January 2008 - Devel - libvirt List Archives

[Libvir] Hang with virt-install graphic=no on FC 6 6.93/F7 RC2
by libvirtuser 23 Dec '25

23 Dec '25

I had this problem and another user confirmed the same behavior on his machine: http://pastie.caboo.se/65417 Haven't tried it with *2925.10* yet. Also, is there a way to get more verbose info about what calls are causing the "configuration file syntax error" errors? Thanks. virt-install How large would you like the disk () to be (in gigabytes)? 4 ERROR: Must provide a file, not a directory for the disk What would you like to use as the disk (path)? /xen/images/x6.img Would you like to enable graphics support? (yes or no) no What is the install location? http://mirror.anl.gov/pub/fedora-linux-core/test/6.93/Fedora/i386/os/ Starting install... libvir: Xen Daemon error : GET operation failed: libvir: error : configuration file syntax error: expecting a name libvir: error : configuration file syntax error: expecting a name libvir: error : configuration file syntax error: expecting a name Retrieving Fedora... 276 kB 00:01 Retrieving vmlinuz... 100% |=========================| 2.1 MB 00:16 Retrieving initrd.img... 100% |=========================| 5.4 MB 00:28 libvir: Xen Daemon error : GET operation failed: libvir: error : configuration file syntax error: expecting a name libvir: error : configuration file syntax error: expecting a name libvir: error : configuration file syntax error: expecting a name Creating domain... 0 B 00:06 Linux version 2.6.20-2925.5.fc7xen (brewbuilder(a)ls20-bc1-14.build.redhat.com) (gcc version 4.1.2 20070317 (Red Hat 4.1.2-5)) #1 SMP Thu Mar 22 13:51:38 EDT 2007 BIOS-provided physical RAM map: sanitize start sanitize bail 0 copy_e820_map() start: 0000000000000000 size: 0000000020800000 end: 0000000020800000 type: 1 Xen: 0000000000000000 - 0000000020800000 (usable) 0MB HIGHMEM available. 520MB LOWMEM available. Using x86 segment limits to approximate NX protection Zone PFN ranges: DMA 0 -> 133120 Normal 133120 -> 133120 HighMem 133120 -> 133120 early_node_map[1] active PFN ranges 0: 0 -> 133120 ACPI in unprivileged domain disabled Built 1 zonelists. Total pages: 132080 Kernel command line: method=http://mirror.anl.gov/pub/fedora-linux-core/test/6.93/Fedora/i386/os/ Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done. Initializing CPU#0 CPU 0 irqstacks, hard=c135f000 soft=c133f000 PID hash table entries: 4096 (order: 12, 16384 bytes) Xen reported: 1694.994 MHz processor. Console: colour dummy device 80x25 Lock dependency validator: Copyright (c) 2006 Red Hat, Inc., Ingo Molnar ... MAX_LOCKDEP_SUBCLASSES: 8 ... MAX_LOCK_DEPTH: 30 ... MAX_LOCKDEP_KEYS: 2048 ... CLASSHASH_SIZE: 1024 ... MAX_LOCKDEP_ENTRIES: 8192 ... MAX_LOCKDEP_CHAINS: 16384 ... CHAINHASH_SIZE: 8192 memory used by lock dependency info: 1064 kB per task-struct memory footprint: 1200 bytes Dentry cache hash table entries: 131072 (order: 7, 524288 bytes) Inode-cache hash table entries: 65536 (order: 6, 262144 bytes) Software IO TLB disabled vmalloc area: e1000000-f4ffe000, maxmem 2d7fe000 Memory: 503936k/532480k available (2030k kernel code, 19948k reserved, 1079k data, 180k init, 0k highmem) virtual kernel memory layout: fixmap : 0xf5315000 - 0xf57fe000 (5028 kB) pkmap : 0xf5000000 - 0xf5200000 (2048 kB) vmalloc : 0xe1000000 - 0xf4ffe000 ( 319 MB) lowmem : 0xc0000000 - 0xe0800000 ( 520 MB) .init : 0xc130e000 - 0xc133b000 ( 180 kB) .data : 0xc11fb9d9 - 0xc1309714 (1079 kB) .text : 0xc1000000 - 0xc11fb9d9 (2030 kB) Checking if this processor honours the WP bit even in supervisor mode... Ok. Calibrating delay using timer specific routine.. 4262.67 BogoMIPS (lpj=8525343) Security Framework v1.0.0 initialized SELinux: Initializing. SELinux: Starting in permissive mode selinux_register_security: Registering secondary module capability Capability LSM initialized as secondary Mount-cache hash table entries: 512 CPU: Trace cache: 12K uops, L1 D cache: 8K CPU: L2 cache: 256K Checking 'hlt' instruction... OK. SMP alternatives: switching to UP code Freeing SMP alternatives: 11k freed Brought up 1 CPUs Grant table initialized NET: Registered protocol family 16 Brought up 1 CPUs PCI: Fatal: No config space access function found PCI: setting up Xen PCI frontend stub Setting up standard PCI resources ACPI: Interpreter disabled. Linux Plug and Play Support v0.97 (c) Adam Belay pnp: PnP ACPI: disabled xen_mem: Initialising balloon driver. usbcore: registered new interface driver usbfs usbcore: registered new interface driver hub usbcore: registered new device driver usb PCI: System does not support PCI PCI: System does not support PCI NetLabel: Initializing NetLabel: domain hash size = 128 NetLabel: protocols = UNLABELED CIPSOv4 NetLabel: unlabeled traffic allowed by default NET: Registered protocol family 2 IP route cache hash table entries: 32768 (order: 5, 131072 bytes) TCP established hash table entries: 131072 (order: 10, 4194304 bytes) TCP bind hash table entries: 65536 (order: 9, 2097152 bytes) TCP: Hash tables configured (established 131072 bind 65536) TCP reno registered checking if image is initramfs... it is Freeing initrd memory: 7256k freed IA-32 Microcode Update Driver: v1.14-xen <tigran(a)veritas.com> audit: initializing netlink socket (disabled) audit(1180379657.981:1): initialized VFS: Disk quotas dquot_6.5.1 Dquot-cache hash table entries: 1024 (order 0, 4096 bytes) SELinux: Registering netfilter hooks io scheduler noop registered io scheduler anticipatory registered io scheduler deadline registered io scheduler cfq registered (default) pci_hotplug: PCI Hot Plug PCI Core version: 0.5 BUG: at kernel/fork.c:994 copy_process() [<c1005d9a>] show_trace_log_lvl+0x1a/0x2f [<c1006343>] show_trace+0x12/0x14 [<c10063be>] dump_stack+0x16/0x18 [<c101d044>] copy_process+0x195/0x1245 [<c101e144>] do_fork+0x50/0x117 [<c1003351>] kernel_thread+0x8e/0x96 [<c102c701>] __call_usermodehelper+0x2d/0x46 [<c102cd4f>] run_workqueue+0x89/0x145 [<c102d711>] worker_thread+0xd5/0x102 [<c102ff27>] kthread+0xb3/0xdc [<c10058db>] kernel_thread_helper+0x7/0x10 ======================= rtc: IRQ 8 is not free. Non-volatile memory driver v1.2 Linux agpgart interface v0.101 (c) Dave Jones RAMDISK driver initialized: 16 RAM disks of 16384K size 4096 blocksize input: Macintosh mouse button emulation as /class/input/input0 Xen virtual console successfully installed as xvc0 Event-channel device installed. usbcore: registered new interface driver libusual usbcore: registered new interface driver hiddev usbcore: registered new interface driver usbhid drivers/usb/input/hid-core.c: v2.6:USB HID core driver PNP: No PS/2 controller found. Probing ports directly. i8042.c: No controller found. mice: PS/2 mouse device common for all mice TCP bic registered Initializing XFRM netlink socket NET: Registered protocol family 1 NET: Registered protocol family 17 Using IPI No-Shortcut mode XENBUS: Device with no driver: device/vbd/51712 XENBUS: Device with no driver: device/vif/0 Freeing unused kernel memory: 180k freed Write protecting the kernel read-only data: 762k ^[Domain installation still in progress. You can reconnect

