January 2008 - Devel - libvirt List Archives

[Libvir] [PATCH] Clean up global name space in examples and tests.
by Jim Meyering 14 Jan '08

14 Jan '08

These don't really matter, other than to remove false-positives if/when we ever automate checks for unnecessarily-global symbols. Clean up global name space in examples and tests. * docs/examples/suspend.c: Declare global "conn" to be static. * tests/qemuxml2argvtest.c: Declare global "driver" to be static. * tests/qemuxml2xmltest.c: Likewise. diff --git a/docs/examples/suspend.c b/docs/examples/suspend.c index 80e4da7..9ca6299 100644 --- a/docs/examples/suspend.c +++ b/docs/examples/suspend.c @@ -15,7 +15,7 @@ #include <stdio.h> #include <libvirt/libvirt.h> -virConnectPtr conn = NULL; /* the hypervisor connection */ +static virConnectPtr conn = NULL; /* the hypervisor connection */ /** * checkDomainState: diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index d4e39b2..22e2e81 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -16,7 +16,7 @@ static char *progname; static char *abs_top_srcdir; -struct qemud_driver driver; +static struct qemud_driver driver; #define MAX_FILE 4096 diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index c42e6fb..2ab1696 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -16,7 +16,7 @@ static char *progname; static char *abs_top_srcdir; -struct qemud_driver driver; +static struct qemud_driver driver; #define MAX_FILE 4096 -- 1.5.4.rc1.11.gd2f82

3 2

[Libvir] [PATCH] Clean up the global name space.
by Jim Meyering 14 Jan '08

14 Jan '08

After mentioning "avoid global variables" as a potential addition to HACKING, I did some quick tests and spotted a few that are easy to eliminate. This is useful even if we don't end up automating the check: Clean up the global name space. * src/qemu_conf.c: Add "static" and "const" attributes to some globals. * src/qemu_conf.h: Update a declaration. * src/qemu_driver.c (qemud_driver): Declare static. diff --git a/src/qemu_conf.c b/src/qemu_conf.c index 15779b7..d03d989 100644 --- a/src/qemu_conf.c +++ b/src/qemu_conf.c @@ -231,35 +231,35 @@ void qemudFreeVM(struct qemud_vm *vm) { /* The list of possible machine types for various architectures, as supported by QEMU - taken from 'qemu -M ?' for each arch */ -static const char *arch_info_x86_machines[] = { +static const char *const arch_info_x86_machines[] = { "pc", "isapc", NULL }; -static const char *arch_info_mips_machines[] = { +static const char *const arch_info_mips_machines[] = { "mips", NULL }; -static const char *arch_info_sparc_machines[] = { +static const char *const arch_info_sparc_machines[] = { "sun4m", NULL }; -static const char *arch_info_ppc_machines[] = { +static const char *const arch_info_ppc_machines[] = { "g3bw", "mac99", "prep", NULL }; /* Feature flags for the architecture info */ -struct qemu_feature_flags arch_info_i686_flags [] = { +static const struct qemu_feature_flags const arch_info_i686_flags [] = { { "pae", 1, 1 }, { "acpi", 1, 1 }, { "apic", 1, 0 }, { NULL, -1, -1 } }; -struct qemu_feature_flags arch_info_x86_64_flags [] = { +static const struct qemu_feature_flags const arch_info_x86_64_flags [] = { { "acpi", 1, 1 }, { "apic", 1, 0 }, { NULL, -1, -1 } }; /* The archicture tables for supported QEMU archs */ -struct qemu_arch_info qemudArchs[] = { +const struct qemu_arch_info const qemudArchs[] = { /* i686 must be in position 0 */ { "i686", 32, arch_info_x86_machines, "qemu", arch_info_i686_flags }, /* x86_64 must be in position 1 */ diff --git a/src/qemu_conf.h b/src/qemu_conf.h index 69342cb..e2c7514 100644 --- a/src/qemu_conf.h +++ b/src/qemu_conf.h @@ -426,11 +426,11 @@ struct qemu_feature_flags { struct qemu_arch_info { const char *arch; int wordsize; - const char **machines; + const char *const *machines; const char *binary; const struct qemu_feature_flags *fflags; }; -extern struct qemu_arch_info qemudArchs[]; +extern const struct qemu_arch_info const qemudArchs[]; #endif /* WITH_QEMU */ diff --git a/src/qemu_driver.c b/src/qemu_driver.c index c96fb45..dee6ba6 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -107,7 +107,7 @@ static int qemudShutdownNetworkDaemon(virConnectPtr conn, struct qemud_driver *driver, struct qemud_network *network); -struct qemud_driver *qemu_driver = NULL; +static struct qemud_driver *qemu_driver = NULL; static -- 1.5.4.rc1.11.gd2f82

3 2

Re: [Libvir] [patch 7/9] Add support for lokkit
by Richard W.M. Jones 14 Jan '08

14 Jan '08

+AC_ARG_ENABLE(iptables-lokkit, Just a minor nit here: My reading of the autoconf info page is that this should be a --with argument, not an --enable argument. [...] Optional Features: ... --enable-bar include bar Optional Packages: ... --with-foo use foo [...] 14.2 Working With External Software =================================== Some packages require, or can optionally use, other software packages that are already installed. The user can give `configure' command line options to specify which such external software to use. The options have one of these forms: --with-PACKAGE[=ARG] --without-PACKAGE [...] 14.3 Choosing Package Options ============================= If a software package has optional compile-time features, the user can give `configure' command line options to specify whether to compile them. The options have one of these forms: --enable-FEATURE[=ARG] --disable-FEATURE 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 4

[Libvir] [patch] qemu/kvm: use_system_powerdown instead of killing the vm hard
by Guido Guenther 14 Jan '08

14 Jan '08

Hi, currently domainShutdown kills qemu/kvm instances hard which is not very filesystem friendly. However recent kvm git acquired system_powerdown to shutdown the system gracefully by simulating an acpi power button press. We can now use this in libvirt: diff --git a/src/qemu_driver.c b/src/qemu_driver.c index f792eba..55adb18 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -1849,6 +1849,27 @@ static int qemudDomainResume(virDomainPtr dom) { } +static int qemudDomainShutdown(virDomainPtr dom) { + struct qemud_driver *driver = (struct qemud_driver *)dom->conn->privateData; + struct qemud_vm *vm = qemudFindVMByID(driver, dom->id); + char* info; + + if (!vm) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_DOMAIN, + "no domain with matching id %d", dom->id); + return -1; + } + + if (qemudMonitorCommand(driver, vm, "system_powerdown", &info) < 0) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + "shutdown operation failed"); + return -1; + } + return 0; + +} + + static int qemudDomainDestroy(virDomainPtr dom) { struct qemud_driver *driver = (struct qemud_driver *)dom->conn->privateData; struct qemud_vm *vm = qemudFindVMByID(driver, dom->id); @@ -2855,7 +2876,7 @@ static virDriver qemuDriver = { qemudDomainLookupByName, /* domainLookupByName */ qemudDomainSuspend, /* domainSuspend */ qemudDomainResume, /* domainResume */ - qemudDomainDestroy, /* domainShutdown */ + qemudDomainShutdown, /* domainShutdown */ NULL, /* domainReboot */ qemudDomainDestroy, /* domainDestroy */ qemudDomainGetOSType, /* domainGetOSType */ Please apply, -- Guido

4 14

[Libvir] Error compiling libvirt 0.4.0 on debian etch
by Solli Honorio 14 Jan '08

