April 2013 - CIM - libvirt List Archives

[PATCH v2 00/12] cimtest updates
by John Ferlan 15 May '13

15 May '13

This is primarily a repost of the previous series: https://www.redhat.com/archives/libvirt-cim/2013-March/msg00051.html and https://www.redhat.com/archives/libvirt-cim/2013-April/msg00014.html The primary difference in this patch vs. the previous sets is to fix the version string checking for nfs server checking in common_util.py and to add patch 12/12 which handles a problem in 'enum_volumes()' in pool.py. I assume patches 1-4 and 6-9 were reviewed without issue. So focus on 5/12 and 10-12/12. John Ferlan (12): Need to check "slp=true", not just "slp" since "slp=false" is possible Change the MAC from "99:" to "88:" Create a temporary directory for disk pool tests Use symbols as named in libvirt-cim for easier reference Fix nfs-server lookup code Fix os_status passing to reporter functions Resolve issues found in test. On Fedora systems default to using 'em1' instead of 'eth1' 19 - resolve issues found in test vxml: Add which volume could not be found to error message Add and utilize virsh_version_cmp pool: Need to handle when there are no volumes in the default pool .../Profile/04_verify_libvirt_cim_slp_profiles.py | 2 +- .../08_CreateDiskResourcePool.py | 30 +++++-- .../09_DeleteDiskPool.py | 19 +++-- .../10_create_storagevolume.py | 6 +- .../11_create_dir_storagevolume_errs.py | 5 +- .../12_create_netfs_storagevolume_errs.py | 5 +- .../13_delete_storagevolume.py | 5 +- .../14_delete_storagevolume_errs.py | 5 +- .../15_DiskPoolAutostart.py | 15 +++- .../SettingsDefineCapabilities/01_forward.py | 1 - .../libvirt-cim/cimtest/VSSD/06_duplicate_uuid.py | 2 +- .../06_addresource.py | 2 +- .../08_modifyresource.py | 4 +- .../13_refconfig_additional_devs.py | 4 +- .../15_mod_system_settings.py | 11 ++- .../18_define_sys_bridge.py | 2 +- .../19_definenetwork_ers.py | 23 +++--- .../22_addmulti_brg_interface.py | 2 +- .../27_definesystem_macvtap_dev.py | 19 ++++- .../28_definesystem_with_vsi_profile.py | 15 ++++ suites/libvirt-cim/lib/XenKvmLib/common_util.py | 93 +++++++++++++++++----- suites/libvirt-cim/lib/XenKvmLib/const.py | 6 +- suites/libvirt-cim/lib/XenKvmLib/pool.py | 10 +-- suites/libvirt-cim/lib/XenKvmLib/rasd.py | 10 ++- suites/libvirt-cim/lib/XenKvmLib/test_xml.py | 2 +- suites/libvirt-cim/lib/XenKvmLib/vxml.py | 4 +- suites/libvirt-cim/lib/XenKvmLib/xm_virt_util.py | 6 ++ suites/libvirt-cim/main.py | 15 ++++ 28 files changed, 237 insertions(+), 86 deletions(-) -- 1.8.1.4

2 19

Re: [Libvirt-cim] [PATCH 3/3] make lldptool command and support output configurable
by John Ferlan 28 Apr '13