2 1

[Libvir] Re: libvirt on mingw
by Richard W.M. Jones 02 Jul '08

02 Jul '08

Brecht Sanders wrote: > Hi, > I saw on the following link: > http://www.mail-archive.com/libvir-list@redhat.com/msg04103.html > that you are also trying to compile libvirt on win32. > I'm also attempting to do this, and I guess I got stuck at the same > point your post was about. > Have you in the mean time found an XDR implementation that compiles on > MinGW and that implements xdr_u_quad_t? > If you did, can you please tell me where to find it? You'll find the answer to this question and more if you look through the libvir-list archives for the current month: https://www.redhat.com/archives/libvir-list/2008-January/thread.html 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 12

[Libvir] [PATCH] Fix USB device name mis-conversion from S-Expr to XML
by Hiroyuki Kaguchi 22 Jun '08

22 Jun '08

"tablet" and "mouse" are set as a value of the /local/domain/<domid>/image/hvm/usbdevice, but libvirt expects "usbdevice" and "usbmouse" as a value. This causes the following. If a USB device is attached by virt-manager, the USB device is not seen from virt-manager. This patch fixes expected value of libvirt Thanks, Hiroyuki Kaguchi Index: xend_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/xend_internal.c,v retrieving revision 1.162 diff -u -r1.162 xend_internal.c --- xend_internal.c 21 Jan 2008 16:29:10 -0000 1.162 +++ xend_internal.c 23 Jan 2008 00:37:02 -0000 @@ -1743,11 +1743,9 @@ node = cur->u.s.car; if (sexpr_lookup(node, "usbdevice")) { tmp = sexpr_node(node, "usbdevice"); - if (tmp && *tmp) { - if (!strcmp(tmp, "usbtablet")) - virBufferAdd(&buf, " <input type='tablet' bus='usb'/>\n", 37); - else if (!strcmp(tmp, "usbmouse")) - virBufferAdd(&buf, " <input type='mouse' bus='usb'/>\n", 36); + if (!strcmp(tmp, "tablet") || + !strcmp(tmp, "mouse")) { + virBufferVSprintf(&buf, " <input type='%s' bus='usb'/>\n", tmp); } } }

3 4

[Libvir] java-libvirt on cvs:libvirt.org
by Daniel Schwager 14 Mar '08

14 Mar '08

Hallo list, i took the java-libvirt port from Tóth István (Thank you - great work !!) and added - the some features from libvirt 0.4.0 - integrate of xsd XML-description / validator - added auto generation of XML-beans related to the XSD file. - project build.xml file - integration of testng - remote building of JNI library - remote testing of junit/testing tests So, I want to publish to port to a public cvs-tree. Is there a little space on the libvirt website for this port ? regards Danny

4 3

[Libvir] RFC: Supporting serial & parallel ports for QEMU (and improving Xen)
by Daniel P. Berrange 26 Feb '08

26 Feb '08

One user's feature request for our QEMU driver is to support serial ports. Easy you might think... you'd be wrong :-) On Xen we have very simple approach. When creating a guest simply add <console/> And it'll cause a serial port to be setup with an autoallocated pty, so you get back <console pty='/dev/pts/5'/> In retrospect calling it 'console' was dumb, but hey we're stuck with that now and its only a tiny naming issue so I don't mind really. We can do the same thing with QEMU quite easily. QEMU, however, supports many many many more ways to hooking up the serial port to Dom0. Indeed so does Xen fullyvirt, but we don't expose this ability. Parallel ports have identical config syntax to serial ports so we might as well deal with both at once. So, here's a stripped down version of the QEMU docs: [quote http://fabrice.bellard.free.fr/qemu/qemu-doc.html#SEC10] `-serial dev' Redirect the virtual serial port to host character device dev. The default device is vc in graphical mode and stdio in non graphical mode. This option can be used several times to simulate up to 4 serials ports. Use -serial none to disable all serial ports. Available character devices are: vc Virtual console pty [Linux only] Pseudo TTY (a new PTY is automatically allocated) none No device is allocated. null void device /dev/XXX [Linux only] Use host tty, e.g. `/dev/ttyS0'. The host serial port parameters are set according to the emulated ones. /dev/parportN [Linux only, parallel port only] Use host parallel port N. Currently only SPP parallel port features can be used. file:filename Write output to filename. No character can be read. stdio [Unix only] standard input/output pipe:filename name pipe filename COMn [Windows only] Use host serial port n udp:[remote_host]:remote_port[@[src_ip]:src_port] This implements UDP Net Console. tcp:[host]:port[,server][,nowait][,nodelay] The TCP Net Console has two modes of operation. telnet:host:port[,server][,nowait][,nodelay] The telnet protocol is used instead of raw tcp sockets. unix:path[,server][,nowait] A unix domain socket is used instead of a tcp socket. [/quote] I don't see any reason to not support all/most of these options. The things I don't like here is that /dev/XXX, vs /dev/parportN, vs COMn differences for connecting guest <-> host passthrough of the devices. I figure it could be simpler if it was just represented as 'n' and we'd translate that to be /dev/ttyS[n] or /dev/parport[n] or COM[n] as needed. The question as ever is how to represent this in XML. For serial ports we'll stick with '<console>', while parallel ports we might as well use a better named '<parallel>'. Next up, I think should use a 'type' attribute on this element to determine the main way ot connecting the device, and then more type specific attributes or sub-elements as needed. If 'type' was not specified then use a default of 'pty', since that gives compatability with existing practice. As an illustrative example /* * Parse the XML definition for a character device * * Top level node will be <console> or <parallel>, but all attributes * and sub-elements are identical. * * type=vc|pty|null|host|file|pipe|udp|tcp|telnet, default is pty * * <console type='vc'/> * * <console type='pty' pty='/dev/pts/3'/> * * <console type='null'/> * * <console type='host' port='3'/> * * <console type='file' path='/some/file'/> * * <console type='pipe' path='/some/file'/> * * <console type='udp'> * <sendto port='12356'/> * </console> * * <console type='udp'> * <sendto addr='127.0.0.1' port='12356'/> * </console> * * <console type='udp'> * <sendto addr='127.0.0.1' port='12356'/> * <bind port='12356'/> * </console> * * <console type='udp'> * <sendto addr='127.0.0.1' port='12356'/> * <bind addr='127.0.0.1' port='12356'/> * </console> * * <console type='tcp'> * <listen port='12356'/> * </console> * * <console type='tcp'> * <listen addr='127.0.0.1' port='12356'> * <nowait/> * <nodelay/> * </listen> * </console> * * <console type='tcp'> * <connect addr='127.0.0.1' port='12356'> * <nodelay/> * </connect> * </console> * * <console type='telnet'> * <listen addr='127.0.0.1' port='12356'/> * </console> * * <console type='telnet'> * <connect addr='127.0.0.1' port='12356'/> * </console> * */ BTW, the udp, tcp, telnet options are only available on QEMU >= 0.9.0. We already have ability to detect / validate that for both Xen & QEMU drivers. NB, whereever there are IP addresses, hostnames can be used too, hence I call the attriute 'addr' instead of 'ip' 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 -=|

