in sexpr, "bootloader_args." The bootloader_args parameter lets you pass arguments -- "mac=aa:12:34:56:78:90" for example -- to a bootloader program that xend invokes when it starts a guest. One such program, pypxeboot, uses the mac= argument to communicate with a dhcp server and retrieve the pxeboot kernel and initrd for a paravirt guest. The patch includes new tests for the bootloader_args XML and sexpr, which pass. Note that for the head of current libvirt, the virsh tests do not pass and virsh will not connect to the xen hypervisor. I don't know if this is my own build environment or some other issue. However the failure is the same with or without this patch, so I am submitting it anyway. Please let me know what you think... --Hugh -- Red Hat Virtualization Group http://redhat.com/virtualization Hugh Brock | virt-manager http://virt-manager.org hbrock(a)redhat.com | virtualization library http://libvirt.org

3 5

[Libvir] tentative next release tomorrow
by Daniel Veillard 07 Jun '07

07 Jun '07

We piled up a lot of patches since 0.2.2, I guess it is time for a new release, so I will try to push 0.2.3 out tomorrow morning. Give a try to the version in CVS (or a recent snapshot), thanks, Daniel -- Red Hat Virtualization group http://redhat.com/virtualization/ Daniel Veillard | virtualization library http://libvirt.org/ 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] output virsh log to file
by Nobuhiro Itou 07 Jun '07

07 Jun '07

Hi, I made the patch which output debug messages and error messages of virsh to the log file. I think that it is necessary to leave logs to the file for the investigation at the failure. Debug messages can take appointed optional information, and error messages will make it easy to specify a cause. Moreover, I'm thinking about outputting virterror information to the virsh log file by the next step. It is simple specifications as follows. * Destination -- <USER HOME>/.virsh/virsh.log * Log size -- 1MB(1024*1024) * Number of Rotations -- 5 * There is process exclusion. First, will you take in this patch ? And, I have a plan to enhance logging functionality. (for example, libvirt error logging) Signed-off-by: Nobuhiro Itou <fj0873gn(a)aa.jp.fujitsu.com> Thanks, Nobuhiro Itou.

6 18