28 Apr '13

For some reason the original email never made it into my inbox, so I've had to cut-n-paste from the mail list archives and I hope I picked the right message-id for the messages to line up properly!!! > From: Xu Wang <cngesaint outlook com> > There's no commit message? > Signed-off-by: Xu Wang <cngesaint outlook com> > --- > libvirt-cim.conf | 14 ++++++++++ > libxkutil/misc_util.c | 18 +++++++++++++ > libxkutil/misc_util.h | 2 + > src/Virt_SwitchService.c | 63 ++++++++++++++++++++++++++++++++++++++-------- > 4 files changed, 86 insertions(+), 11 deletions(-) > > diff --git a/libvirt-cim.conf b/libvirt-cim.conf > index 3244ee3..396dac9 100644 > --- a/libvirt-cim.conf > +++ b/libvirt-cim.conf > @@ -38,3 +38,17 @@ > # Default value: false > # > # force_use_qemu = false; > + > +# lldptool_query_options (string) > +# Defines the command used in SwitchService to query VEPA support, will be > +# used as "lldptool -i [INTERFACE] [OPTIONS]" > +# > +# lldptool_query_options = "-t -g ncb -V evbcfg"; > + > +# vsi_support_key_string (string) > +# Defines the string used in SwitchService to search in lldptool's output > +# When lldptool updates its output, please update the new output into the > +# output set. add comma between items. > +# > +# vsi_support_key_string = "{ supported forwarding mode: (0x40) reflective relay," > +# " supported capabilities: (0x7) RTE ECP VDP}"; The code later indicates all the strings have to be there; however, the comments here do not match that. There's also no mention here that at most 8 lines can be added. Hopefully we never need to go beyond that number (however arbitrary 8 is). Is the semicolon mandatory? Seems to me whatever format is required needs to be well documented. With respect to the strings and the parsing done, should I 'assume' that the config_lookup_string() will know how to read the config file where elements span multiple lines.... In particular, is "line1," "line2" read/returned as "line1,line2"? I would think a continuation "\" marker would be necessary: "line1," \ "line2" I would also think the comma would be outside the quotes, but don't have experience with the API in question. Since I don't have a way to test this - I'm curious while coding this up if you checked that both lines were read when parsing? Secondarily, there's some extraneous spaces which perhaps need to be cleaned up. Since the code later on will separate based on ",". What happens if message #3 comes along some day as: "support feature1, feature2, and feature3: (0x83) mumbly fratz" > diff --git a/libxkutil/misc_util.c b/libxkutil/misc_util.c > index 4c0b0a1..6ce8dca 100644 > --- a/libxkutil/misc_util.c > +++ b/libxkutil/misc_util.c > @@ -244,6 +244,24 @@ const char *get_mig_ssh_tmp_key(void) > return prop.value_string; > } > > +const char *get_lldptool_query_options(void) > +{ > + static LibvirtcimConfigProperty prop = { > + "lldptool_query_options", CONFIG_STRING, {0}, 0}; > + > + libvirt_cim_config_get(&prop); > + return prop.value_string; > +} > + > +const char *get_vsi_support_key_string(void) > +{ > + static LibvirtcimConfigProperty prop = { > + "vsi_support_key_string", CONFIG_STRING, {0}, 0};; > + > + libvirt_cim_config_get(&prop); > + return prop.value_string; > +} > + > virConnectPtr connect_by_classname(const CMPIBroker *broker, > const char *classname, > CMPIStatus *s) > diff --git a/libxkutil/misc_util.h b/libxkutil/misc_util.h > index 9e6b419..4eb588d 100644 > --- a/libxkutil/misc_util.h > +++ b/libxkutil/misc_util.h > @@ -155,6 +155,8 @@ int virt_set_status(const CMPIBroker *broker, > /* get libvirt-cim config */ > const char *get_mig_ssh_tmp_key(void); > bool get_force_use_qemu(void); > +const char *get_lldptool_query_options(void); > +const char *get_vsi_support_key_string(void); > > /* > * Local Variables: > diff --git a/src/Virt_SwitchService.c b/src/Virt_SwitchService.c > index 8991426..f7bafbf 100644 > --- a/src/Virt_SwitchService.c > +++ b/src/Virt_SwitchService.c > @@ -46,12 +46,26 @@ static CMPIStatus check_vsi_support(char *command) > CMPIStatus s = {CMPI_RC_OK, NULL}; > char buff[MAX_LEN]; > FILE *stream = NULL; > - const char *searchStr[] = {" supported forwarding mode: " > - "(0x40) reflective relay", > - " supported capabilities: " > - "(0x07) RTE ECP VDP", > - NULL}; > - int matched = 0; > + char *searchStr[8]; /* maximum items of vsi support output */ > + int count = 0; > + const char *user_settings = get_vsi_support_key_string(); > + char *vsi_support_key_string = NULL; > + char *delim = "{},"; > + int matched = 0; > + char *temp = NULL; > + > + if (!user_settings) { > + user_settings = "{ supported forwarding mode: " > + "(0x40) reflective relay," > + " supported capabilities: " > + "(0x7) RTE ECP VDP}"; > + } > + > + vsi_support_key_string = strdup(user_settings); There's a memory leak hear if 'user_settings' is returned from the api. That code will strdup() whatever is returned, it's returned here, strdup()'d again and leaked. > + if (vsi_support_key_string == NULL) { > + CU_DEBUG("strdup vsi_support_key_string failed!"); I think you need to call cu_statusf here; otherwise, 's' returns as {CMPI_RC_OK, NULL} > + goto out; > + } > > // Run lldptool command to find vsi support. > stream = popen(command, "r"); > @@ -63,6 +77,25 @@ static CMPIStatus check_vsi_support(char *command) > goto out; > } > > + /* Slice vsi_support_key_string into items */ > + searchStr[count] = strtok_r(vsi_support_key_string, delim, &temp); > + if (searchStr[count] == NULL) { > + CU_DEBUG("searchStr fetch failed when calling strtok_r!"); > + } else { > + CU_DEBUG("searchStr[%d]: %s", count, searchStr[count]); > + count++; > + } > + > + while ((searchStr[count] = strtok_r(NULL, delim, &temp))) { > + if (count >= 7) { > + CU_DEBUG("WARN: searchStr is full, left aborted!"); > + break; > + } else { > + CU_DEBUG("searchStr[%d]: %s", count, searchStr[count]); > + count++; > + } > + } > + What if count == 0? That is can the element in the conf file be "{}"? > // Read the output of the command. > while (fgets(buff, MAX_LEN, stream) != NULL) { > int i = 0; > @@ -81,16 +114,18 @@ static CMPIStatus check_vsi_support(char *command) > } > /* All the search strings were found in the output of this > command. */ > - if (matched == 2) { > + if (matched == count) { > cu_statusf(_BROKER, &s, CMPI_RC_OK, "VSI supported"); > - goto out;; > + goto out; > } > } > + > cu_statusf(_BROKER, &s, > CMPI_RC_ERR_NOT_FOUND, > "No VSI Support found"); > > - out: > + out: > + free(vsi_support_key_string); > if (stream != NULL) > pclose(stream); > return s; > @@ -214,6 +249,7 @@ static CMPIStatus get_switchservice(const CMPIObjectPath *reference, > int i; > char **if_list; > char cmd[MAX_LEN]; > + const char *lldptool_query_options = NULL; > > *_inst = NULL; > conn = connect_by_classname(broker, CLASSNAME(reference), &s); > @@ -257,10 +293,15 @@ static CMPIStatus get_switchservice(const CMPIObjectPath *reference, > > CU_DEBUG("Found %d interfaces", count); > > + lldptool_query_options = get_lldptool_query_options(); > + if (!lldptool_query_options) { > + lldptool_query_options = "-t -g ncb -V evbcfg"; > + } Memory leak if we are successful from get_lldptool_query_options() John > > for (i=0; i<count; i++) { > - sprintf(cmd, "lldptool -i %s -t -V evbcfg", if_list[i]); > - CU_DEBUG("running command %s ...", cmd); > + sprintf(cmd, "lldptool -i %s %s", > + if_list[i], lldptool_query_options); > + CU_DEBUG("running command [%s]", cmd); > s = check_vsi_support(cmd); > if (s.rc == CMPI_RC_OK) { > vsi = true; > -- > 1.7.1

