March 2011 - Devel - libvirt List Archives

[libvirt] TPM support in libvirt
by Stefan Berger 31 Mar '11

31 Mar '11

Hello! I'd like to start a discussion and get some feedback on support for TPM (Trusted Platform Module) management in libvirt. I have been posting patches to the Qemu and SeaBIOS mailing lists that will provide the ground work for TPM support in Qemu, though I anticipate that it will take a while until the code goes into those repositories. So everything I am saying related to command lines etc. can still change. Nevertheless, the next obvious layer for support is libvirt. Well, and I do have patches... First some background: the TPM is a crypto device built into many modern laptops and desktops. It's a chip connected to the motherboard and has firmware for running the TPM specifications. I would say it has a full processor on its own to run this firmware. It also has some very limited space for persistent storage to permanently store for example the owner's password (that he can establish with the TPM), store private keys and it also has some area for NVRAM spaces where the user can put any data into it he wants to. So now, on the qemu command line the TPM support (currently) looks as follows: ... -tpm ? displays: Supported TPM types (choose only one): builtin Qemu's built-in TPM; requires 63kb of block storage I use this for reading the size of the necessary (QCoW2) blockstorage from it, 63kb. To start a VM with TPM support: ... -tpm type=builtin,path=/tmp/tpmstate.bin The above indicates the type of TPM to use. Currently there is only one type available, which is the 'builtin' one. It also provides a path to persistent storage, here /tmp/tpmstate.bin. The persistent storage file is actually a Qcow2 file so we can properly support snapshotting of the image and the TPM and go back to different states. Obviously we are now adding another 'disk' to the VM, so that the TPM can store its own persistent data. Now a user has to make sure to always keep his image file and this TPM 'disk' together, particularly once he stores vital keys in the TPM. So now for modeling this in libvirt I thought of this XML here for a user-provided Qcow2 image: <tpm type='built-in'> <storage file='/tmp/tpmstate.bin'/> </tpm> In this case the user would create the above QCoW2 /tmp/tmpstate.bin at the appropriate size (63kb). This XML here is for a libvirt-created QCoW2 image: <tpm type='built-in'> <storage/> </tpm> This then causes libvirt to create a QCoW2 image of 63kb size (reads the 63kb from .. tpm ?) once the VM is about to start and produces for example this command line: -tpm type=builtin,path=/var/lib/libvirt/tpm/a4d7cd22-da89-3094-6212-079a48a309a1.bin The UUID then corresponds to the UUID of the VM. Since only one TPM per VM is supported, this should work just fine. Now to support an encrypted QCoW2 image, the following XML could be used: <tpm type='built-in'> <storage> <encryption format='qcow'> <secret type='passphrase' uuid='13ea49f7-2553-7308-5168-e337ade36232'/> </encryption> </storage> </tpm> Here the user has created a 'secret' and passes that secret's UUID into the XML above. So this is similar to how storage is dealt with with the difference that no path to the storage is provided and also internally no comparisons for storage paths are being made. I did also not want to force the user to create storage spaces (volumes) on his own, but let libvirt handle this implicitly for the TPM (use qemu-img to create the QCoW2). Also, multiple VMs' TPMs could share the same secret. Please let me know of any comments regarding this. I'll post the patches as RFC later on. Regards, Stefan

2 2

[libvirt] [PATCH] Fix domain events C example on Win32
by Daniel P. Berrange 31 Mar '11