5 8

[Libvir] [PATCH] Fix the function that remove the element of the hash table.
by Hiroyuki Kaguchi 07 Feb '08

07 Feb '08

There are two logic error and a unnecessary else-statement in virHashRemoveSet function. This patch fix the following. (1/3) The logic error that use released memory area. (2/3) The logic error that doesn't remove elements. (3/3) Unnecessary else-statement. (1/3) The logic error that use released memory area. This bug causes the crash of virt-manager or virsh. If "virsh undefine" command is executed, "Segmentation Fault" is caused. The result of executing "valgrind virsh" command is: virsh # list --all ==3337== Invalid read of size 4 ==3337== at 0x4016A21: virHashRemoveSet (hash.c:562) ==3337== by 0x403ACF4: xenXMConfigCacheRefresh (xm_internal.c:458) ==3337== by 0x4040D58: xenXMNumOfDefinedDomains (xm_internal.c:2518) ==3337== by 0x40245C3: xenUnifiedNumOfDefinedDomains (xen_unified.c:1004) ==3337== by 0x4013730: virConnectNumOfDefinedDomains (libvirt.c:2446) ==3337== by 0x804AE81: cmdList (virsh.c:577) ==3337== by 0x8052E6E: vshCommandRun (virsh.c:4052) ==3337== by 0x8054E2D: main (virsh.c:5036) ==3337== Address 0x4282AB8 is 0 bytes inside a block of size 16 free'd ==3337== at 0x4004FDA: free (vg_replace_malloc.c:233) ==3337== by 0x40169A0: virHashRemoveSet (hash.c:545) ==3337== by 0x403ACF4: xenXMConfigCacheRefresh (xm_internal.c:458) ==3337== by 0x4040D58: xenXMNumOfDefinedDomains (xm_internal.c:2518) ==3337== by 0x40245C3: xenUnifiedNumOfDefinedDomains (xen_unified.c:1004) ==3337== by 0x4013730: virConnectNumOfDefinedDomains (libvirt.c:2446) ==3337== by 0x804AE81: cmdList (virsh.c:577) ==3337== by 0x8052E6E: vshCommandRun (virsh.c:4052) ==3337== by 0x8054E2D: main (virsh.c:5036) The current removing logic is: 1. search for node that be removed in the hash table. The removing cursor(pointer) point to "element-B". +-hash table---+ | +----------+ | +----------+ +----------+ | |element-A |-|-->|element-B |-->|element-C | | +----------+ | +----------+ +----------+ | |element-D | | | +----------+ | | . | | . | +--------------+ 2. free "element-B". The removing cursor point to invalid address +-hash table---+ | +----------+ | +----------+ | |element-A |-|-->|element-C | | +----------+ | +----------+ | |element-D | | | +----------+ | | . | | . | +--------------+ 3. try to move the removing cursor to next node. At this time, the error occurs. (2/3) The logic error that doesn't remove elements. The current removing logic is: 1. search for node that will be removed in the hash table. The removing cursor point to "element-A". +-hash table---+ | +----------+ | +----------+ +----------+ | |element-A |-|-->|element-B |-->|element-C | | +----------+ | +----------+ +----------+ | |element-D | | | +----------+ | | . | | . | +--------------+ 2. copy "element-B" to "element-A". +-hash table---+ | +----------+ | +----------+ +----------+ | |element-B'|-|-->|element-B |-->|element-C | | +----------+ | +----------+ +----------+ | |element-D | | | +----------+ | | . | | . | +--------------+ 3. remove "element-B" The removing cursor point to "element-D". "element-C" is skipped. +-hash table---+ | +----------+ | +----------+ | |element-B'|-|-->|element-C | | +----------+ | +----------+ | |element-D | | | +----------+ | | . | | . | +--------------+ (3/3) Unnecessary else-statement. There is else-statement that set NULL to the NULL pointer. Thanks, Hiroyuki Kaguchi Index: hash.c =================================================================== RCS file: /data/cvs/libvirt/src/hash.c,v retrieving revision 1.27 diff -u -r1.27 hash.c --- hash.c 21 Jan 2008 16:29:10 -0000 1.27 +++ hash.c 28 Jan 2008 06:48:09 -0000 @@ -543,6 +543,7 @@ if (prev) { prev->next = entry->next; free(entry); + entry = prev; } else { if (entry->next == NULL) { entry->valid = 0; @@ -553,6 +554,7 @@ sizeof(virHashEntry)); free(entry); entry = NULL; + i--; } } table->nbElems--; @@ -560,8 +562,6 @@ prev = entry; if (entry) { entry = entry->next; - } else { - entry = NULL; } } }

4 6

[Libvir] bug in xen_unified.c
by Marcus Meissner 07 Feb '08

07 Feb '08

Hi, spotted by gcc 4.3, in libvirt 0.4.0, xen_unified.c:xenUnifiedDomainSuspend() if (priv->opened[i] && which should probably be: if (priv->opened[XEN_UNIFIED_HYPERVISOR_OFFSET] && because it is outside of the loop, and the other indices in that if (). --- src/xen_unified.c 2008/01/28 22:20:22 1.1 +++ src/xen_unified.c 2008/01/28 22:20:26 @@ -699,7 +699,7 @@ drivers[i]->domainDestroy (dom) == 0) return 0; - if (priv->opened[i] && + if (priv->opened[XEN_UNIFIED_HYPERVISOR_OFFSET] && drivers[XEN_UNIFIED_HYPERVISOR_OFFSET]->domainDestroy && drivers[XEN_UNIFIED_HYPERVISOR_OFFSET]->domainDestroy (dom) == 0) return 0; Ciao, Marcus -- Working, but not speaking, for the following german company: SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nuernberg)