14 Jan '08

Hi, when I'm trying to compile the libvirt 0.4.0 on Debian etch I'm getting the error below, I have no problem with version 0.3.3. I've look at line 578 of /usr/include/xen/xen.h and this error don't make sense for me. Does any anyone had this problem on debian etch too ? Maybe someone can help-me with this error ? Thanks for attention, Solli Honorio <error> .... In file included from /usr/include/xen/dom0_ops.h:31, from xen_unified.c:29: /usr/include/xen/xen.h:578: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'char' /usr/include/xen/xen.h:579: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'short' /usr/include/xen/xen.h:580: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'int' /usr/include/xen/xen.h:581: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'long' make[2]: *** [libvirt_la-xen_unified.lo] Error 1 make[2]: Leaving directory `/usr/src/libvirt-0.4.0/src' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/usr/src/libvirt-0.4.0' make: *** [all] Error 2 </error>

2 1

[Libvir] [patch 1/1] Reduce the number of times lokkit is run
by Mark McLoughlin 14 Jan '08

14 Jan '08

plain text document attachment (libvirt-iptables-save-rules.patch) With --enable-iptables-lokkit, the rules are saved to disk and lokkit --custom-rules is run each time a single rule is added or removed. This patch moves this code into a new iptablesSaveRules() function so that we can do it for all rules at once. Signed-off-by: Mark McLoughlin <markmc(a)redhat.com> Index: libvirt/src/iptables.c =================================================================== --- libvirt.orig/src/iptables.c 2008-01-04 15:30:31.000000000 +0000 +++ libvirt.orig/src/iptables.c 2008-01-04 15:30:31.000000000 +0000 @@ -282,6 +282,25 @@ writeRules(const char *path, #endif /* ENABLE_IPTABLES_LOKKIT */ static void +iptRulesSave(iptRules *rules) +{ +#ifdef ENABLE_IPTABLES_LOKKIT + int err; + + if ((err = writeRules(rules->path, rules->rules, rules->nrules))) { + qemudLog(QEMUD_WARN, "Failed to saves iptables rules to %s : %s", + rules->path, strerror(err)); + return; + } + + if (rules->nrules > 0) + notifyRulesUpdated(rules->table, rules->path); + else + notifyRulesRemoved(rules->table, rules->path); +#endif /* ENABLE_IPTABLES_LOKKIT */ +} + +static void iptRuleFree(iptRule *rule) { if (rule->rule) @@ -321,20 +340,6 @@ iptRulesAppend(iptRules *rules, rules->nrules++; -#ifdef ENABLE_IPTABLES_LOKKIT - { - int err; - - if ((err = virFileMakePath(rules->dir))) - return err; - - if ((err = writeRules(rules->path, rules->rules, rules->nrules))) - return err; - } - - notifyRulesUpdated(rules->table, rules->path); -#endif /* ENABLE_IPTABLES_LOKKIT */ - return 0; } @@ -359,20 +364,6 @@ iptRulesRemove(iptRules *rules, rules->nrules--; -#ifdef ENABLE_IPTABLES_LOKKIT - { - int err; - - if ((err = writeRules(rules->path, rules->rules, rules->nrules))) - return err; - } - - if (rules->nrules > 0) - notifyRulesUpdated(rules->table, rules->path); - else - notifyRulesRemoved(rules->table, rules->path); -#endif /* ENABLE_IPTABLES_LOKKIT */ - return 0; } @@ -658,6 +649,22 @@ iptablesContextFree(iptablesContext *ctx free(ctx); } +/** + * iptablesSaveRules: + * @ctx: pointer to the IP table context + * + * Saves all the IP table rules associated with a context + * to disk so that if iptables is restarted, the rules + * will automatically be reload. + */ +void +iptablesSaveRules(iptablesContext *ctx) +{ + iptRulesSave(ctx->input_filter); + iptRulesSave(ctx->forward_filter); + iptRulesSave(ctx->nat_postrouting); +} + static void iptRulesReload(iptRules *rules) { Index: libvirt/src/iptables.h =================================================================== --- libvirt.orig/src/iptables.h 2007-12-13 09:05:31.000000000 +0000 +++ libvirt.orig/src/iptables.h 2007-12-13 09:05:31.000000000 +0000 @@ -29,6 +29,7 @@ typedef struct _iptablesContext iptables iptablesContext *iptablesContextNew (void); void iptablesContextFree (iptablesContext *ctx); +void iptablesSaveRules (iptablesContext *ctx); void iptablesReloadRules (iptablesContext *ctx); int iptablesAddTcpInput (iptablesContext *ctx, Index: libvirt/src/qemu_driver.c =================================================================== --- libvirt.orig/src/qemu_driver.c 2007-12-13 09:05:31.000000000 +0000 +++ libvirt.orig/src/qemu_driver.c 2007-12-13 09:05:31.000000000 +0000 @@ -1009,8 +1009,10 @@ qemudAddIptablesRules(virConnectPtr conn /* The remaining rules are only needed for IP forwarding */ - if (!network->def->forward) + if (!network->def->forward) { + iptablesSaveRules(driver->iptables); return 1; + } /* allow forwarding packets from the bridge interface */ if ((err = iptablesAddForwardAllowOut(driver->iptables, @@ -1044,6 +1046,8 @@ qemudAddIptablesRules(virConnectPtr conn goto err10; } + iptablesSaveRules(driver->iptables); + return 1; err10: @@ -1100,6 +1104,7 @@ qemudRemoveIptablesRules(struct qemud_dr iptablesRemoveTcpInput(driver->iptables, network->bridge, 53); iptablesRemoveUdpInput(driver->iptables, network->bridge, 67); iptablesRemoveTcpInput(driver->iptables, network->bridge, 67); + iptablesSaveRules(driver->iptables); } static int --

3 2

[Libvir] PATCH: BZ 427107: fix crash wrt auth callback
by Daniel P. Berrange 14 Jan '08

14 Jan '08

If the application does not supply an authentication callback, and tries to connect to a server with PolicyKit auth turned on it will try to deference a NULL pointer with predictably crashtastic results: https://bugzilla.redhat.com/show_bug.cgi?id=427107 This patch has been tested by bug reporter to fix the issue Index: src/remote_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/remote_internal.c,v retrieving revision 1.49 diff -u -p -r1.49 remote_internal.c --- src/remote_internal.c 17 Dec 2007 17:41:02 -0000 1.49 +++ src/remote_internal.c 31 Dec 2007 20:30:19 -0000 @@ -3347,24 +3347,26 @@ remoteAuthPolkit (virConnectPtr conn, st }; remoteDebug(priv, "Client initialize PolicyKit authentication"); - for (i = 0 ; i < auth->ncredtype ; i++) { - if (auth->credtype[i] == VIR_CRED_EXTERNAL) - allowcb = 1; - } + if (auth && auth->cb) { + /* Check if the neccessary credential type for PolicyKit is supported */ + for (i = 0 ; i < auth->ncredtype ; i++) { + if (auth->credtype[i] == VIR_CRED_EXTERNAL) + allowcb = 1; + } - /* Run the authentication callback */ - if (allowcb) { - if (auth && auth->cb && - (*(auth->cb))(&cred, 1, auth->cbdata) < 0) { - __virRaiseError (in_open ? NULL : conn, NULL, NULL, VIR_FROM_REMOTE, - VIR_ERR_AUTH_FAILED, VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, - "Failed to collect auth credentials"); - return -1; + if (allowcb) { + /* Run the authentication callback */ + if ((*(auth->cb))(&cred, 1, auth->cbdata) < 0) { + __virRaiseError (in_open ? NULL : conn, NULL, NULL, VIR_FROM_REMOTE, + VIR_ERR_AUTH_FAILED, VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, + "Failed to collect auth credentials"); + return -1; + } } else { - remoteDebug(priv, "No auth callback provided for PolicyKit"); + remoteDebug(priv, "Client auth callback does not support PolicyKit"); } } else { - remoteDebug(priv, "Client auth callback does not support PolicyKit"); + remoteDebug(priv, "No auth callback provided"); } memset (&ret, 0, sizeof ret); 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 6