2 2

[PATCH 1/3] make force use qemu configurable
by cngesaint＠outlook.com 27 Apr '13

27 Apr '13

From: Wenchao Xia <xiawenc(a)linux.vnet.ibm.com> Since in nested KVM, libvirt-cim doesn't handle it well now, add this option to make it run well with qemu wchi help development and test. Signed-off-by: Xu Wang <cngesaint(a)outlook.com> --- libvirt-cim.conf | 8 ++++++++ libxkutil/misc_util.c | 8 ++++++++ libxkutil/misc_util.h | 1 + src/Virt_VirtualSystemManagementService.c | 7 +++++++ 4 files changed, 24 insertions(+), 0 deletions(-) diff --git a/libvirt-cim.conf b/libvirt-cim.conf index 37d7b0f..3244ee3 100644 --- a/libvirt-cim.conf +++ b/libvirt-cim.conf @@ -30,3 +30,11 @@ # Default value: NULL, that is not set. # # migrate_ssh_temp_key = "/root/vm_migrate_tmp_id_rsa"; + +# force_use_qemu (bool) +# Since in nested KVM, libvirt-cim doesn't handler it well now, so add this +# option to make it run well with qemu which help development and test. +# Possible values: {true,false} +# Default value: false +# +# force_use_qemu = false; diff --git a/libxkutil/misc_util.c b/libxkutil/misc_util.c index 00eb4b1..4c0b0a1 100644 --- a/libxkutil/misc_util.c +++ b/libxkutil/misc_util.c @@ -227,6 +227,14 @@ static int is_read_only(void) return prop.value_bool; } +bool get_force_use_qemu(void) +{ + static LibvirtcimConfigProperty prop = { + "force_use_qemu", CONFIG_BOOL, {0}, 0}; + libvirt_cim_config_get(&prop); + return prop.value_bool; +} + const char *get_mig_ssh_tmp_key(void) { static LibvirtcimConfigProperty prop = { diff --git a/libxkutil/misc_util.h b/libxkutil/misc_util.h index 0f52290..9e6b419 100644 --- a/libxkutil/misc_util.h +++ b/libxkutil/misc_util.h @@ -154,6 +154,7 @@ int virt_set_status(const CMPIBroker *broker, /* get libvirt-cim config */ const char *get_mig_ssh_tmp_key(void); +bool get_force_use_qemu(void); /* * Local Variables: diff --git a/src/Virt_VirtualSystemManagementService.c b/src/Virt_VirtualSystemManagementService.c index cbb646d..4e93ef0 100644 --- a/src/Virt_VirtualSystemManagementService.c +++ b/src/Virt_VirtualSystemManagementService.c @@ -394,6 +394,13 @@ static bool system_has_kvm(const char *pfx) virConnectPtr conn; char *caps = NULL; bool kvm = false; + bool force_use_qemu = get_force_use_qemu(); + + /* hack for nested KVM */ + if (force_use_qemu) { + CU_DEBUG("Enter force use qemu mode!"); + return false; + } conn = connect_by_classname(_BROKER, pfx, &s); if ((conn == NULL) || (s.rc != CMPI_RC_OK)) { -- 1.7.1