31 Mar '11

printf on Win32 does not neccessarily support %lld and we don't have GNULIBs wrapper for printf(). Switch to use asprintf() for which we do have a gnulib wrapper with %lld support * examples/domain-events/events-c/event-test.c: Fix formatting of %lld on Win32 * cfg.mk: Don't require use of virAsprintf since this is an example app for out of tree users to follow --- cfg.mk | 2 +- examples/domain-events/events-c/event-test.c | 10 ++++++++-- 2 files changed, 9 insertions(+), 3 deletions(-) diff --git a/cfg.mk b/cfg.mk index ac419f7..0440425 100644 --- a/cfg.mk +++ b/cfg.mk @@ -564,7 +564,7 @@ exclude_file_name_regexp--sc_prohibit_always_true_header_tests = \ (^docs|^python/(libvirt-override|typewrappers)\.c$$) exclude_file_name_regexp--sc_prohibit_asprintf = \ - ^(bootstrap.conf$$|po/|src/util/util\.c$$) + ^(bootstrap.conf$$|po/|src/util/util\.c$$|examples/domain-events/events-c/event-test\.c$$) exclude_file_name_regexp--sc_prohibit_close = \ (\.py$$|^docs/|(src/util/files\.c|src/libvirt\.c)$$) diff --git a/examples/domain-events/events-c/event-test.c b/examples/domain-events/events-c/event-test.c index 7d05dd8..44e94ec 100644 --- a/examples/domain-events/events-c/event-test.c +++ b/examples/domain-events/events-c/event-test.c @@ -169,8 +169,14 @@ static int myDomainEventRTCChangeCallback(virConnectPtr conn ATTRIBUTE_UNUSED, long long offset, void *opaque ATTRIBUTE_UNUSED) { - printf("%s EVENT: Domain %s(%d) rtc change %lld\n", __func__, virDomainGetName(dom), - virDomainGetID(dom), offset); + char *str = NULL; + /* HACK: use asprintf since we have gnulib's wrapper for %lld on Win32 + * but don't have a printf() replacement with %lld */ + if (asprintf(&str, "%s EVENT: Domain %s(%d) rtc change %lld\n", __func__, virDomainGetName(dom), + virDomainGetID(dom), offset) < 0) + return 0; + + printf("%s", str); return 0; } -- 1.7.4

2 2

[libvirt] [PATCH] maint: avoid locale-sensitivity in string case comparisons
by Eric Blake 31 Mar '11