3 3

[Libvir] Pthreads error in windows compilation
by Gabriel Kaufmann 07 Feb '08

07 Feb '08

Hi, I received the error In file included from libvirt.c:32: internal.h: At top level: internal.h:177: error: expected specifier-qualifier-list before 'pthread_mutex_t' In order to fix this problem, I installed the 'pthreads' package on my MinGW. Then added include pthread.h in src/internal.h Is this the right action? Best Regards, Gabriel Kaufmann Software Engineer Gabriel.Kaufmann(a)ericom.com<mailto:Gabriel.Kaufmann@ericom.com> Ericom Software Tel (Dir): +972 2 591 1700 Ext 754 Tel (Main): +972 2 591 1700 http://www.ericom.com Access Done Right Empower Enterprise-Wide Access to Microsoft(r) Terminal Server, Virtual Desktops and Legacy Applications

4 6

[Libvir] [PATCH] change a Disk/Nic of inactive domain
by S.Sakamoto 06 Feb '08

06 Feb '08

Hi The libvirt on Xen-3.03 does not have the function that add/change/delete a Disk/NIC of inactive domain. So, I made a patch which enabled add/change/delete of Disk/NIC of inactive domain. This patch will be able to change "disk" or "vif" parameter in configuration file with "attach-disk","attach-interface", "attach-device", "detach-disk", "detach-interface" and "detach-device" command, like "setmem" or "setvcpus". Thanks, Shigeki Sakamoto.

5 9

[Libvir] virsh vncdisplay fix
by John Levon 03 Feb '08

03 Feb '08

I have an old HVM guest defined in 3.0.4, and now I'm running 3.1.2 and libvirt 0.4.0. The config looks like this: (domain (image (hvm (vnc 1) (vncdisplay 0) (device (vfb (location localhost:5900) (vncdisplay 0) (uuid ...) The libvirt code looks like this: 1757 /* Graphics device (HVM <= 3.0.4, or PV <= 3.0.3) vnc config */ 1758 if ((hvm && xendConfigVersion < 4) || 1759 (!hvm && xendConfigVersion < 3)) { 1760 tmp = sexpr_fmt_node(root, "domain/image/%s/vnc", hvm ? "hvm" : "linux"); So we don't see the 'vnc 1' at the top. But the code to parse device/vfb doesn't expect this config either: 1685 tmp = sexpr_node(node, "device/vfb/type"); The patch below replaces the later version check with something that copes correctly with the "upgrade" case. regards john 2008-01-15 John Levon <levon(a)movementarian.org> * src/xend_internal.c: fix VNC parsing of old Xen domains on newer Xen versions. diff -prauN libvirt-0.4.0/src/xend_internal.c libvirt-new/src/xend_internal.c --- libvirt-0.4.0/src/xend_internal.c 2007-12-17 15:05:27.000000000 -0800 +++ libvirt-new/src/xend_internal.c 2008-01-14 18:42:08.895060570 -0800 @@ -1371,6 +1371,7 @@ xend_parse_sexp_desc(virConnectPtr conn, unsigned char uuid[VIR_UUID_BUFLEN]; char uuidstr[VIR_UUID_STRING_BUFLEN]; int vif_index = 0; + int found_graphics = 0; if (root == NULL) { /* ERROR */ @@ -1683,6 +1684,7 @@ xend_parse_sexp_desc(virConnectPtr conn, if (tmp && !strcmp(tmp, "sdl")) { virBufferVSprintf(&buf, " <input type='mouse' bus='%s'/>\n", hvm ? "ps2": "xen"); virBufferAdd(&buf, " <graphics type='sdl'/>\n", 27); + found_graphics = 1; } else if (tmp && !strcmp(tmp, "vnc")) { int port = xenStoreDomainGetVNCPort(conn, domid); const char *listenAddr = sexpr_node(node, "device/vfb/vnclisten"); @@ -1700,6 +1702,7 @@ xend_parse_sexp_desc(virConnectPtr conn, if (keymap) virBufferVSprintf(&buf, " keymap='%s'", keymap); virBufferAdd(&buf, "/>\n", 3); + found_graphics = 1; } } } @@ -1751,8 +1754,7 @@ xend_parse_sexp_desc(virConnectPtr conn, } /* Graphics device (HVM <= 3.0.4, or PV <= 3.0.3) vnc config */ - if ((hvm && xendConfigVersion < 4) || - (!hvm && xendConfigVersion < 3)) { + if (!found_graphics) { tmp = sexpr_fmt_node(root, "domain/image/%s/vnc", hvm ? "hvm" : "linux"); if (tmp != NULL) { if (tmp[0] == '1') { @@ -1780,6 +1782,7 @@ xend_parse_sexp_desc(virConnectPtr conn, if (keymap) virBufferVSprintf(&buf, " keymap='%s'", keymap); virBufferAdd(&buf, "/>\n", 3); + found_graphics = 1; } } @@ -1789,6 +1792,7 @@ xend_parse_sexp_desc(virConnectPtr conn, if (tmp[0] == '1') { virBufferVSprintf(&buf, " <input type='mouse' bus='%s'/>\n", hvm ? "ps2" : "xen"); virBufferAdd(&buf, " <graphics type='sdl'/>\n", 27 ); + found_graphics = 1; } } }

2 5

[Libvir] [PATCH] Pull vc-list-files from gnulib, and update.
by Jim Meyering 01 Feb '08