[Libvir] VMC - Virtual Machine Console
by Meng Kuan 14 Jan '08

14 Jan '08

Greetings, I would like to announce the availability of VMC (Virtual Machine Console). VMC is an attempt to provide an opensource, web-based VM management infrastructure. It uses libvirt as the underlying library to manage para-virtualized Xen VMs. You can find out more on our "Introduction to VMC" page: http://www.sxven.com/vmc List of current features and future plans: http://www.sxven.com/vmc/features To get started, we have made available a "VMC Install" document: http://www.sxven.com/vmc/gettingstarted We invite people to take a look at VMC and tell us what you like and what you don't like. If you have any problems, questions or suggestions please feel free to contact us at dev(a)sxven.com or post them on our forum: http://forum.sxven.com/ Best regards, Meng Kuan

1 0

Re: [Libvir] [patch 0/9] Add support for Fedora lokkit's --custom-rules option
by Mark McLoughlin 10 Jan '08

10 Jan '08

Hey, Sounds like this series is good to commit, but just to be clear: 1) Need Dan's virRun() and virReadFile() patches in first 2) Patch 5/9 wasn't ACK-ed 3) Also need ACKs for the "use virRun()" and "add --enable-iptables-lokkit=check" patches Cheers, Mark.

3 6

[Libvir] [PATCH] an irregular value is set as 'CPU Affinity' by virsh vcpupin
by Saori Fukuta 09 Jan '08

09 Jan '08

Hi, There is a difference between 'CPU Affinity' of 'virsh vcpuinfo' and 'CPU Affinity' of 'xm vcpu-list' when I set the vcpu affinity by 'virsh vcpupin'. i.e. 'xm vcpu-list' shows wrong value. But there is no difference between them when I set by 'xm vcpu-pin'. (a) set by 'virsh vcpupin' # virsh vcpupin guest_dom 0 1 # virsh vcpuinfo guest_dom VCPU: 0 CPU: 1 State: blocked CPU time: 17.1s CPU Affinity: -y------ # xm vcpu-list guest_dom Name ID VCPUs CPU State Time(s) CPU Affinity guest_dom 13 0 1 -b- 17.1 any cpu (b) set by 'xm vcpu-pin' # xm vcpu-pin guest_dom 0 0 # virsh vcpuinfo guest_dom VCPU: 0 CPU: 0 State: blocked CPU time: 17.1s CPU Affinity: y------- # xm vcpu-list guest_dom Name ID VCPUs CPU State Time(s) CPU Affinity guest_dom 13 0 0 -b- 17.1 0 I guess that the allocated memory to cpumap must be 8-byte for Xen because "xc_vcpu_setaffinity(a)libxc/xc_domain.c" is using uint64_t as cpumap. But the libvirt is allocating only maplen size for cpumap. virsh +--- cpumaplen ---+ (more than 1-byte) xen +----------------- 8-byte -----------------+ (uint64_t) Then, an irregular value is set to the buffer that is more than the number of physical CPU, because libvirt initialize maplen size only (i.e. not initialize 8-byte buffer). So, I changed the size of cpumap and the value of nr_cpus at xen_internal.c. Regards, Saori Fukuta

2 2

Re: [Libvir] [patch 6/9] Move iptablesSpawn()
by Daniel P. Berrange 09 Jan '08

09 Jan '08

On Fri, Jan 04, 2008 at 03:57:31PM +0000, Mark McLoughlin wrote: > The next patch requires iptablesSpawn() higher up in > the file. This patch just moves the code around; there > is no functional change. The util.c file has a virExec function, although its a little more cumbersome to use than iptablesSpawn. So for the storage APIs I've got a new virRun() function in util.c which is very similar to your iptablesSpawn code. How about we just make iptables use the virRun method in the attached patch ? diff -r a0ae4f87480b src/util.c --- a/src/util.c Thu Dec 06 15:35:43 2007 -0500 +++ b/src/util.c Thu Dec 06 15:39:34 2007 -0500 @@ -34,6 +34,7 @@ #include <errno.h> #include <sys/types.h> #include <sys/stat.h> +#include <sys/wait.h> #include <string.h> #include "libvirt/virterror.h" @@ -196,6 +197,43 @@ virExecNonBlock(virConnectPtr conn, int *retpid, int infd, int *outfd, int *errfd) { return(_virExec(conn, argv, retpid, infd, outfd, errfd, 1)); +} + +/** + * @conn connection to report errors against + * @argv NULL terminated argv to run + * @status optional variable to return exit status in + * + * Run a command without using the shell. + * + * If status is NULL, then return 0 if the command run and + * exited with 0 status; Otherwise return -1 + * + * If status is not-NULL, then return 0 if the command ran. + * The status variable is filled with the command exit status + * and should be checked by caller for success. Return -1 + * only if the command could not be run. + */ +int +virRun(virConnectPtr conn, + char **argv, + int *status) { + int childpid, exitstatus, ret; + + if ((ret = virExec(conn, argv, &childpid, -1, NULL, NULL)) < 0) + return ret; + + while ((ret = waitpid(childpid, &exitstatus, 0) == -1) && errno == EINTR); + if (ret == -1) + return -1; + + if (status == NULL) { + errno = EINVAL; + return (WIFEXITED(exitstatus) && WEXITSTATUS(exitstatus) == 0) ? 0 : -1; + } else { + *status = exitstatus; + return 0; + } } /* Like read(), but restarts after EINTR */ diff -r a0ae4f87480b src/util.h --- a/src/util.h Thu Dec 06 15:35:43 2007 -0500 +++ b/src/util.h Thu Dec 06 15:39:34 2007 -0500 @@ -28,6 +28,7 @@ int virExec(virConnectPtr conn, char **argv, int *retpid, int infd, int *outfd, int *errfd); int virExecNonBlock(virConnectPtr conn, char **argv, int *retpid, int infd, int *outfd, int *errfd); +int virRun(virConnectPtr conn, char **argv, int *status); int saferead(int fd, void *buf, size_t count); ssize_t safewrite(int fd, const void *buf, size_t count); Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|

3 3

Re: [Libvir] [patch 1/2] Do not try and delete built-in iptables chains
by Daniel P. Berrange 08 Jan '08

08 Jan '08

On Mon, Jan 07, 2008 at 06:12:37PM +0000, Mark McLoughlin wrote: > Previously, when we supported adding our rules to chains > with a custom prefix (e.g. libvirt-Fedora-POSTROUTING) > we needed to be able to handle adding and deleting these > custom chains. > > Now that we only use built-in iptables chains, we don't > need to add or delete them - in fact, deleting them is > not allowed. ACK 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 -=|

1 0

Re: [Libvir] [patch 2/2] Enable stdout/stderr spew with --enable-debug
by Daniel P. Berrange 08 Jan '08

08 Jan '08

On Mon, Jan 07, 2008 at 06:12:38PM +0000, Mark McLoughlin wrote: > Change virExec() such that with --enable-debug, stdout > and stderr no longer go to /dev/null. ACK 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 -=|

1 0

[Libvir] PATCH: BZ 426425: Fix truncated reading of config files
by Daniel P. Berrange 08 Jan '08