31 Mar '11

strcase{cmp/str} have the drawback of being sensitive to the global locale; this is unacceptable in a library setting. Prefer a hard-coded C locale alternative for all but virsh, which is user facing and where the global locale isn't changing externally. * .gnulib: Update to latest, for c-strcasestr change. * bootstrap.conf (gnulib_modules): Drop strcasestr, add c-strcase and c-strcasestr. * cfg.mk (sc_avoid_strcase): New rule. (exclude_file_name_regexp--sc_avoid_strcase): New exception. * src/internal.h (STRCASEEQ, STRCASENEQ, STRCASEEQLEN) (STRCASENEQLEN): Adjust offenders. * src/qemu/qemu_monitor_text.c (qemuMonitorTextEjectMedia): Likewise. * tools/virsh.c (namesorter): Document exception. --- Inspired by today's earlier patch that started using strcasester. This also goes along with our policy of no ctype, just c-ctype. .gnulib | 2 +- bootstrap.conf | 3 ++- cfg.mk | 7 +++++++ src/internal.h | 10 ++++++---- src/qemu/qemu_monitor_text.c | 3 ++- tools/virsh.c | 1 + 6 files changed, 19 insertions(+), 7 deletions(-) diff --git a/.gnulib b/.gnulib index 422ab2e..790645d 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit 422ab2e0d70ed348e2fd0a82558be38e5859011a +Subproject commit 790645d837f8084991421107fba639b110d58335 diff --git a/bootstrap.conf b/bootstrap.conf index 6e10828..733c354 100644 --- a/bootstrap.conf +++ b/bootstrap.conf @@ -22,6 +22,8 @@ gnulib_modules=' areadlink base64 c-ctype +c-strcase +c-strcasestr canonicalize-lgpl chown close @@ -63,7 +65,6 @@ sigpipe snprintf socket stpcpy -strcasestr strchrnul strndup strerror diff --git a/cfg.mk b/cfg.mk index ac419f7..f802cee 100644 --- a/cfg.mk +++ b/cfg.mk @@ -349,6 +349,11 @@ sc_avoid_ctype_macros: halt="don't use ctype macros (use c-ctype.h)" \ $(_sc_search_regexp) +sc_avoid_strcase: + @prohibit='\bstrn?case(cmp|str) *\(' \ + halt="don't use raw strcase functions (use c-strcase instead)" \ + $(_sc_search_regexp) + sc_prohibit_virBufferAdd_with_string_literal: @prohibit='\<virBufferAdd *\([^,]+, *"[^"]' \ halt='use virBufferAddLit, not virBufferAdd, with a string literal' \ @@ -547,6 +552,8 @@ _makefile_at_at_check_exceptions = ' && !/(SCHEMA|SYSCONF)DIR/' syntax-check: $(top_srcdir)/HACKING # List all syntax-check exemptions: +exclude_file_name_regexp--sc_avoid_strcase = ^tools/virsh\.c$$ + _src1=libvirt|fdstream|qemu/qemu_monitor|util/(command|util)|xen/xend_internal exclude_file_name_regexp--sc_avoid_write = \ ^(src/($(_src1))|daemon/libvirtd|tools/console)\.c$$ diff --git a/src/internal.h b/src/internal.h index be97801..2afbd8d 100644 --- a/src/internal.h +++ b/src/internal.h @@ -44,6 +44,8 @@ # include "libvirt_internal.h" +# include "c-strcase.h" + /* On architectures which lack these limits, define them (ie. Cygwin). * Note that the libvirt code should be robust enough to handle the * case where actual value is longer than these limits (eg. by setting @@ -64,13 +66,13 @@ /* String equality tests, suggested by Jim Meyering. */ # define STREQ(a,b) (strcmp(a,b) == 0) -# define STRCASEEQ(a,b) (strcasecmp(a,b) == 0) +# define STRCASEEQ(a,b) (c_strcasecmp(a,b) == 0) # define STRNEQ(a,b) (strcmp(a,b) != 0) -# define STRCASENEQ(a,b) (strcasecmp(a,b) != 0) +# define STRCASENEQ(a,b) (c_strcasecmp(a,b) != 0) # define STREQLEN(a,b,n) (strncmp(a,b,n) == 0) -# define STRCASEEQLEN(a,b,n) (strncasecmp(a,b,n) == 0) +# define STRCASEEQLEN(a,b,n) (c_strncasecmp(a,b,n) == 0) # define STRNEQLEN(a,b,n) (strncmp(a,b,n) != 0) -# define STRCASENEQLEN(a,b,n) (strncasecmp(a,b,n) != 0) +# define STRCASENEQLEN(a,b,n) (c_strncasecmp(a,b,n) != 0) # define STRPREFIX(a,b) (strncmp(a,b,strlen(b)) == 0) # define STRSKIP(a,b) (STRPREFIX(a,b) ? (a) + strlen(b) : NULL) diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index e0e3292..168c60f 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -33,6 +33,7 @@ #include "qemu_monitor_text.h" #include "qemu_command.h" #include "c-ctype.h" +#include "c-strcasestr.h" #include "memory.h" #include "logging.h" #include "driver.h" @@ -934,7 +935,7 @@ int qemuMonitorTextEjectMedia(qemuMonitorPtr mon, /* If the command failed qemu prints: * device not found, device is locked ... * No message is printed on success it seems */ - if (strcasestr(reply, "device ")) { + if (c_strcasestr(reply, "device ")) { qemuReportError(VIR_ERR_OPERATION_FAILED, _("could not eject media on %s: %s"), devname, reply); goto cleanup; diff --git a/tools/virsh.c b/tools/virsh.c index faeaf47..3c759b9 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -432,6 +432,7 @@ static int namesorter(const void *a, const void *b) { const char **sa = (const char**)a; const char **sb = (const char**)b; + /* User visible sort, so we want locale-specific case comparison. */ return strcasecmp(*sa, *sb); } -- 1.7.4

2 2

[libvirt] [RFC] Add flag for virsh undefine to remove/wipe the disk devices
by Osier Yang 31 Mar '11