01 Feb '08

Now that I've added the vc-list-files script to gnulib, this patch makes libvirt's bootstrap pull it from there. Changes from the previous version of the script: I reintroduced support for mercurial (only lightly tested), and rearranged the directory-existence tests so git is first. Here's the full patch, including a renaming in vasnprintf.c: ---------------- Pull vc-list-files from gnulib, and update. * bootstrap (gnulib_tool): Add the vc-list-files module. --- bootstrap | 3 ++- build-aux/vc-list-files | 19 +++++++++++++------ gnulib/lib/Makefile.am | 9 ++++++++- gnulib/lib/vasnprintf.c | 18 ++++++++++-------- gnulib/m4/gnulib-cache.m4 | 4 ++-- gnulib/m4/gnulib-comp.m4 | 1 + 6 files changed, 36 insertions(+), 18 deletions(-) diff --git a/bootstrap b/bootstrap index 751b244..1627086 100755 --- a/bootstrap +++ b/bootstrap @@ -77,7 +77,8 @@ $gnulib_tool \ --tests-base=gnulib/tests \ --import physmem getaddrinfo \ sys_stat vasprintf strndup \ - strsep poll gettext getpass + strsep poll gettext getpass \ + vc-list-files rm -f \ .gitignore \ diff --git a/build-aux/vc-list-files b/build-aux/vc-list-files index 72a6f54..1e158bb 100755 --- a/build-aux/vc-list-files +++ b/build-aux/vc-list-files @@ -1,7 +1,7 @@ #!/bin/sh -# List the specified version-controlled files. +# List version-controlled file names. -# Copyright (C) 2006, 2007 Free Software Foundation, Inc. +# Copyright (C) 2006-2008 Free Software Foundation, Inc. # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by @@ -18,7 +18,8 @@ # List the specified version-controlled files. -# With no argument, list them all. +# With no argument, list them all. With a single DIRECTORY argument, +# list the version-controlled files in that directory. # This script must be run solely from the top of a $srcdir build directory. # If there's an argument, it must be a single, "."-relative directory name. @@ -28,12 +29,17 @@ dir= case $# in 0) ;; 1) dir=$1 ;; - *) echo "$0: too many arguments" 1>&2; exit 1 ;; + *) echo "$0: too many arguments" 1>&2 + echo "Usage: $0 [DIR]" 1>&2; exit 1;; esac test "x$dir" = x && dir=. -if test -d CVS; then +if test -d .git; then + exec git ls-files "$dir" +elif test -d .hg; then + exec hg locate "$dir/*" +elif test -d CVS; then if test -x build-aux/cvsu; then build-aux/cvsu --find --types=AFGM "$dir" else @@ -46,5 +52,6 @@ if test -d CVS; then $(find ${*-*} -name Entries -print) /dev/null; fi else - git-ls-files "$dir" + echo "$0: Failed to determine type of version control used in "`pwd` 1>&2 + exit 1 fi diff --git a/gnulib/lib/Makefile.am b/gnulib/lib/Makefile.am index 64fc398..8f545ca 100644 --- a/gnulib/lib/Makefile.am +++ b/gnulib/lib/Makefile.am @@ -9,7 +9,7 @@ # the same distribution terms as the rest of that program. # # Generated by gnulib-tool. -# Reproduce by: gnulib-tool --import --dir=. --lib=libgnu --source-base=gnulib/lib --m4-base=gnulib/m4 --doc-base=doc --aux-dir=build-aux --with-tests --lgpl=2 --libtool --macro-prefix=gl getaddrinfo getpass gettext physmem poll strndup strsep sys_stat vasprintf +# Reproduce by: gnulib-tool --import --dir=. --lib=libgnu --source-base=gnulib/lib --m4-base=gnulib/m4 --doc-base=doc --aux-dir=build-aux --with-tests --lgpl=2 --libtool --macro-prefix=gl getaddrinfo getpass gettext physmem poll strndup strsep sys_stat vasprintf vc-list-files AUTOMAKE_OPTIONS = 1.5 gnits @@ -719,6 +719,13 @@ EXTRA_libgnu_la_SOURCES += asprintf.c vasprintf.c ## end gnulib module vasprintf +## begin gnulib module vc-list-files + + +EXTRA_DIST += $(top_srcdir)/build-aux/vc-list-files + +## end gnulib module vc-list-files + ## begin gnulib module wchar BUILT_SOURCES += $(WCHAR_H) diff --git a/gnulib/lib/vasnprintf.c b/gnulib/lib/vasnprintf.c index 6c8d5dc..205a9bf 100644 --- a/gnulib/lib/vasnprintf.c +++ b/gnulib/lib/vasnprintf.c @@ -1,5 +1,5 @@ /* vsprintf with automatic memory allocation. - Copyright (C) 1999, 2002-2007 Free Software Foundation, Inc. + Copyright (C) 1999, 2002-2008 Free Software Foundation, Inc. This program is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by @@ -95,7 +95,7 @@ #if (NEED_PRINTF_DOUBLE || NEED_PRINTF_INFINITE_DOUBLE) && !defined IN_LIBINTL # include <math.h> -# include "isnan.h" +# include "isnand.h" #endif #if (NEED_PRINTF_LONG_DOUBLE || NEED_PRINTF_INFINITE_LONG_DOUBLE) && !defined IN_LIBINTL @@ -106,7 +106,7 @@ #if (NEED_PRINTF_DIRECTIVE_A || NEED_PRINTF_DOUBLE) && !defined IN_LIBINTL # include <math.h> -# include "isnan.h" +# include "isnand.h" # include "printf-frexp.h" #endif @@ -236,7 +236,7 @@ decimal_point_char () static int is_infinite_or_zero (double x) { - return isnan (x) || x + x == x; + return isnand (x) || x + x == x; } #endif @@ -2327,7 +2327,7 @@ VASNPRINTF (DCHAR_T *resultbuf, size_t *lengthp, # if NEED_PRINTF_DIRECTIVE_A || NEED_PRINTF_DOUBLE double arg = a.arg[dp->arg_index].a.a_double; - if (isnan (arg)) + if (isnand (arg)) { if (dp->conversion == 'A') { @@ -2676,7 +2676,7 @@ VASNPRINTF (DCHAR_T *resultbuf, size_t *lengthp, if (dp->conversion == 'f' || dp->conversion == 'F') { double arg = a.arg[dp->arg_index].a.a_double; - if (!(isnan (arg) || arg + arg == arg)) + if (!(isnand (arg) || arg + arg == arg)) { /* arg is finite and nonzero. */ int exponent = floorlog10 (arg < 0 ? -arg : arg); @@ -3080,7 +3080,7 @@ VASNPRINTF (DCHAR_T *resultbuf, size_t *lengthp, { double arg = a.arg[dp->arg_index].a.a_double; - if (isnan (arg)) + if (isnand (arg)) { if (dp->conversion >= 'A' && dp->conversion <= 'Z') { @@ -4327,7 +4327,7 @@ VASNPRINTF (DCHAR_T *resultbuf, size_t *lengthp, if (prec_ourselves) { /* Handle the precision. */ - TCHAR_T *prec_ptr = + TCHAR_T *prec_ptr = # if USE_SNPRINTF (TCHAR_T *) (result + length); # else @@ -4654,6 +4654,7 @@ VASNPRINTF (DCHAR_T *resultbuf, size_t *lengthp, not have this limitation. */ return result; +#if USE_SNPRINTF overflow: if (!(result == resultbuf || result == NULL)) free (result); @@ -4662,6 +4663,7 @@ VASNPRINTF (DCHAR_T *resultbuf, size_t *lengthp, CLEANUP (); errno = EOVERFLOW; return NULL; +#endif out_of_memory: if (!(result == resultbuf || result == NULL)) diff --git a/gnulib/m4/gnulib-cache.m4 b/gnulib/m4/gnulib-cache.m4 index d55e07a..357f170 100644 --- a/gnulib/m4/gnulib-cache.m4 +++ b/gnulib/m4/gnulib-cache.m4 @@ -15,11 +15,11 @@ # Specification in the form of a command-line invocation: -# gnulib-tool --import --dir=. --lib=libgnu --source-base=gnulib/lib --m4-base=gnulib/m4 --doc-base=doc --aux-dir=build-aux --with-tests --lgpl=2 --libtool --macro-prefix=gl getaddrinfo getpass gettext physmem poll strndup strsep sys_stat vasprintf +# gnulib-tool --import --dir=. --lib=libgnu --source-base=gnulib/lib --m4-base=gnulib/m4 --doc-base=doc --aux-dir=build-aux --with-tests --lgpl=2 --libtool --macro-prefix=gl getaddrinfo getpass gettext physmem poll strndup strsep sys_stat vasprintf vc-list-files # Specification in the form of a few gnulib-tool.m4 macro invocations: gl_LOCAL_DIR([]) -gl_MODULES([getaddrinfo getpass gettext physmem poll strndup strsep sys_stat vasprintf]) +gl_MODULES([getaddrinfo getpass gettext physmem poll strndup strsep sys_stat vasprintf vc-list-files]) gl_AVOID([]) gl_SOURCE_BASE([gnulib/lib]) gl_M4_BASE([gnulib/m4]) diff --git a/gnulib/m4/gnulib-comp.m4 b/gnulib/m4/gnulib-comp.m4 index 1b91a1e..8896a38 100644 --- a/gnulib/m4/gnulib-comp.m4 +++ b/gnulib/m4/gnulib-comp.m4 @@ -225,6 +225,7 @@ AC_DEFUN([gltests_LIBSOURCES], [ AC_DEFUN([gl_FILE_LIST], [ build-aux/config.rpath build-aux/link-warning.h + build-aux/vc-list-files lib/alloca.in.h lib/asnprintf.c lib/asprintf.c -- 1.5.4.rc5.1.g0fa73