08 Jan '08

The virConfReadFile API has a fixed 4096 byte buffer it reads the config file into, regardless of the config file size and silently drops any data exceeding this. This manifested itself in the 0.4.0 release by the fact that toggling the 'auth_unix_ro' config param from 'polkit' to 'none' had no effect at all. It turns out that with the user friendly verbose comments I added to the config file /etc/libvirt/libvirtd.cofn, the 'auth_unix_ro' param occurred beyond the 4096 byte mark in the file and was thus never read. In fixing this I decided to make virConfReadFile call to the recently added virFileReadAll() method. Except the latter takes a pre-allocated buffer as its input. So I've switched virFileReadAll() to pass in a char **, and auto-allocate the file as needed. As a sanity check there is also a 'maxlen' param to stop it reading stupidly huge files - for reading large files virFileReadll() shouldn't be used - they should be read incrementally. Finally, also added a larger test data file to trap this error case src/conf.c | 27 ++--- src/qemu_conf.c | 6 - src/util.c | 22 ++-- src/util.h | 4 tests/confdata/libvirtd.conf | 222 +++++++++++++++++++++++++++++++++++++++++++ tests/confdata/libvirtd.out | 180 ++++++++++++++++++++++++++++++++++ tests/test_conf.sh | 3 7 files changed, 441 insertions(+), 23 deletions(-) Index: src/conf.c =================================================================== RCS file: /data/cvs/libvirt/src/conf.c,v retrieving revision 1.15 diff -u -p -u -p -r1.15 conf.c --- src/conf.c 11 Dec 2007 21:57:29 -0000 1.15 +++ src/conf.c 3 Jan 2008 18:58:47 -0000 @@ -21,6 +21,7 @@ #include "internal.h" #include "buf.h" #include "conf.h" +#include "util.h" /************************************************************************ * * @@ -693,6 +694,9 @@ error: * * ************************************************************************/ +/* 10 MB limit on config file size as a sanity check */ +#define MAX_CONFIG_FILE_SIZE (1024*1024*10) + /** * __virConfReadFile: * @filename: the path to the configuration file. @@ -705,26 +709,25 @@ error: virConfPtr __virConfReadFile(const char *filename) { - char content[4096]; - int fd; + char *content; int len; + virConfPtr conf; if (filename == NULL) { virConfError(NULL, VIR_ERR_INVALID_ARG, __FUNCTION__, 0); return(NULL); } - fd = open(filename, O_RDONLY); - if (fd < 0) { + + if ((len = virFileReadAll(filename, MAX_CONFIG_FILE_SIZE, &content)) < 0) { virConfError(NULL, VIR_ERR_OPEN_FAILED, filename, 0); - return(NULL); + return NULL; } - len = read(fd, content, sizeof(content)); - close(fd); - if (len <= 0) { - virConfError(NULL, VIR_ERR_READ_FAILED, filename, 0); - return(NULL); - } - return(virConfParse(filename, content, len)); + + conf = virConfParse(filename, content, len); + + free(content); + + return conf; } /** Index: src/qemu_conf.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_conf.c,v retrieving revision 1.24 diff -u -p -u -p -r1.24 qemu_conf.c --- src/qemu_conf.c 11 Dec 2007 21:57:29 -0000 1.24 +++ src/qemu_conf.c 3 Jan 2008 18:58:48 -0000 @@ -2604,7 +2604,7 @@ int qemudScanConfigDir(struct qemud_driv } while ((entry = readdir(dir))) { - char xml[QEMUD_MAX_XML_LEN]; + char *xml; char path[PATH_MAX]; char autostartLink[PATH_MAX]; @@ -2626,13 +2626,15 @@ int qemudScanConfigDir(struct qemud_driv continue; } - if (virFileReadAll(path, xml, QEMUD_MAX_XML_LEN) < 0) + if (virFileReadAll(path, QEMUD_MAX_XML_LEN, &xml) < 0) continue; if (isGuest) qemudLoadConfig(driver, entry->d_name, path, xml, autostartLink); else qemudLoadNetworkConfig(driver, entry->d_name, path, xml, autostartLink); + + free(xml); } closedir(dir); Index: src/util.c =================================================================== RCS file: /data/cvs/libvirt/src/util.c,v retrieving revision 1.10 diff -u -p -u -p -r1.10 util.c --- src/util.c 7 Dec 2007 14:45:39 -0000 1.10 +++ src/util.c 3 Jan 2008 18:58:48 -0000 @@ -272,8 +272,8 @@ ssize_t safewrite(int fd, const void *bu int virFileReadAll(const char *path, - char *buf, - unsigned int buflen) + int maxlen, + char **buf) { FILE *fh; struct stat st; @@ -296,20 +296,28 @@ int virFileReadAll(const char *path, goto error; } - if (st.st_size >= (buflen-1)) { - virLog("File '%s' is too large", path); + if (st.st_size > maxlen) { + virLog("File '%s' is too large %d, max %d", path, st.st_size, maxlen); goto error; } - if ((ret = fread(buf, st.st_size, 1, fh)) != 1) { + *buf = malloc(st.st_size + 1); + if (*buf == NULL) { + virLog("Failed to allocate data"); + goto error; + } + + if ((ret = fread(*buf, st.st_size, 1, fh)) != 1) { + free(buf); + buf = NULL; virLog("Failed to read config file '%s': %s", path, strerror(errno)); goto error; } - buf[st.st_size] = '\0'; + (*buf)[st.st_size] = '\0'; - ret = 0; + ret = st.st_size; error: if (fh) Index: src/util.h =================================================================== RCS file: /data/cvs/libvirt/src/util.h,v retrieving revision 1.4 diff -u -p -u -p -r1.4 util.h --- src/util.h 3 Dec 2007 14:30:47 -0000 1.4 +++ src/util.h 3 Jan 2008 18:58:48 -0000 @@ -33,8 +33,8 @@ int saferead(int fd, void *buf, size_t c ssize_t safewrite(int fd, const void *buf, size_t count); int virFileReadAll(const char *path, - char *buf, - unsigned int buflen); + int maxlen, + char **buf); int virFileMatchesNameSuffix(const char *file, const char *name, Index: tests/test_conf.sh =================================================================== RCS file: /data/cvs/libvirt/tests/test_conf.sh,v retrieving revision 1.2 diff -u -p -u -p -r1.2 test_conf.sh --- tests/test_conf.sh 14 Nov 2007 10:35:59 -0000 1.2 +++ tests/test_conf.sh 3 Jan 2008 18:58:48 -0000 @@ -8,6 +8,9 @@ do diff $outfile conftest.$$ > /dev/null if [ $? != 0 ] then + if [ -n "$DEBUG_TESTS" ]; then + diff -u $outfile conftest.$$ + fi echo "$f FAILED" NOK=1 else Index: tests/confdata/libvirtd.conf =================================================================== RCS file: tests/confdata/libvirtd.conf diff -N tests/confdata/libvirtd.conf --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/confdata/libvirtd.conf 3 Jan 2008 18:58:48 -0000 @@ -0,0 +1,222 @@ +# Master libvirt daemon configuration file +# +# For further information consult http://libvirt.org/format.html + + +################################################################# +# +# Network connectivitiy controls +# + +# Flag listening for secure TLS connections on the public TCP/IP port. +# NB, must pass the --listen flag to the libvirtd process for this to +# have any effect. +# +# It is neccessary to setup a CA and issue server certificates before +# using this capability. +# +# This is enabled by default, uncomment this to disable it +listen_tls = 0 + +# Listen for unencrypted TCP connections on the public TCP/IP port. +# NB, must pass the --listen flag to the libvirtd process for this to +# have any effect. +# +# Using the TCP socket requires SASL authentication by default. Only +# SASL mechanisms which support data encryption are allowed. This is +# DIGEST_MD5 and GSSAPI (Kerberos5) +# +# This is disabled by default, uncomment this to enable it. +listen_tcp = 1 + + + +# Override the port for accepting secure TLS connections +# This can be a port number, or service name +# +tls_port = "16514" + +# Override the port for accepting insecure TCP connections +# This can be a port number, or service name +# +tcp_port = "16509" + + + +# Flag toggling mDNS advertizement of the libvirt service. +# +# Alternatively can disable for all services on a host by +# stopping the Avahi daemon +# +# This is enabled by default, uncomment this to disable it +mdns_adv = 0 + +# Override the default mDNS advertizement name. This must be +# unique on the immediate broadcast network. +# +# The default is "Virtualization Host HOSTNAME", where HOSTNAME +# is subsituted for the short hostname of the machine (without domain) +# +mdns_name = "Virtualization Host Joe Demo" + + +################################################################# +# +# UNIX socket access controls +# + +# Set the UNIX domain socket group ownership. This can be used to +# allow a 'trusted' set of users access to management capabilities +# without becoming root. +# +# This is restricted to 'root' by default. +unix_sock_group = "libvirt" + +# Set the UNIX socket permissions for the R/O socket. This is used +# for monitoring VM status only +# +# Default allows any user. If setting group ownership may want to +# restrict this to: +unix_sock_ro_perms = "0777" + +# Set the UNIX socket permissions for the R/W socket. This is used +# for full management of VMs +# +# Default allows only root. If PolicyKit is enabled on the socket, +# the default will change to allow everyone (eg, 0777) +# +# If not using PolicyKit and setting group ownership for access +# control then you may want to relax this to: +unix_sock_rw_perms = "0770" + + + +################################################################# +# +# Authentication. +# +# - none: do not perform auth checks. If you can connect to the +# socket you are allowed. This is suitable if there are +# restrictions on connecting to the socket (eg, UNIX +# socket permissions), or if there is a lower layer in +# the network providing auth (eg, TLS/x509 certificates) +# +# - sasl: use SASL infrastructure. The actual auth scheme is then +# controlled from /etc/sasl2/libvirt.conf. For the TCP +# socket only GSSAPI & DIGEST-MD5 mechanisms will be used. +# For non-TCP or TLS sockets, any scheme is allowed. +# +# - polkit: use PolicyKit to authenticate. This is only suitable +# for use on the UNIX sockets. The default policy will +# require a user to supply their own password to gain +# full read/write access (aka sudo like), while anyone +# is allowed read/only access. +# +# Set an authentication scheme for UNIX read-only sockets +# By default socket permissions allow anyone to connect +# +# To restrict monitoring of domains you may wish to enable +# an authentication mechanism here +auth_unix_ro = "none" + +# Set an authentication scheme for