4 6

Re: [Libvirt-cim] [PATCH 9/9] 19 - resolve issues found in test
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)outlook.com> --- .../cimtest/VirtualSystemManagementService/19_definenetwork_ers.py | 1 - 1 file changed, 1 deletion(-) diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/19_definenetwork_ers.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/19_definenetwork_ers.py index 4cda327..cf461d1 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/19_definenetwork_ers.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/19_definenetwork_ers.py @@ -89,7 +89,6 @@ def main(): expected_values['invalid']['bridge'] = inv_br_str else: - logger.error('DEBUG libvirt_version=%s', libvirt_version) if libvirt_version >= "0.7.0": expected_values['empty']['network'] = inv_empty_network expected_values['invalid']['network'] = inv_empty_network -- 1.8.1.4

1 0

Re: [Libvirt-cim] [PATCH 8/9] On Fedora systems default to using 'em1' instead of 'eth1'
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)outlook.com> Fedora changed the default device naming scheme, see http://fedoraproject.org/wiki/Features/ConsistentNetworkDeviceNaming So if we're running on Fedora, let's "default" to "em1" although that doesn't guarantee that we will find what we're looking for. This code probably needs some mechanism to detect with interfaces are available, then query those interfaces to find one that supports the feature we want. --- .../27_definesystem_macvtap_dev.py | 19 ++++++++++++++++++- .../28_definesystem_with_vsi_profile.py | 15 +++++++++++++++ 2 files changed, 33 insertions(+), 1 deletion(-) diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/27_definesystem_macvtap_dev.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/27_definesystem_macvtap_dev.py index 8c20781..36bf52f 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/27_definesystem_macvtap_dev.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/27_definesystem_macvtap_dev.py @@ -32,6 +32,7 @@ import sys from CimTest.Globals import logger from CimTest.ReturnCodes import FAIL, PASS, SKIP +from VirtLib import utils from XenKvmLib.classes import get_typed_class, inst_to_mof from XenKvmLib.rasd import get_default_rasds from XenKvmLib.const import do_main, get_provider_version @@ -101,6 +102,7 @@ def verify_net_rasd(ip, virt, target_dev, source_dev, guest_name): @do_main(sup_types) def main(): options = main.options + server = options.ip status = FAIL @@ -113,7 +115,22 @@ def main(): cxml = get_class(options.virt)(test_dom) target_dev = "vepa" - source_dev = "eth1" + + # Fedora changed the default device naming scheme, see + # http://fedoraproject.org/wiki/Features/ConsistentNetworkDeviceNaming + # + # So if we're running on Fedora, let's "default" to "em1" although that + # doesn't guarantee that we will find what we're looking for. This code + # probably needs some mechanism to detect with interfaces are available, + # then query those interfaces to find one that supports the feature we + # want. If not are found and XFAIL could be generated indicating so. + # + cmd = 'cat /etc/issue | grep -v ^$ | egrep "Fedora"' + rc, out = utils.run_remote(server, cmd) + if rc == 0: + source_dev = "em1" + else: + source_dev = "eth1" guest_defined = False diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/28_definesystem_with_vsi_profile.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/28_definesystem_with_vsi_profile.py index d14dda3..2b108f9 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/28_definesystem_with_vsi_profile.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/28_definesystem_with_vsi_profile.py @@ -35,6 +35,7 @@ import sys from CimTest.Globals import logger from CimTest.ReturnCodes import FAIL, PASS, SKIP, XFAIL_RC, XFAIL +from VirtLib import utils from XenKvmLib.classes import get_typed_class, inst_to_mof from XenKvmLib.rasd import get_default_rasds from XenKvmLib.const import do_main, get_provider_version @@ -173,6 +174,20 @@ def main(): 'VSITypeIDVersion' : "1" } + # Fedora changed the default device naming scheme, see + # http://fedoraproject.org/wiki/Features/ConsistentNetworkDeviceNaming + # + # So if we're running on Fedora, let's "default" to "em1" although that + # doesn't guarantee that we will find what we're looking for. This code + # probably needs some mechanism to detect with interfaces are available, + # then query those interfaces to find one that supports the feature we + # want. If not are found and XFAIL could be generated indicating so. + # + cmd = 'cat /etc/issue | grep -v ^$ | egrep "Fedora"' + rc, out = utils.run_remote(server, cmd) + if rc == 0: + vsi_defaults['SourceDevice'] = "em1" + nrasd_cn = get_typed_class(virt, 'NetResourceAllocationSettingData') status = FAIL cxml = None -- 1.8.1.4