3 4

[Libvir] avoid used-uninitialized errors in test driver
by Jim Meyering 01 Feb '08

01 Feb '08

Here's another distracting and easily avoided error: $ valgrind --leak-check=full ./virsh --connect test://$PWD/../docs/testnode.xml list ==13150== Conditional jump or move depends on uninitialised value(s) ==13150== at 0x4177DE: testLoadDomain (test.c:324) ==13150== by 0x417BA4: testLoadDomainFromFile (test.c:401) ==13150== by 0x418E52: testOpenFromFile (test.c:797) ==13150== by 0x4192C9: testOpen (test.c:913) ==13150== by 0x40FA20: do_open (libvirt.c:572) ==13150== by 0x40FDE4: virConnectOpenAuth (libvirt.c:681) ==13150== by 0x40D5B1: vshInit (virsh.c:4464) ==13150== by 0x40E67C: main (virsh.c:4985) ==13150== ==13150== Conditional jump or move depends on uninitialised value(s) ==13150== at 0x417F0D: testLoadNetwork (test.c:486) ==13150== by 0x4183AA: testLoadNetworkFromFile (test.c:577) ==13150== by 0x418F81: testOpenFromFile (test.c:822) ==13150== by 0x4192C9: testOpen (test.c:913) ==13150== by 0x40FA20: do_open (libvirt.c:572) ==13150== by 0x40FDE4: virConnectOpenAuth (libvirt.c:681) ==13150== by 0x40D5B1: vshInit (virsh.c:4464) ==13150== by 0x40E67C: main (virsh.c:4985) ==13150== ==13150== Conditional jump or move depends on uninitialised value(s) ==13150== at 0x41955B: testNumOfDomains (test.c:1020) ==13150== by 0x4104B9: virConnectNumOfDomains (libvirt.c:940) ==13150== by 0x405363: cmdList (virsh.c:557) ==13150== by 0x40C5F9: vshCommandRun (virsh.c:4032) ==13150== by 0x40E6AF: main (virsh.c:4991) Id Name State ---------------------------------- 1 fv0 running 2 fc4 running This patch fixes it: * src/test.c (testOpenFromFile): avoid used-uninitialized errors in test driver --- src/test.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/test.c b/src/test.c index 85170d9..fe5da41 100644 --- a/src/test.c +++ b/src/test.c @@ -672,7 +672,7 @@ static int testOpenFromFile(virConnectPtr conn, xmlNodePtr *domains, *networks = NULL; xmlXPathContextPtr ctxt = NULL; virNodeInfoPtr nodeInfo; - testConnPtr privconn = malloc(sizeof(*privconn)); + testConnPtr privconn = calloc(1, sizeof(*privconn)); if (!privconn) { testError(NULL, NULL, NULL, VIR_ERR_NO_MEMORY, "testConn"); return VIR_DRV_OPEN_ERROR; -- 1.5.4.rc5.1.g0fa73

2 1

[Libvir] libvirt.c: avoid a double-free upon do_open failure
by Jim Meyering 01 Feb '08

01 Feb '08

With a contrived example using more than 20 (the max permitted by the testing framework) domains, I got a double-free error: ==4821== Invalid free() / delete / delete[] ==4821== at 0x4A0560B: free (vg_replace_malloc.c:233) ==4821== by 0x4167F1: virReleaseConnect (hash.c:717) ==4821== by 0x4168E7: virUnrefConnect (hash.c:746) ==4821== by 0x40FCEB: do_open (libvirt.c:621) ==4821== by 0x40FDF1: virConnectOpenAuth (libvirt.c:682) ==4821== by 0x40D5B1: vshInit (virsh.c:4464) ==4821== by 0x40E67C: main (virsh.c:4985) ==4821== Address 0x4C3BA68 is 0 bytes inside a block of size 60 free'd ==4821== at 0x4A0560B: free (vg_replace_malloc.c:233) ==4821== by 0x40FCB3: do_open (libvirt.c:618) ==4821== by 0x40FDF1: virConnectOpenAuth (libvirt.c:682) ==4821== by 0x40D5B1: vshInit (virsh.c:4464) ==4821== by 0x40E67C: main (virsh.c:4985) error: failed to connect to the hypervisor ... here's one way to fix it: diff --git a/src/libvirt.c b/src/libvirt.c index defadc1..c19565f 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -615,7 +615,6 @@ do_open (const char *name, return ret; failed: - free (ret->name); if (ret->driver) ret->driver->close (ret); if (uri) xmlFreeURI(uri); virUnrefConnect(ret); At first, rather than removing the offending free, I inserted this line just after it: ret->name = NULL; which avoids leaking ->name even if some driver-specific close function fails to clean up properly. But IMHO if such a function doesn't clean up properly then *it* should be fixed, not all callers.