UNIX read-write sockets +# By default socket permissions only allow root. If PolicyKit +# support was compiled into libvirt, the default will be to +# use 'polkit' auth. +# +# If the unix_sock_rw_perms are changed you may wish to enable +# an authentication mechanism here +auth_unix_rw = "none" + +# Change the authentication scheme for TCP sockets. +# +# If you don't enable SASL, then all TCP traffic is cleartext. +# Don't do this outside of a dev/test scenario. For real world +# use, always enable SASL and use the GSSAPI or DIGEST-MD5 +# mechanism in /etc/sasl2/libvirt.conf +auth_tcp = "sasl" + +# Change the authentication scheme for TLS sockets. +# +# TLS sockets already have encryption provided by the TLS +# layer, and limited authentication is done by certificates +# +# It is possible to make use of any SASL authentication +# mechanism as well, by using 'sasl' for this option +auth_tls = "none" + + + +################################################################# +# +# TLS x509 certificate configuration +# + + +# Override the default server key file path +# +key_file = "/etc/pki/libvirt/private/serverkey.pem" + +# Override the default server certificate file path +# +cert_file = "/etc/pki/libvirt/servercert.pem" + +# Override the default CA certificate path +# +ca_file = "/etc/pki/CA/cacert.pem" + +# Specify a certificate revocation list. +# +# Defaults to not using a CRL, uncomment to enable it +crl_file = "/etc/pki/CA/crl.pem" + + + +################################################################# +# +# Authorization controls +# + + +# Flag to disable verification of client certificates +# +# Client certificate verification is the primary authentication mechanism. +# Any client which does not present a certificate signed by the CA +# will be rejected. +# +# Default is to always verify. Uncommenting this will disable +# verification - make sure an IP whitelist is set +tls_no_verify_certificate = 1 + + +# A whitelist of allowed x509 Distinguished Names +# This list may contain wildcards such as +# +# "C=GB,ST=London,L=London,O=Red Hat,CN=*" +# +# See the POSIX fnmatch function for the format of the wildcards. +# +# NB If this is an empty list, no client can connect, so comment out +# entirely rather than using empty list to disable these checks +# +# By default, no DN's are checked +tls_allowed_dn_list = ["DN1", "DN2"] + + +# A whitelist of allowed SASL usernames. The format for usernames +# depends on the SASL authentication mechanism. Kerberos usernames +# look like username@REALM +# +# This list may contain wildcards such as +# +# "*(a)EXAMPLE.COM" +# +# See the POSIX fnmatch function for the format of the wildcards. +# +# NB If this is an empty list, no client can connect, so comment out +# entirely rather than using empty list to disable these checks +# +# By default, no Username's are checked +sasl_allowed_username_list = ["joe(a)EXAMPLE.COM", "fred(a)EXAMPLE.COM" ] + + Index: tests/confdata/libvirtd.out =================================================================== RCS file: tests/confdata/libvirtd.out diff -N tests/confdata/libvirtd.out --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/confdata/libvirtd.out 3 Jan 2008 18:58:48 -0000 @@ -0,0 +1,180 @@ +# Master libvirt daemon configuration file +# +# For further information consult http://libvirt.org/format.html +################################################################# +# +# Network connectivitiy controls +# +# Flag listening for secure TLS connections on the public TCP/IP port. +# NB, must pass the --listen flag to the libvirtd process for this to +# have any effect. +# +# It is neccessary to setup a CA and issue server certificates before +# using this capability. +# +# This is enabled by default, uncomment this to disable it +listen_tls = 0 +# Listen for unencrypted TCP connections on the public TCP/IP port. +# NB, must pass the --listen flag to the libvirtd process for this to +# have any effect. +# +# Using the TCP socket requires SASL authentication by default. Only +# SASL mechanisms which support data encryption are allowed. This is +# DIGEST_MD5 and GSSAPI (Kerberos5) +# +# This is disabled by default, uncomment this to enable it. +listen_tcp = 1 +# Override the port for accepting secure TLS connections +# This can be a port number, or service name +# +tls_port = "16514" +# Override the port for accepting insecure TCP connections +# This can be a port number, or service name +# +tcp_port = "16509" +# Flag toggling mDNS advertizement of the libvirt service. +# +# Alternatively can disable for all services on a host by +# stopping the Avahi daemon +# +# This is enabled by default, uncomment this to disable it +mdns_adv = 0 +# Override the default mDNS advertizement name. This must be +# unique on the immediate broadcast network. +# +# The default is "Virtualization Host HOSTNAME", where HOSTNAME +# is subsituted for the short hostname of the machine (without domain) +# +mdns_name = "Virtualization Host Joe Demo" +################################################################# +# +# UNIX socket access controls +# +# Set the UNIX domain socket group ownership. This can be used to +# allow a 'trusted' set of users access to management capabilities +# without becoming root. +# +# This is restricted to 'root' by default. +unix_sock_group = "libvirt" +# Set the UNIX socket permissions for the R/O socket. This is used +# for monitoring VM status only +# +# Default allows any user. If setting group ownership may want to +# restrict this to: +unix_sock_ro_perms = "0777" +# Set the UNIX socket permissions for the R/W socket. This is used +# for full management of VMs +# +# Default allows only root. If PolicyKit is enabled on the socket, +# the default will change to allow everyone (eg, 0777) +# +# If not using PolicyKit and setting group ownership for access +# control then you may want to relax this to: +unix_sock_rw_perms = "0770" +################################################################# +# +# Authentication. +# +# - none: do not perform auth checks. If you can connect to the +# socket you are allowed. This is suitable if there are +# restrictions on connecting to the socket (eg, UNIX +# socket permissions), or if there is a lower layer in +# the network providing auth (eg, TLS/x509 certificates) +# +# - sasl: use SASL infrastructure. The actual auth scheme is then +# controlled from /etc/sasl2/libvirt.conf. For the TCP +# socket only GSSAPI & DIGEST-MD5 mechanisms will be used. +# For non-TCP or TLS sockets, any scheme is allowed. +# +# - polkit: use PolicyKit to authenticate. This is only suitable +# for use on the UNIX sockets. The default policy will +# require a user to supply their own password to gain +# full read/write access (aka sudo like), while anyone +# is allowed read/only access. +# +# Set an authentication scheme for UNIX read-only sockets +# By default socket permissions allow anyone to connect +# +# To restrict monitoring of domains you may wish to enable +# an authentication mechanism here +auth_unix_ro = "none" +# Set an authentication scheme for UNIX read-write sockets +# By default socket permissions only allow root. If PolicyKit +# support was compiled into libvirt, the default will be to +# use 'polkit' auth. +# +# If the unix_sock_rw_perms are changed you may wish to enable +# an authentication mechanism here +auth_unix_rw = "none" +# Change the authentication scheme for TCP sockets. +# +# If you don't enable SASL, then all TCP traffic is cleartext. +# Don't do this outside of a dev/test scenario. For real world +# use, always enable SASL and use the GSSAPI or DIGEST-MD5 +# mechanism in /etc/sasl2/libvirt.conf +auth_tcp = "sasl" +# Change the authentication scheme for TLS sockets. +# +# TLS sockets already have encryption provided by the TLS +# layer, and limited authentication is done by certificates +# +# It is possible to make use of any SASL authentication +# mechanism as well, by using 'sasl' for this option +auth_tls = "none" +################################################################# +# +# TLS x509 certificate configuration +# +# Override the default server key file path +# +key_file = "/etc/pki/libvirt/private/serverkey.pem" +# Override the default server certificate file path +# +cert_file = "/etc/pki/libvirt/servercert.pem" +# Override the default CA certificate path +# +ca_file = "/etc/pki/CA/cacert.pem" +# Specify a certificate revocation list. +# +# Defaults to not using a CRL, uncomment to enable it +crl_file = "/etc/pki/CA/crl.pem" +################################################################# +# +# Authorization controls +# +# Flag to disable verification of client certificates +# +# Client certificate verification is the primary authentication mechanism. +# Any client which does not present a certificate signed by the CA +# will be rejected. +# +# Default is to always verify. Uncommenting this will disable +# verification - make sure an IP whitelist is set +tls_no_verify_certificate = 1 +# A whitelist of allowed x509 Distinguished Names +# This list may contain wildcards such as +# +# "C=GB,ST=London,L=London,O=Red Hat,CN=*" +# +# See the POSIX fnmatch function for the format of the wildcards. +# +# NB If this is an empty list, no client can connect, so comment out +# entirely rather than using empty list to disable these checks +# +# By default, no DN's are checked +tls_allowed_dn_list = [ "DN1", "DN2" ] +# A whitelist of allowed SASL usernames. The format for usernames +# depends on the SASL authentication mechanism. Kerberos usernames +# look like username@REALM +# +# This list may contain wildcards such as +# +# "*(a)EXAMPLE.COM" +# +# See the POSIX fnmatch function for the format of the wildcards. +# +# NB If this is an empty list, no client can connect, so comment out +# entirely rather than using empty list to disable these checks +# +# By default, no Username's are checked +sasl_allowed_username_list = [ "joe(a)EXAMPLE.COM", "fred(a)EXAMPLE.COM" ] 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 5