31 Mar '11

Hi, All I'm thinking to introduce a new flag (something like --remove-disks, --wipe-disks) for "virsh undefine", so that the user can choose whether to remove/wipe the disk devices or not, have seen this requirement in many places, @libvirt-users, public #virt, and also we have a bug of this function. So, IMHO this is a reasonable requirement, following is the rough thoughts: 1) General idea. As we don't have a API which can get all the disk devices of a domain, perhaps need to write functions to parse domain xml to extract the disks' path (this is annoyed, but seems don't other way), and then lookup them by storage volume API (virStorageVolLookupByPath), and then can remove or wipe the volume by (virStorageVolDelete/virStorageVolWipe). And for the disk path which doesn't belong to any storage pool, simply remove it by "unlink()"? 2) Which type of devices can not be removed/wiped. * Can't delete/wipe ISCSI/SCSI vol. * Vol doesn't exists (which will throw an warning when do virStorageVolLookupByPath). * Have no write permission on the parent directory of the disk path. * Can't delete/wipe the disk device which is passthrough'ed from host, (e.g. /dev/sr0 as a CDROM device for guest) * The storage pool which the disk device belongs to as a vol is marked as "share" * The storage pool which the disk device belongs as a vol is readonly * can't delete disk device of network type. * Any others? For these situations, we need to do checking and throw straightforward warnings to tell user why it can't be removed/wiped. Any idea is welcomed. Thanks. CC'ed Eric, as I saw you talked about this somewhere. (#virt?), :-) Regards Osier

6 13

[libvirt] [libvirt-php][PATCH] Fix virConnectClose failed issue on list_domains
by warp.kawada＠gmail.com 31 Mar '11

31 Mar '11

This patch for libvirt-php Pre-4.2 version. I checked with libvirt 0.8.8 version. I fixed irConnectClose failed issue on list_domains. And web example codes updated for 0.8.8 API. This issue is: $domains = libvirt_list_domains($conn); PHP Warning: main(): virConnectClose failed with 1 on destructor: (null) And if ssh connect to remote host then some netcat processes remain on remote host. # ps axu|grep libvirt|grep 'nc -U' apache 27867 0.0 0.0 7332 368 ? Ss 12:09 0:00 nc -U /var/run/libvirt/libvirt-sock apache 27871 0.0 0.0 7332 372 ? Ss 12:09 0:00 nc -U /var/run/libvirt/libvirt-sock apache 27875 0.0 0.0 7332 368 ? Ss 12:09 0:00 nc -U /var/run/libvirt/libvirt-sock Thanks for libvirt-php. Regards, == Yukihiro Kawada diff --git a/examples/libvirt.php b/examples/libvirt.php index 10e7e38..8a54acd 100644 --- a/examples/libvirt.php +++ b/examples/libvirt.php @@ -22,7 +22,7 @@ } function get_hostname() { - return libvirt_get_hostname($this->conn); + return libvirt_connect_get_hostname($this->conn); } function get_domain_object($nameRes) { @@ -178,30 +178,23 @@ } function get_uri() { - return libvirt_get_uri($this->conn); + return libvirt_connect_get_uri($this->conn); } function get_domain_count() { - $ac = libvirt_get_active_domain_count($this->conn); - $ic = libvirt_get_inactive_domain_count($this->conn); - $tc = libvirt_get_domain_count($this->conn); - - return array( - 'active' => $ac, - 'inactive' => $ic, - 'total' => $tc - ); + return libvirt_domain_get_counts($this->conn); } function get_domains() { $domNames = array(); $doms = libvirt_list_domains($this->conn); - foreach ($doms as $dom) { - $tmp = libvirt_domain_get_uuid_string($dom); - $domNames[$tmp] = libvirt_domain_get_name($dom); + foreach ($doms as $nam) { + $dom = libvirt_domain_lookup_by_name($this->conn, $nam); + $tmp = libvirt_domain_get_uuid_string($dom); + $domNames[$tmp] = $nam; } - + unset($doms); ksort($domNames); return $domNames; } @@ -316,7 +309,7 @@ $ret = array(); if ($name != false) { - $dom=libvirt_domain_lookup_by_name($this->conn, $name); + $dom = libvirt_domain_lookup_by_name($this->conn, $name); if (!$dom) return false; diff --git a/src/libvirt-php.cb/src/libvirt-php.c index 5e77994..f5a927b 100644 --- a/src/libvirt-php.c +++ b/src/libvirt-php.c @@ -3467,7 +3467,7 @@ PHP_FUNCTION(libvirt_list_domains) int *ids; char **names; const char *name; - int i; + int i, rv; virDomainPtr domain=NULL; GET_CONNECTION_FROM_ARGS("r",&zconn); @@ -3490,6 +3490,12 @@ PHP_FUNCTION(libvirt_list_domains) } efree(ids); + rv = virDomainFree (domain); // Y.Kawada + if (rv != 0) { + php_error_docref(NULL TSRMLS_CC, E_WARNING,"virDomainFree failed with %i on list_domain: %s", rv, LIBVIRT_G (last_error)); + } + domain = NULL; + expectedcount=virConnectNumOfDefinedDomains (conn->conn); names=emalloc(expectedcount*sizeof(char *)); count=virConnectListDefinedDomains (conn->conn,names ,expectedcount);