2 1

[Libvir] [PATCH] test.c: Avoid segfault upon malloc failure, and plug a leak.
by Jim Meyering 01 Feb '08

01 Feb '08

Testing exposed a tiny leak: $ valgrind --leak-check=full ./virsh --connect \ test://$PWD/../docs/testnode.xml save fc4 /dev/null ... ==11077== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 4 from 1) ==11077== malloc/free: in use at exit: 24,677 bytes in 367 blocks. ==11077== malloc/free: 23,521 allocs, 23,154 frees, 3,424,316 bytes allocated. ==11077== For counts of detected errors, rerun with: -v ==11077== searching for pointers to 367 not-freed blocks. ==11077== checked 538,400 bytes. ==11077== ==11077== 4,000 bytes in 1 blocks are definitely lost in loss record 26 of 27 ==11077== at 0x4A059F6: malloc (vg_replace_malloc.c:149) ==11077== by 0x41C53A: virBufferNew (buf.c:132) ==11077== by 0x41AB22: testDomainDumpXML (test.c:1473) ==11077== by 0x41A1B5: testDomainSave (test.c:1284) ==11077== by 0x41120B: virDomainSave (libvirt.c:1342) ==11077== by 0x406169: cmdSave (virsh.c:1057) ==11077== by 0x40C5F9: vshCommandRun (virsh.c:4032) ==11077== by 0x40E6AF: main (virsh.c:4991) ==11077== ==11077== LEAK SUMMARY: ==11077== definitely lost: 4,000 bytes in 1 blocks. ==11077== possibly lost: 0 bytes in 0 blocks. ==11077== still reachable: 20,677 bytes in 366 blocks. ==11077== suppressed: 0 bytes in 0 blocks. Here's the fix: test.c: Avoid segfault upon malloc failure, and plug a leak. * src/test.c (testDomainSave): Time permitting, I will add tests to exercise as many of the commands as possible. --- src/test.c | 9 ++++++++- 1 files changed, 8 insertions(+), 1 deletions(-) diff --git a/src/test.c b/src/test.c index 35e41a3..85170d9 100644 --- a/src/test.c +++ b/src/test.c @@ -1,7 +1,7 @@ /* * test.c: A "mock" hypervisor for use by application unit tests * - * Copyright (C) 2006-2007 Red Hat, Inc. + * Copyright (C) 2006-2008 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -1281,6 +1281,11 @@ static int testDomainSave(virDomainPtr domain, GET_DOMAIN(domain, -1); xml = testDomainDumpXML(domain, 0); + if (xml == NULL) { + testError(domain->conn, domain, NULL, VIR_ERR_INTERNAL_ERROR, + "cannot allocate space for metadata"); + return (-1); + } if ((fd = open(path, O_CREAT|O_TRUNC|O_WRONLY, S_IRUSR|S_IWUSR)) < 0) { testError(domain->conn, domain, NULL, VIR_ERR_INTERNAL_ERROR, @@ -1303,9 +1308,11 @@ static int testDomainSave(virDomainPtr domain, if (write(fd, xml, len) != len) { testError(domain->conn, domain, NULL, VIR_ERR_INTERNAL_ERROR, "cannot write metadata"); + free(xml); close(fd); return (-1); } + free(xml); if (close(fd) < 0) { testError(domain->conn, domain, NULL, VIR_ERR_INTERNAL_ERROR, "cannot save domain data"); -- 1.5.4.rc5.1.g0fa73

2 1

[Libvir] [PATCH] fix compilation with USE_XEN == 0
by Guido Guenther 31 Jan '08

31 Jan '08

Hi, statstest.c needs to check if USE_XEN > 0, otherwise compilation breaks on --without-xen builds. Please apply if appropriate. Patch is against current cvs. Cheers, -- Guido

3 4

[Libvir] Re: libvirt on mingw
by Richard W.M. Jones 31 Jan '08

31 Jan '08

Brecht Sanders wrote: > I wasn't able to run virt-manager yet because of Python dependancy > issues, but I'm sure I'll be able to work that out. > Did you have virt-manager running on win32 yet? No, I wasn't able to get virt-manager running. Probably the same problems as you are having: building Python modules (specifically, libvirtmod.dll) which can work from the Win32 version of python, with gcc/MinGW, proves to be very hard. If you get it working, please let me know. 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] virsh capabilities cmd: plug a leak
by Jim Meyering 31 Jan '08

31 Jan '08

Here's another: $ valgrind --leak-check=full \ ./virsh --connect test://$PWD/../docs/testnode.xml capabilities ... ==16392== 368 bytes in 1 blocks are definitely lost in loss record 13 of 27 ==16392== at 0x4A059F6: malloc (vg_replace_malloc.c:149) ==16392== by 0x387BC7CC11: strdup (strdup.c:43) ==16392== by 0x4194D2: testGetCapabilities (test.c:1006) ==16392== by 0x412B43: virConnectGetCapabilities (libvirt.c:2119) ==16392== by 0x408047: cmdCapabilities (virsh.c:2018) ==16392== by 0x40C5F9: vshCommandRun (virsh.c:4032) ==16392== by 0x40E6AF: main (virsh.c:4991) ==16392== Here's the patch: * src/virsh.c (cmdCapabilities): Plug a small leak. diff --git a/src/virsh.c b/src/virsh.c index c09dc8d..0d93fb6 100644 --- a/src/virsh.c +++ b/src/virsh.c @@ -2020,6 +2020,7 @@ cmdCapabilities (vshControl * ctl, vshCmd * cmd ATTRIBUTE_UNUSED) return FALSE; } vshPrint (ctl, "%s\n", caps); + free (caps); return TRUE; } -- 1.5.4.rc5.1.g0fa73

3 2

Re: [Libvir] Libvirt on windows
by Richard W.M. Jones 31 Jan '08

31 Jan '08

Gabriel Kaufmann wrote: > I tool libvirt-0.dll and I am using it in VC. > In order to do this, I had to dynamically load the DLL and map all its functions. This is somewhat outside my experience. My best suggestion (guess) is to try adding __declspec(dllexport) around some libvirt functions to see if this causes them to be exported to VC++-compiled programs. On gcc declspec doesn't appear to be necessary. See also: http://www.cygwin.com/ml/cygwin/2000-05/msg00430.html 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] Re: libvirt on mingw
by Richard W.M. Jones 30 Jan '08

30 Jan '08

Brecht Sanders wrote: > Hello again, > Note that I am using gcc (mingw) 3.4.5, and not 4 as you recommend. > The attached patch got rid of the warnings though. > Maybe there is a compiler switch to fix it in a cleaner way though. > Anyway, with the attached patch it does compile. > Basically it casts to void* before converting to a different type. Cheers for the patch, I'll take a look at it tomorrow ... There is a reason why you need to use gcc 4 although it's been so long since I did this that I can't remember exactly why now. Possibly it was some compiler feature that we use in 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] avoid virsh hang due to missing virDomainFree(dom) call
by Jim Meyering 30 Jan '08