[Libvir] [PATCH] Xen: Support cpu_weight and cpu_cap for Xen.
by Tatsuro Enokura 08 Jan '08

08 Jan '08

Hi, I set the information of cpu_weight/cpu_cap to configuration file or sxp format, and execute "virsh start". But the information of weight/cap is lost. libvirt doesn't support cpu_weight and cpu_cap for XML format. see also Bz#337591 https://bugzilla.redhat.com/show_bug.cgi?id=337591 I make a patch to add weight/cap attributes as vcpus element (eg. <vcpus weight='512' cap='280'>4</vcpus>) c.f. the thread at: https://www.redhat.com/archives/libvir-list/2007-October/msg00044.html Signed-off-by: Tatsuro Enokura <fj7716hz(a)aa.jp.fujitsu.com> Thanks, Tatsuro Enokura Index: libvirt/src/xend_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/xend_internal.c,v retrieving revision 1.151 diff -u -p -r1.151 xend_internal.c --- libvirt/src/xend_internal.c 22 Oct 2007 20:28:55 -0000 1.151 +++ libvirt/src/xend_internal.c 24 Oct 2007 06:02:27 -0000 @@ -1438,8 +1438,19 @@ xend_parse_sexp_desc(virConnectPtr conn, if ((cur_mem >= MIN_XEN_GUEST_SIZE) && (cur_mem != max_mem)) virBufferVSprintf(&buf, " <currentMemory>%d</currentMemory>\n", cur_mem); - virBufferVSprintf(&buf, " <vcpu>%d</vcpu>\n", + virBufferVSprintf(&buf, " <vcpu"); + if (sexpr_node(root, "domain/cpu_weight") != NULL) { + virBufferVSprintf(&buf, " weight='%d'", + sexpr_int(root, "domain/cpu_weight")); + } + if (sexpr_node(root, "domain/cpu_cap") != NULL) { + virBufferVSprintf(&buf, " cap='%d'", + sexpr_int(root, "domain/cpu_cap")); + } + virBufferVSprintf(&buf, ">%d</vcpu>\n", sexpr_int(root, "domain/vcpus")); + + tmp = sexpr_node(root, "domain/on_poweroff"); if (tmp != NULL) virBufferVSprintf(&buf, " <on_poweroff>%s</on_poweroff>\n", tmp); Index: libvirt/src/xm_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/xm_internal.c,v retrieving revision 1.41 diff -u -p -r1.41 xm_internal.c --- libvirt/src/xm_internal.c 10 Oct 2007 17:55:38 -0000 1.41 +++ libvirt/src/xm_internal.c 24 Oct 2007 06:02:28 -0000 @@ -661,9 +661,14 @@ char *xenXMDomainFormatXML(virConnectPtr virBufferVSprintf(buf, " <memory>%ld</memory>\n", val * 1024); + virBufferVSprintf(buf, " <vcpu"); + if (xenXMConfigGetInt(conf, "cpu_weight", &val) == 0) + virBufferVSprintf(buf, " weight='%ld'", val); + if (xenXMConfigGetInt(conf, "cpu_cap", &val) == 0) + virBufferVSprintf(buf, " cap='%ld'", val); if (xenXMConfigGetInt(conf, "vcpus", &val) < 0) val = 1; - virBufferVSprintf(buf, " <vcpu>%ld</vcpu>\n", val); + virBufferVSprintf(buf, ">%ld</vcpu>\n", val); if (xenXMConfigGetString(conf, "on_poweroff", &str) < 0) @@ -1820,6 +1825,12 @@ virConfPtr xenXMParseXMLToConfig(virConn if (xenXMConfigSetIntFromXPath(conn, conf, ctxt, "vcpus", "string(/domain/vcpu)", 0, 1, "cannot set vcpus config parameter") < 0) goto error; + if (xenXMConfigSetIntFromXPath(conn, conf, ctxt, "cpu_weight", "string(/domain/vcpu/@weight)", 0, 1, + "cannot set cpu_weight config paramerter") < 0) + goto error; + if (xenXMConfigSetIntFromXPath(conn, conf, ctxt, "cpu_cap", "string(/domain/vcpu/@cap)", 0, 1, + "cannot set cpu_cap config paramerter") < 0) + goto error; obj = xmlXPathEval(BAD_CAST "string(/domain/os/type)", ctxt); if ((obj != NULL) && (obj->type == XPATH_STRING) && Index: libvirt/src/xml.c =================================================================== RCS file: /data/cvs/libvirt/src/xml.c,v retrieving revision 1.94 diff -u -p -r1.94 xml.c --- libvirt/src/xml.c 23 Oct 2007 15:31:33 -0000 1.94 +++ libvirt/src/xml.c 24 Oct 2007 06:02:28 -0000 @@ -1529,6 +1529,12 @@ virDomainParseXMLDesc(virConnectPtr conn vcpus = (unsigned int) f; } virBufferVSprintf(&buf, "(vcpus %u)", vcpus); + if (virXPathNumber("number(/domain/vcpu[1]/@weight)", ctxt, &f) == 0) { + virBufferVSprintf(&buf, "(cpu_weight %ld)", (long) f); + } + if (virXPathNumber("number(/domain/vcpu[1]/@cap)", ctxt, &f) == 0) { + virBufferVSprintf(&buf, "(cpu_cap %ld)", (long) f); + } str = virXPathString("string(/domain/vcpu/@cpuset)", ctxt); if (str != NULL) {

