June 2007 - Devel - libvirt List Archives

[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 15 Jun '07

15 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