1 0

Re: [Libvirt-cim] [PATCH 7/9] Resolve issues found in test.
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)outlook.com> 1. Use a MAC that isn't a multicast MAC. 2. Error messages changed in libvirt "0.9.8", so we need to account for that --- .../19_definenetwork_ers.py | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/19_definenetwork_ers.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/19_definenetwork_ers.py index ad7c1ce..4cda327 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/19_definenetwork_ers.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/19_definenetwork_ers.py @@ -37,7 +37,7 @@ from XenKvmLib.xm_virt_util import virsh_version sup_types = ['Xen', 'KVM', 'XenFV'] default_dom = 'brgtest_domain' -nmac = '99:aa:bb:cc:ee:ff' +nmac = '88:aa:bb:cc:ee:ff' npool_name = default_network_name + str(random.randint(1, 100)) brg_name = "br" + str(random.randint(1, 100)) @@ -89,11 +89,16 @@ def main(): expected_values['invalid']['bridge'] = inv_br_str else: + logger.error('DEBUG libvirt_version=%s', libvirt_version) if libvirt_version >= "0.7.0": expected_values['empty']['network'] = inv_empty_network expected_values['invalid']['network'] = inv_empty_network - expected_values['invalid']['bridge'] = " Failed to add tap "\ - "interface to bridge" + if libvirt_version >= "0.9.8": + expected_values['invalid']['bridge'] = "Cannot get interface "\ + "MTU on 'invalid'" + else: + expected_values['invalid']['bridge'] = " Failed to add tap "\ + "interface to bridge" tc_scen = { @@ -114,7 +119,8 @@ def main(): status = PASS for nettype in nettypes: for tc, field in tc_scen.iteritems(): - logger.error("DEBUG nettype is %s, field is %s, tc is %s", nettype, field, tc) + logger.error("DEBUG nettype is %s, field is %s, tc is %s", + nettype, field, tc) cxml = vxml.get_class(options.virt)(default_dom, mac=nmac, ntype=nettype, net_name=field) @@ -122,7 +128,7 @@ def main(): try: ret = cxml.cim_define(options.ip) - if not ret: + if not ret: status = verify_error(exp_rc, exp_desc, cxml) if status != PASS: # There are few libvirt version between 0.7.0 @@ -138,7 +144,7 @@ def main(): cxml.err_rc, cxml.err_desc)) continue ret = cxml.cim_start(options.ip) - if ret: + if ret: status = verify_error(exp_rc, exp_desc, cxml) cxml.undefine(options.ip) if status != PASS: -- 1.8.1.4

1 0

Re: [Libvirt-cim] [PATCH 6/9] Fix os_status passing to reporter functions
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)outlook.com> If the status returned is an errno value, then os_status will be set outside of the bounds of the 'rc' array in Reporter.py which will cause a KeyError exception. Find, message, and adjust - we're failing anyway. --- suites/libvirt-cim/main.py | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/suites/libvirt-cim/main.py b/suites/libvirt-cim/main.py index a4e33e6..e5f3526 100644 --- a/suites/libvirt-cim/main.py +++ b/suites/libvirt-cim/main.py @@ -23,12 +23,14 @@ # from time import time +from time import sleep from optparse import OptionParser import os import sys sys.path.append('../../lib') import TestSuite from CimTest.Globals import logger, log_param +from CimTest.ReturnCodes import PASS, FAIL, XFAIL, SKIP import commands from VirtLib import groups import ConfigParser @@ -274,6 +276,14 @@ def main(options, args): os_status = os.WEXITSTATUS(status) + # status should be from our test; however, if there's an OS level + # failure, it could be set to errno. But that's included in our + # output, so just set it to FAIL; otherwise, we get a KeyError + # in Reporter.py when trying to index it's 'rc' record + if os_status not in (PASS, FAIL, XFAIL, SKIP): + logger.error("Changing os_status from %d to FAIL", os_status) + os_status = FAIL + testsuite.print_results(test['group'], test['test'], os_status, output) exec_time = end_time - start_time @@ -282,6 +292,11 @@ def main(options, args): if options.print_exec_time: print_exec_time(testsuite, exec_time, " Test execution time:") + # Give ourselves a 3 second pause before running the next + # test to help ensure we have cleaned things up properly just + # in case the cimserver is a little slow to respond + sleep(3) + testsuite.debug("%s\n" % div) if options.print_exec_time: -- 1.8.1.4