30 Jan '08

Without the following patch, this command would hang printf 'domuuid fc4\ndomstate fc4\n' \ | ./virsh --connect test://$PWD/../docs/testnode.xml with this stack trace: __lll_lock_wait ... _L_lock_105 ... __pthread_mutex_lock ... virUnrefDomain (domain=0x6a8b30) at hash.c:884 virDomainFree (domain=0x6a8b30) at libvirt.c:1211 cmdDomstate (ctl=0x7fffea723390, cmd=0x6a8b10) at virsh.c:677 vshCommandRun (ctl=0x7fffea723390, cmd=0x6a8b10) at virsh.c:4033 main (argc=3, argv=0x7fffea7234e8) at virsh.c:5015 The problem is that cmdDomuuid didn't call virDomainFree(dom), so cmdDomstate hangs trying to do the Unref. I then did a mind-numbing audit of all of the other cmd* functions in virsh.c and found two other cases in which a "dom" may not be released. Here's the patch: Avoid virsh hang due to missing virDomainFree(dom) calls * src/virsh.c (cmdDomuuid): Add missing virDomainFree call. (cmdAttachDevice): Likewise. (cmdDetachDevice): Likewise. --- src/virsh.c | 9 +++++++-- 1 files changed, 7 insertions(+), 2 deletions(-) diff --git a/src/virsh.c b/src/virsh.c index 0d93fb6..487f256 100644 --- a/src/virsh.c +++ b/src/virsh.c @@ -2161,6 +2161,7 @@ cmdDomuuid(vshControl * ctl, vshCmd * cmd) else vshError(ctl, FALSE, "%s", _("failed to get domain UUID")); + virDomainFree(dom); return TRUE; } @@ -3038,8 +3039,10 @@ cmdAttachDevice(vshControl * ctl, vshCmd * cmd) return FALSE; } - if (virFileReadAll(from, VIRSH_MAX_XML_FILE, &buffer) < 0) + if (virFileReadAll(from, VIRSH_MAX_XML_FILE, &buffer) < 0) { + virDomainFree(dom); return FALSE; + } ret = virDomainAttachDevice(dom, buffer); free (buffer); @@ -3092,8 +3095,10 @@ cmdDetachDevice(vshControl * ctl, vshCmd * cmd) return FALSE; } - if (virFileReadAll(from, VIRSH_MAX_XML_FILE, &buffer) < 0) + if (virFileReadAll(from, VIRSH_MAX_XML_FILE, &buffer) < 0) { + virDomainFree(dom); return FALSE; + } ret = virDomainDetachDevice(dom, buffer); free (buffer); -- 1.5.4.rc5.1.g0fa73

2 5

[Libvir] plug small leaks upon failure
by Jim Meyering 30 Jan '08

30 Jan '08

with more than 20 domains, I get this: $ valgrind --leak-check=full ./virsh --connect test://$PWD/../docs/testnode.xml list ... error: failed to connect to the hypervisor ... ==13756== 5 bytes in 1 blocks are definitely lost in loss record 1 of 32 ==13756== at 0x4A059F6: malloc (vg_replace_malloc.c:149) ==13756== by 0x387BC7CC11: strdup (strdup.c:43) ==13756== by 0x41D9F2: virXPathString (xml.c:486) ==13756== by 0x4174D3: testLoadDomain (test.c:248) ==13756== by 0x417BA4: testLoadDomainFromFile (test.c:401) ==13756== by 0x418E57: testOpenFromFile (test.c:797) ==13756== by 0x4192CE: testOpen (test.c:913) ==13756== by 0x40FA20: do_open (libvirt.c:572) ==13756== by 0x40FDE4: virConnectOpenAuth (libvirt.c:681) ==13756== by 0x40D5B1: vshInit (virsh.c:4464) ==13756== by 0x40E67C: main (virsh.c:4985) ==13756== ==13756== ==13756== 12,996 (352 direct, 12,644 indirect) bytes in 1 blocks are definitely lost in loss record 31 of 32 ==13756== at 0x4A059F6: malloc (vg_replace_malloc.c:149) ==13756== by 0x38864880E6: xmlXPathNewContext (in /usr/lib64/libxml2.so.2.6.31) ==13756== by 0x417494: testLoadDomain (test.c:242) ==13756== by 0x417BA4: testLoadDomainFromFile (test.c:401) ==13756== by 0x418E57: testOpenFromFile (test.c:797) ==13756== by 0x4192CE: testOpen (test.c:913) ==13756== by 0x40FA20: do_open (libvirt.c:572) ==13756== by 0x40FDE4: virConnectOpenAuth (libvirt.c:681) ==13756== by 0x40D5B1: vshInit (virsh.c:4464) ==13756== by 0x40E67C: main (virsh.c:4985) ==13756== ==13756== LEAK SUMMARY: ==13756== definitely lost: 357 bytes in 2 blocks. ==13756== indirectly lost: 12,644 bytes in 32 blocks. ==13756== possibly lost: 0 bytes in 0 blocks. ==13756== still reachable: 20,614 bytes in 362 blocks. ==13756== suppressed: 0 bytes in 0 blocks. ==13756== Reachable blocks (those to which a pointer was found) are not shown. ==13756== To see them, rerun with: --leak-check=full --show-reachable Here's the fix: * test.c (testLoadDomain): Avoid leaks upon failure. --- src/test.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/test.c b/src/test.c index fe5da41..ab83f65 100644 --- a/src/test.c +++ b/src/test.c @@ -327,7 +327,7 @@ static int testLoadDomain(virConnectPtr conn, } } if (handle < 0) - return (-1); + goto error; privconn->domains[handle].active = 1; privconn->domains[handle].id = domid; -- 1.5.4.rc5.1.g0fa73

2 1

[Libvir] [PATCH] typo
by Atsushi SAKAI 30 Jan '08

30 Jan '08

Hi, I finally succeed for building libvirt on windows(MinGW/MSYS). Thank you for coopration. During that work I found several typo in libvirt document. Thanks Atsushi SAKAI

2 5

[Libvir] tiny patch, FYI: Let "make syntax-check" pass.
by Jim Meyering 30 Jan '08

30 Jan '08

FYI I've just checked this in: Let "make syntax-check" pass. * .x-sc_avoid_if_before_free: Exempt ChangeLog. * tests/statstest.c: Include <config.h>, not "config.h". diff --git a/.x-sc_avoid_if_before_free b/.x-sc_avoid_if_before_free index 627dcdf..f83ae7b 100644 --- a/.x-sc_avoid_if_before_free +++ b/.x-sc_avoid_if_before_free @@ -2,3 +2,4 @@ ^gnulib/lib/printf-parse\.c$ ^gnulib/lib/vasnprintf\.c$ ^build-aux/find-unnecessary-if-before-free$ +^ChangeLog$ diff --git a/tests/statstest.c b/tests/statstest.c index 7fa6f46..4c19833 100644 --- a/tests/statstest.c +++ b/tests/statstest.c @@ -1,4 +1,4 @@ -#include "config.h" +#include <config.h> #include <stdio.h> #include <stdlib.h>

1 0

Fwd: Re: [Libvir] update after changing XML
by Richard W.M. Jones 29 Jan '08

29 Jan '08

-- 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] VIR_DOMAIN_NOSTATE
by Russ Blaine 29 Jan '08

29 Jan '08

On Solaris-based xen 3.1.2 I am seeing virDomainGetInfo() often returning VIR_DOMAIN_NOSTATE in dominfo.state for domUs that are powered on and running. What exactly does VIR_DOMAIN_NOSTATE mean? Why does it exist? I suspect it is a catch-all in the API for hosts that don't entirely have their act together, that do not always return a valid guest state. Is that the case? Is anyone else seeing this on Xen? -- ----------------------------------------------------- Russ Blaine | Solaris Kernel | russell.blaine(a)sun.com

4 10