3 2

[libvirt] daily report
by Wen Congyang 31 Mar '11

31 Mar '11

- discuss the bug of libvirt that will cause libvirtd crashed.

1 1

[libvirt] Can't get connection to libvirt if one of the guests' monitor is occupied by other app
by Osier Yang 31 Mar '11

31 Mar '11

Hi, As $subject, e.g. the monitor is being used by netcat. Is it expected? Should we only skip to reconnect the domain monitor which is being locked by other app with a timeout specified on "qemuMonitorOpenUnix" (it blocks when trying to connect to the unix socket)? so that it can get a connection, and could manage other domains. How to produce: 1) # service libvirtd stop 2) # nc -U /var/lib/libvirt/qemu/$guest.monitor 3) # virsh (hangs forever if nc process is not killed) Regards Osier

2 1

[libvirt] [PATCH] qemu: Ignore libvirt debug messages in qemu log
by Jiri Denemark 31 Mar '11

31 Mar '11

qemu driver uses a 4K buffer for reading qemu log file. This is enough when only qemu's output is present in the log file. However, when debugging messages are turned on, intermediate libvirt process fills the log with a bunch of debugging messages before it executes qemu binary. In such a case the buffer may become too small. However, we are not really interested in libvirt messages so they can be filtered out from the buffer. --- src/qemu/qemu_process.c | 47 +++++++++++++++++++++++++++++++++++++---------- 1 files changed, 37 insertions(+), 10 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index e31e1b4..bc117a4 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -762,11 +762,21 @@ qemuProcessReadLogOutput(virDomainObjPtr vm, { int retries = (timeout*10); int got = 0; + char *debug = NULL; + int ret = -1; + char *filter_next = buf; + buf[0] = '\0'; + if (virAsprintf(&debug, ": %d: debug : ", vm->pid) < 0) { + virReportOOMError(); + return -1; + } + while (retries) { - ssize_t func_ret, ret; + ssize_t func_ret, bytes; int isdead = 0; + char *eol; func_ret = func(vm, buf, fd); @@ -775,32 +785,46 @@ qemuProcessReadLogOutput(virDomainObjPtr vm, /* Any failures should be detected before we read the log, so we * always have something useful to report on failure. */ - ret = saferead(fd, buf+got, buflen-got-1); - if (ret < 0) { + bytes = saferead(fd, buf+got, buflen-got-1); + if (bytes < 0) { virReportSystemError(errno, _("Failure while reading %s log output"), what); - return -1; + goto cleanup; } - got += ret; + got += bytes; buf[got] = '\0'; + + /* Filter out debug messages from intermediate libvirt process */ + while ((eol = strchr(filter_next, '\n'))) { + char *p = strstr(filter_next, debug); + if (p && p < eol) { + memmove(filter_next, eol + 1, got - (eol - buf)); + got -= eol + 1 - filter_next; + } else { + filter_next = eol + 1; + } + } + if (got == buflen-1) { qemuReportError(VIR_ERR_INTERNAL_ERROR, _("Out of space while reading %s log output: %s"), what, buf); - return -1; + goto cleanup; } if (isdead) { qemuReportError(VIR_ERR_INTERNAL_ERROR, _("Process exited while reading %s log output: %s"), what, buf); - return -1; + goto cleanup; } - if (func_ret <= 0) - return func_ret; + if (func_ret <= 0) { + ret = func_ret; + goto cleanup; + } usleep(100*1000); retries--; @@ -809,7 +833,10 @@ qemuProcessReadLogOutput(virDomainObjPtr vm, qemuReportError(VIR_ERR_INTERNAL_ERROR, _("Timed out while reading %s log output: %s"), what, buf); - return -1; + +cleanup: + VIR_FREE(debug); + return ret; } -- 1.7.4.1