1 0

Re: [Libvirt-cim] [PATCH 5/9] Fix nfs-server lookup code
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)outlook.com> I have a small confusing about checking linux version under /etc/issue. Can we use search() to find if 'Red Hat' or 'Fedora'is contained in the output? It seems to make code shorter and logic simpler:-) The 'get_nfs_bin()', 'nfs_config()', and 'nfs_netfs_setup()' API's needed adjustment to handle using systemd rather than looking for /etc/init.d/nfs --- suites/libvirt-cim/lib/XenKvmLib/common_util.py | 77 ++++++++++++++++++++----- 1 file changed, 61 insertions(+), 16 deletions(-) diff --git a/suites/libvirt-cim/lib/XenKvmLib/common_util.py b/suites/libvirt-cim/lib/XenKvmLib/common_util.py index f145811..43e5e2c 100644 --- a/suites/libvirt-cim/lib/XenKvmLib/common_util.py +++ b/suites/libvirt-cim/lib/XenKvmLib/common_util.py @@ -493,27 +493,65 @@ def parse_instance_id(instid): def get_nfs_bin(server): cmd = 'cat /etc/issue | grep -v ^$ | egrep "Red Hat|Fedora"' rc, out = utils.run_remote(server, cmd) + is_systemd = 0 if rc != 0: #SLES nfs_server_bin = "/etc/init.d/nfsserver" else: + # Default, but allow it to be changed nfs_server_bin = "/etc/init.d/nfs" - return nfs_server_bin - -def nfs_config(server, nfs_server_bin): - cmd = "ps aux | grep -v -e nfsiod -e grep | grep nfsd" + # Check for 'systemd' being used + # Fedora 15 seems to have been the first place this was the default + # RHEL 7 and beyond will also use the mechanism. + # Since 'out' returns the fetched string, let's parse it a bit more + # looking for the number after the string 'release' which happens to + # be the version. On Fedora systems it's + # "Fedora release 18 (Spherical Cow)" while on RHEL systems it's: + # "Red Hat Enterprise Linux Server release 6.4 (Santiago)" + # + elems = out.split() + if elems[0] == 'Fedora' or (elems[0] == 'Red' and elems[1] == 'Hat'): + for i in range(1, len(elems)): + if elems[i] == 'release': + if (elems[0] == 'Fedora' and int(elems[i+1]) >= 15) or \ + (elems[0] == 'Red' and int(elems[i+1]) >= 7): + # Handle this differently - the command would be + # "systemctl {start|restart|status} nfs" + nfs_server_bin = "systemctl %s nfs" + is_systemd = 1 + break + + return nfs_server_bin, is_systemd + +def nfs_config(server, nfs_server_bin, is_systemd): + if is_systemd == 0: + cmd = "ps aux | grep -v -e nfsiod -e grep | grep nfsd" + else: + cmd = "systemctl | grep nfs-server" rc, out = utils.run_remote(server, cmd) # if NFS services is not found on the machine, start it.. if rc != PASS : # Check if NFS server is installed ... - if not os.path.exists(nfs_server_bin): - logger.error("NFS server '%s' does not seem to be installed "\ - "on '%s'", nfs_server_bin, server) - return SKIP + if is_systemd == 0: + if not os.path.exists(nfs_server_bin): + logger.error("NFS server '%s' does not seem to be installed "\ + "on '%s'", nfs_server_bin, server) + return SKIP + else: + # Works on Fedora and RHEL6 + cmd = "rpm -q nfs-utils" + rc, out = utils.run_remote(server, cmd) + if rc != PASS : + logger.error("NFS server package nfs-utils does not seem "\ + "to be installed on '%s'", server) + return SKIP # Start the nfs server ... - nfs_server_cmd = "%s start" % nfs_server_bin + if is_systemd == 0: + nfs_server_cmd = "%s start" % nfs_server_bin + else: + nfs_server_cmd = nfs_server_bin % "start" rc, out = utils.run_remote(server, nfs_server_cmd) if rc != PASS: logger.error("Could not start the nfsserver on '%s'", server) @@ -573,13 +611,16 @@ def netfs_cleanup(server, pool_attr): move_file(back_exports_file, exports_file) # restart the nfs server - nfs_server_bin = get_nfs_bin(server) - nfs_server_cmd = "%s restart" % nfs_server_bin + nfs_server_bin, is_systemd = get_nfs_bin(server) + if is_systemd == 0: + nfs_server_cmd = "%s restart" % nfs_server_bin + else: + nfs_server_cmd = nfs_server_bin % "restart" rc, out = utils.run_remote(server, nfs_server_cmd) if rc != PASS: logger.error("Could not restart NFS server on '%s'" % server) -def netfs_config(server, nfs_server_bin, dest_dir_to_mnt): +def netfs_config(server, nfs_server_bin, dest_dir_to_mnt, is_systemd): src_dir_for_mnt = mkdtemp() try: @@ -601,7 +642,10 @@ def netfs_config(server, nfs_server_bin, dest_dir_to_mnt): % (src_dir_for_mnt, dest_dir_to_mnt)) # Restart the nfs server.... - nfs_server_cmd = "%s restart" % nfs_server_bin + if is_systemd == 0: + nfs_server_cmd = "%s restart" % nfs_server_bin + else: + nfs_server_cmd = nfs_server_bin % "restart" rc, out = utils.run_remote(server, nfs_server_cmd) if rc != PASS: raise Exception("Could not restart NFS server on '%s'" % server) @@ -615,12 +659,12 @@ def netfs_config(server, nfs_server_bin, dest_dir_to_mnt): return PASS, src_dir_for_mnt def nfs_netfs_setup(server): - nfs_server_bin = get_nfs_bin(server) + nfs_server_bin, is_systemd = get_nfs_bin(server) dest_dir = mkdtemp() # Before going ahead verify that nfs server is available on machine.. - ret = nfs_config(server, nfs_server_bin) + ret = nfs_config(server, nfs_server_bin, is_systemd) if ret != PASS: logger.error("Failed to configure NFS on '%s'", server) logger.info("Trying to look for nfs mounted dir on '%s'...", server) @@ -631,7 +675,8 @@ def nfs_netfs_setup(server): else: return PASS, server, src_dir, dest_dir else: - ret, src_dir = netfs_config(server, nfs_server_bin, dest_dir) + ret, src_dir = netfs_config(server, nfs_server_bin, \ + dest_dir, is_systemd) if ret != PASS: logger.error("Failed to configure netfs on '%s'", server) return ret, None, None, None -- 1.8.1.4