6 11

[Libvir] OCaml bindings 0.4.0.1 released
by Richard W.M. Jones 08 Jan '08

08 Jan '08

I'm pleased to announce the latest release of the libvirt OCaml bindings and programs. http://libvirt.org/ocaml/ There's a large list of changes since the last release, so best thing is to refer to the Changelog itself: http://libvirt.org/ocaml/ChangeLog.txt Source code is available from here: http://libvirt.org/sources/ocaml/ 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] OCaml bindings - Windows installer
by Richard W.M. Jones 08 Jan '08

08 Jan '08

This is a Windows installer for the OCaml libvirt bindings and programs: http://libvirt.org/sources/ocaml/ocaml-libvirt-0.4.0.1.exe If someone has a 'virgin' Windows system and could test that the above installer works, particularly for Windows < Vista and for Windows which has not had any GTK/MinGW development tools installed. About 90% of the size of the installer is down to the GTK DLLs and configuration files which need to be carried along to support the GTK graphical program (Virt Control). It's possible that I haven't placed the GTK files in exactly the right place, so instead it is using my own GTK development environment. Screenshots showing the installer in action: http://annexia.org/tmp/wininstaller-1-desktop.png Desktop, nothing installed, showing the installer icon. http://annexia.org/tmp/wininstaller-2-packages.png List of required, recommended and optional subpackages. http://annexia.org/tmp/wininstaller-3-instpath.png Select the installation path. http://annexia.org/tmp/wininstaller-4-installdone.png Install complete. Notice the desktop icons, ... http://annexia.org/tmp/wininstaller-5-menu.png ... and the menu entries and uninstaller program. http://annexia.org/tmp/wininstaller-6-virtctrl.png Running Virt Control. http://annexia.org/tmp/wininstaller-7-uninstall.png After uninstall. Notice that the desktop icons have gone. The installer uses NSIS (the Nullsoft Scriptable Install System, http://nsis.sf.net/) and is reasonably well integrated into the autoconf / make build system. In particular if you are compiling under Windows and NSIS is installed, then a simple './configure --with-nsis; make wininstaller' will build a Windows installer. This is something to look at if libvirt wants/needs a Windows installer. 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

Re: [Libvir] [patch 2/2] Enable stdout/stderr spew with --enable-debug
by Daniel Veillard 08 Jan '08

08 Jan '08

On Mon, Jan 07, 2008 at 06:12:38PM +0000, Mark McLoughlin wrote: > Change virExec() such that with --enable-debug, stdout > and stderr no longer go to /dev/null. > Looks fine to me, but I didn't see [1/2] numbering problem ? I didn't see any bounced mail in the list either, 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 1

[Libvir] [patch 2/2] Enable stdout/stderr spew with --enable-debug
by Mark McLoughlin 07 Jan '08

07 Jan '08

Change virExec() such that with --enable-debug, stdout and stderr no longer go to /dev/null. Signed-off-by: Mark McLoughlin <markmc(a)redhat.com> Index: libvirt/src/util.c =================================================================== --- libvirt.orig/src/util.c 2008-01-07 18:00:03.000000000 +0000 +++ libvirt.orig/src/util.c 2008-01-07 18:00:03.000000000 +0000 @@ -157,10 +157,17 @@ _virExec(virConnectPtr conn, if (dup2(infd >= 0 ? infd : null, STDIN_FILENO) < 0) _exit(1); +#ifndef ENABLE_DEBUG if (dup2(pipeout[1] > 0 ? pipeout[1] : null, STDOUT_FILENO) < 0) _exit(1); if (dup2(pipeerr[1] > 0 ? pipeerr[1] : null, STDERR_FILENO) < 0) _exit(1); +#else /* ENABLE_DEBUG */ + if (pipeout[1] > 0 && dup2(pipeout[1], STDOUT_FILENO) < 0) + _exit(1); + if (pipeerr[1] > 0 && dup2(pipeerr[1], STDERR_FILENO) < 0) + _exit(1); +#endif /* ENABLE_DEBUG */ close(null); if (pipeout[1] > 0) --

1 0

[Libvir] [patch 1/2] Do not try and delete built-in iptables chains
by Mark McLoughlin 07 Jan '08