2 2

[libvirt] How to get the IP address of a Domain?
by 徐滕 30 Mar '11

30 Mar '11

HI,all: I'm new to libvirt, and have some questions about How to get the IP address of a Domain? I searched the old mails int the mail-list ,found one way to do this,it is first get the mac address of the Domain,then used some sniffer or arp tools to fetch the IP address , but: Q1: Is there any APIs of libvirt to get the domain's IP address directly? If does not exist now , have any plan to add this api or some extra tools to do this ,just like the vm-tools on VMware ESX? Q2: How to get the domain's mac? I download the virt-manager src, searching it's source to find how to do this : in details.py , this line seems to get the mac address: in function update_hwlist(hwtype, info, name, icon_name): Line 2475 for net in self.vm.get_network_devices(): mac = net.macaddr update_hwlist(HW_LIST_TYPE_NIC, net, "NIC %s" % mac[-9:], "network-idle") then ,I looked into domains.py, in function get_network_devices(): def get_network_devices(self, refresh_if_necc=True): return self._build_device_list("interface", refresh_if_necc) next, go to function _build_device_list() def _build_device_list(self, device_type, refresh_if_necc=True, inactive=False): guest = self._get_guest(refresh_if_necc=refresh_if_necc, inactive=inactive) devs = guest.get_devices(device_type) count = 0 for dev in devs: dev.vmmindex = count count += 1 return devs finally ,in funciton _get_guest(self, inactive=False, refresh_if_necc=True) def _get_guest(self, inactive=False, refresh_if_necc=True): raise NotImplementedError() this puzzled me , only raise a exception NotImplentedError(),but not deal with it . How to get the mac address? Sorry ,I 'm new to Python either... Really appreciate for your help! wade

4 3

[libvirt] [PATCH] docs: mention C89 syntax preferences
by Eric Blake 30 Mar '11

30 Mar '11

* docs/hacking.html.in (Code formatting): Document that // comment and declaration-after-statement are discouraged. * HACKING: Regenerate. --- Any comments on the wording chosen here? Should this go in for 0.9.0? HACKING | 6 ++++++ docs/hacking.html.in | 10 ++++++++++ 2 files changed, 16 insertions(+), 0 deletions(-) diff --git a/HACKING b/HACKING index b2ceaa7..686328c 100644 --- a/HACKING +++ b/HACKING @@ -127,6 +127,12 @@ Note that sometimes you'll have to post-process that output further, by piping it through "expand -i", since some leading TABs can get through. Usually they're in macro definitions or strings, and should be converted anyhow. +Libvirt requires a C99 compiler for various reasons. However, most of the code +base prefers to stick to C89 syntax unless there is a compelling reason +otherwise. For example, it is preferable to use "/* */" comments rather than +"//". Also, when declaring local variables, the prevailing style has been to +declare them at the beginning of a scope, rather than immediately before use. + Curly braces ============ diff --git a/docs/hacking.html.in b/docs/hacking.html.in index e0484dd..dd8eeb8 100644 --- a/docs/hacking.html.in +++ b/docs/hacking.html.in @@ -152,6 +152,16 @@ anyhow. </p> + <p> + Libvirt requires a C99 compiler for various reasons. However, + most of the code base prefers to stick to C89 syntax unless + there is a compelling reason otherwise. For example, it is + preferable to use <code>/* */</code> comments rather + than <code>//</code>. Also, when declaring local variables, the + prevailing style has been to declare them at the beginning of a + scope, rather than immediately before use. + </p> + <h2><a name="curly_braces">Curly braces</a></h2> -- 1.7.4

2 2