1 0

Re: [Libvirt-cim] [PATCH 4/9] Use symbols as named in libvirt-cim for easier reference
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)outlook.com> --- .../VirtualSystemManagementService/15_mod_system_settings.py | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/15_mod_system_settings.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/15_mod_system_settings.py index e576fca..0be3ae0 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/15_mod_system_settings.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/15_mod_system_settings.py @@ -37,7 +37,10 @@ from XenKvmLib.xm_virt_util import domain_list, active_domain_list, \ sup_types = ['Xen', 'KVM', 'XenFV', 'LXC'] default_dom = 'rstest_domain' cpu = 2 -RECOVERY_VAL = 3 +# AutomaticRecoveryAction +NONE_VAL = 2 +RESTART_VAL = 3 + DEFINED_STATE = 3 bug = "00008" f9_bug = "00010" @@ -113,7 +116,7 @@ def main(): raise Expcetion("Failed to get the VSSD instance for %s" % \ default_dom) - val = pywbem.cim_types.Uint16(RECOVERY_VAL) + val = pywbem.cim_types.Uint16(RESTART_VAL) inst['AutomaticRecoveryAction'] = val vssd = inst_to_mof(inst) @@ -135,9 +138,9 @@ def main(): raise Exception("Failed to get the VSSD instance for %s" % \ default_dom) - if inst.AutomaticRecoveryAction != RECOVERY_VAL: + if inst.AutomaticRecoveryAction != RESTART_VAL: logger.error("Exp AutomaticRecoveryAction=%d, got %d", - RECOVERY_VAL, inst.AutomaticRecoveryAction) + RESTART_VAL, inst.AutomaticRecoveryAction) raise Exception("%s not updated properly" % default_dom) status = PASS -- 1.8.1.4