07 Jan '08

Previously, when we supported adding our rules to chains with a custom prefix (e.g. libvirt-Fedora-POSTROUTING) we needed to be able to handle adding and deleting these custom chains. Now that we only use built-in iptables chains, we don't need to add or delete them - in fact, deleting them is not allowed. (Note: this was the only user of the iptablesSpawn(NO_ERRORS) stuff - we didn't want error spew when adding/deleting the chains failed as expected in most cases) Signed-off-by: Mark McLoughlin <markmc(a)redhat.com> Index: libvirt/src/iptables.c =================================================================== --- libvirt.orig/src/iptables.c 2008-01-07 17:26:42.000000000 +0000 +++ libvirt.orig/src/iptables.c 2008-01-07 17:26:42.000000000 +0000 @@ -386,53 +386,6 @@ iptRulesNew(const char *table, return NULL; } -static int -iptablesAddRemoveChain(iptRules *rules, int action) -{ - char **argv; - int retval = ENOMEM; - int n, status; - - n = 1 + /* /sbin/iptables */ - 2 + /* --table foo */ - 2; /* --new-chain bar */ - - if (!(argv = calloc(n + 1, sizeof(*argv)))) - goto error; - - n = 0; - - if (!(argv[n++] = strdup(IPTABLES_PATH))) - goto error; - - if (!(argv[n++] = strdup("--table"))) - goto error; - - if (!(argv[n++] = strdup(rules->table))) - goto error; - - if (!(argv[n++] = strdup(action == ADD ? "--new-chain" : "--delete-chain"))) - goto error; - - if (!(argv[n++] = strdup(rules->chain))) - goto error; - - if (virRun(NULL, argv, &status) < 0) - retval = errno; - - retval = 0; - - error: - if (argv) { - n = 0; - while (argv[n]) - free(argv[n++]); - free(argv); - } - - return retval; -} - static char * argvToString(char **argv) { @@ -521,19 +474,11 @@ iptablesAddRemoveRule(iptRules *rules, i goto error; } - if (action == ADD && - (retval = iptablesAddRemoveChain(rules, action))) - goto error; - if (virRun(NULL, argv, NULL) < 0) { retval = errno; goto error; } - if (action == REMOVE && - (retval = iptablesAddRemoveChain(rules, action))) - goto error; - if (action == ADD) { retval = iptRulesAppend(rules, rule, argv, command_idx); rule = NULL; @@ -641,11 +586,6 @@ iptRulesReload(iptRules *rules) rule->argv[rule->command_idx] = orig; } - if ((retval = iptablesAddRemoveChain(rules, REMOVE)) || - (retval = iptablesAddRemoveChain(rules, ADD))) - qemudLog(QEMUD_WARN, "Failed to re-create chain '%s' in table '%s': %s", - rules->chain, rules->table, strerror(retval)); - for (i = 0; i < rules->nrules; i++) if (virRun(NULL, rules->rules[i].argv, NULL) < 0) qemudLog(QEMUD_WARN, "Failed to add iptables rule '%s' to chain '%s' in table '%s': %s", --

1 0

[Libvir] [patch 0/2] Couple more misc iptables related patches
by Mark McLoughlin 07 Jan '08

07 Jan '08

Hey, Dan's suggestion to enable debug spew from spawned processes with --enable-debug reminded me that the only reason I wanted the NO_ERRORS stuff in the iptables code was to avoid spewage when trying to add/delete builtin iptables chains. Cheers, Mark. --

1 0

Re: [Libvir] [patch 7/9] Add support for lokkit
by Daniel P. Berrange 07 Jan '08

07 Jan '08

On Fri, Jan 04, 2008 at 03:57:32PM +0000, Mark McLoughlin wrote: > Add support for integrating our iptables support with Fedora's > iptables configuration using the lokkit --custom-rules command. > > Basically, we write out our rules to /var/lib/libvirt/iptables > and run lokkit --custom-rules so that if e.g. iptables is > restarted or the user edits their firewall configuration, then > libvirt's rules get reloaded. Ahh, that's very nice to have. > +dnl > +dnl ensure that Fedora's system-config-firewall knows > +dnl about libvirt's iptables rules > +dnl > +AC_ARG_ENABLE(iptables-lokkit, > + AC_HELP_STRING([--enable-iptables-lokkit=no/yes], > + [enable registering libvirt's iptables rules with Fedora's lokkit]), > + [],[enable_iptables_lokkit=no]) > +if test x"$enable_iptables_lokkit" = x"yes"; then > + AC_DEFINE(ENABLE_IPTABLES_LOKKIT, [], [whether support for Fedora's lokkit is enabled]) > + AC_PATH_PROG(LOKKIT_PATH, lokkit, /usr/sbin/lokkit) > + AC_DEFINE_UNQUOTED(LOKKIT_PATH, "$LOKKIT_PATH", [path to lokkit binary]) > +fi > + Could we make the configure script a little more clever so that it is a tri-state and can auto-detect whether lokkit is available. - enable_iptables_lokkit=no - force disable - enable_iptables_lokkit=yes - check if it is supported, and error if not - enable_iptables_lokkit=check - check if it is supported and enable or disable as needed With 'check' being the default. This makes it 'do the right' thing by default, and lets the user have a strict override if neccessary. ACK to the rest of the patch 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 2

Re: [Libvir] [patch 5/9] Include the iptables command and chain name in the saved rules file
by Daniel P. Berrange 07 Jan '08

07 Jan '08

On Fri, Jan 04, 2008 at 03:57:30PM +0000, Mark McLoughlin wrote: > lokkit --custom-rules expects the passed file to include the > iptables command and chain name (e.g. "--inset INPUT") rather > than just the rest of the arguments. > > Add both of those to what will be saved to the rules file > and simplify the resulting code by splitting out a > argvToString() helper function. > > The one complication is that when we're removing a rule > we need to make sure we don't search for it using > "--delete" rather than "--insert". For that reason, > only change the argument to "--delete" once we've > constructed the string we use to search through the > existing rules. ACK. 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 -=|

1 0

[Libvir] OCaml mlvirtmanager ported to Windows
by Richard W.M. Jones 05 Jan '08

05 Jan '08

(NB. This is _not_ the real virt-manager). Screenshots: http://www.annexia.org/tmp/mlvirtmanager-win-1.png http://www.annexia.org/tmp/mlvirtmanager-win-2.png http://www.annexia.org/tmp/mlvirtmanager-win-3.png The code is in the Mercurial repository, http://hg.et.redhat.com/virt/applications/virt-top--devel . You will need to install lablgtk2 binary from: http://wwwfun.kurims.kyoto-u.ac.jp/soft/olabl/lablgtk.html and Windows Gtk development binary from: http://gladewin32.sourceforge.net/ and then follow the installation instructions for lablgtk2 given here: http://wwwfun.kurims.kyoto-u.ac.jp/soft/olabl/install-win32.txt After that, ocaml-libvirt should automatically detect that you've got Gtk / lablgtk2 installed (you might need to restart the MSys shell so it gets the new $PATH), and will offer to compile mlvirtmanager. 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] A few more Windows / MinGW fixups
by Richard W.M. Jones 05 Jan '08

05 Jan '08

A few fairly miscellaneous fixups for Windows (MinGW) which fix shared library builds and fix error handling for sockets. (1) XDR functions on MinGW come from a library called 'libxdr', not 'librpc'. (2) To build a DLL under MinGW we need to pass the -no-undefined flag to the linker. (3) Socket compatibility header file replaces <winsock2.h> inclusion. This just defines a portable 'socket_errno()' function which returns errno in the normal case, or WSAGetLastError() in the Windows case. (4) Use socket_errno() instead of errno in a few cases (but only when the code can be compiled under Windows, ie. only in the remote client case). Example -- a dynamically linked virsh.exe (linked to libvirt-0.dll) accessing a remote libvirtd: $ src/.libs/virsh.exe -c test+tcp://192.168.2.128/default list Id Name State ---------------------------------- 1 test running If you want to compile this under Windows, you will need: (a) MinGW & MSYS (select the "candidate" versions of MinGW tools if you are using Vista or W2K8, since otherwise nothing works because of a known bug). (b) Install gcc 4 experimental package from MinGW site. It's called gcc-sjlj for reasons which escape me. (c) Install GnuTLS binary from http://www.gnu.org/software/gnutls/download.html. I'm using GnuTLS Windows binary 1.6.3 and I had to hand-hack the *.la files in that distribution because they contain incorrect paths. (d) Compile and install latest libxml2 from http://xmlsoft.org/. I'm using 2.6.30. (e) Compile and install my XDR package for Windows (http://www.annexia.org/tmp/xdr-4.0-mingw5.tar.gz) (f) ./configure --without-xen --without-qemu --without-sasl --without-libvirtd (g) make 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 4