1 0

Re: [Libvirt-cim] [PATCH 3/9] Create a temporary directory for disk pool tests
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)outlook.com> From: cngesaint(a)outlook.com To: cngesaint(a)outlook.com Subject: [Libvirt-cim] [PATCH 3/9] Create a temporary directory for disk pool tests Date: Wed, 24 Apr 2013 06:42:05 +0000 We cannot have two storage pools referencing the same path. Since the cimtest-diskpool is already created at /var/lib/libvirt/images, we'll use that (e.g. _image_dir symbol) to create a subdirectory for the test. We'll also delete that directory when we're done. --- .../08_CreateDiskResourcePool.py | 26 +++++++++++++++++----- .../09_DeleteDiskPool.py | 19 +++++++++++----- .../15_DiskPoolAutostart.py | 11 ++++++++- 3 files changed, 43 insertions(+), 13 deletions(-) diff --git a/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/08_CreateDiskResourcePool.py b/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/08_CreateDiskResourcePool.py index c82b5b0..636f59c 100644 --- a/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/08_CreateDiskResourcePool.py +++ b/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/08_CreateDiskResourcePool.py @@ -48,6 +48,7 @@ # -Date: 26.05.2009 import sys +import os from CimTest.Globals import logger from XenKvmLib.xm_virt_util import virsh_version from CimTest.ReturnCodes import FAIL, PASS, SKIP @@ -55,8 +56,9 @@ from XenKvmLib.const import do_main, platform_sup from XenKvmLib.classes import get_typed_class from XenKvmLib.common_util import destroy_diskpool, nfs_netfs_setup, \ netfs_cleanup -from XenKvmLib.pool import create_pool, verify_pool, undefine_diskpool -from XenKvmLib.const import get_provider_version +from XenKvmLib.pool import create_pool, verify_pool, undefine_diskpool, \ + DIR_POOL, NETFS_POOL +from XenKvmLib.const import get_provider_version, _image_dir libvirt_disk_pool_support=837 libvirt_netfs_pool_support=869 @@ -94,19 +96,20 @@ def main(): curr_cim_rev, changeset = get_provider_version(virt, server) if curr_cim_rev >= libvirt_disk_pool_support: - dp_types["DISK_POOL_DIR"] = 1 + dp_types["DISK_POOL_DIR"] = DIR_POOL if curr_cim_rev >= libvirt_netfs_pool_support: - dp_types["DISK_POOL_NETFS"] = 3 + dp_types["DISK_POOL_NETFS"] = NETFS_POOL if len(dp_types) == 0 : - logger.info("No disk pool types in list , hence skipping the test...") + logger.info("No disk pool types in list, hence skipping the test...") return SKIP status = FAIL pool_attr = None # For now the test case support only the creation of # dir type disk pool, netfs later change to fs and disk pooltypes etc - for key, value in dp_types.iteritems(): + for key, value in dp_types.iteritems(): + del_path = False try: logger.info("Verifying '%s'.....", key) test_pool = key @@ -115,6 +118,15 @@ def main(): if status != PASS: return FAIL + # Cannot have two pools that use the same location/path, so + # since cimtest-diskpool already exists + if key == 'DISK_POOL_DIR': + path = os.path.join(_image_dir, 'temppool') + if not os.path.exists(path): + os.mkdir(path) + del_path = True + pool_attr["Path"] = path + status = create_pool(server, virt, test_pool, pool_attr, mode_type=value, pool_type= "DiskPool") @@ -152,6 +164,8 @@ def main(): if key == 'DISK_POOL_NETFS': netfs_cleanup(server, pool_attr) + if del_path: + os.rmdir(path) return status if __name__ == "__main__": diff --git a/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/09_DeleteDiskPool.py b/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/09_DeleteDiskPool.py index 31e3f22..8bd15e2 100644 --- a/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/09_DeleteDiskPool.py +++ b/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/09_DeleteDiskPool.py @@ -44,6 +44,7 @@ # -Date: 26.05.2009 import sys +import os import pywbem from XenKvmLib import rpcs_service from CimTest.Globals import logger @@ -88,13 +89,17 @@ def main(): elif curr_cim_rev >= libvirt_cim_child_pool_rev: + del_path = False try: - pool_attr = { "Path" : _image_dir } + path = os.path.join(_image_dir, 'deltest') + if not os.path.exists(path): + os.mkdir(path) + del_path = True + pool_attr = { "Path" : path } status = create_pool(server, virt, test_pool, pool_attr, pool_type="DiskPool", mode_type=TYPE) if status != PASS: - logger.error("Failed to create diskpool '%s'", test_pool) - return status + raise Exception("Failed to create diskpool '%s'" % test_pool) status = verify_pool(server, virt, test_pool, pool_attr, pool_type="DiskPool") @@ -112,8 +117,8 @@ def main(): break if pool_settings == None: - logger.error("Failed to get poolsettings for '%s'", test_pool) - return FAIL + raise Exception("Failed to get poolsettings for '%s'" \ + % test_pool) rpcs_conn.DeleteResourcePool(Pool = pool_settings) pool = EnumInstances(server, dp) @@ -127,8 +132,10 @@ def main(): logger.error("Exception details: %s", details) destroy_diskpool(server, virt, test_pool) undefine_diskpool(server, virt, test_pool) - return FAIL + status = FAIL + if del_path: + os.rmdir(path) return status if __name__ == "__main__": diff --git a/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/15_DiskPoolAutostart.py b/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/15_DiskPoolAutostart.py index b6b758c..b7e72a8 100644 --- a/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/15_DiskPoolAutostart.py +++ b/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/15_DiskPoolAutostart.py @@ -23,6 +23,7 @@ # -Date: 04.14.2011 import sys +import os from pywbem import cim_types from CimTest.Globals import logger from XenKvmLib.xm_virt_util import virsh_version @@ -77,12 +78,17 @@ def main(): pool_attr = None key = 'DISK_POOL_DIR' value = 1 + del_path = False try: logger.info("Verifying '%s'.....", key) test_pool = key - pool_attr = { "Path" : "/var/lib/libvirt/images", + pool_attr = { "Path" : "/var/lib/libvirt/images/autotest", "Autostart" : cim_types.Uint16(1) } + if not os.path.exists(pool_attr["Path"]): + os.mkdir(pool_attr["Path"]) + del_path = True + status = create_pool(server, virt, test_pool, pool_attr, mode_type=value, pool_type= "DiskPool") @@ -112,6 +118,9 @@ def main(): status = FAIL logger.error("Exception details: %s", details) + if del_path: + os.rmdir(pool_attr["Path"]) + return status if __name__ == "__main__": -- 1.8.1.4

1 0

Re: [Libvirt-cim] [PATCH 2/9] Change the MAC from "99:" to "88:"
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)outlook.com> From: cngesaint(a)outlook.com To: cngesaint(a)outlook.com Subject: [Libvirt-cim] [PATCH 2/9] Change the MAC from "99:" to "88:" Date: Wed, 24 Apr 2013 06:19:43 +0000 Using a MAC starting with "99:" is not legal. That ends up being a Multicast Address since the low order bit of the first byte of the address (in hex) is one. That is "99" is 0x10011001" in hex, where the least significant bit is a "1" (one) indicating a multicast. --- suites/libvirt-cim/cimtest/VSSD/06_duplicate_uuid.py | 2 +- .../cimtest/VirtualSystemManagementService/06_addresource.py | 2 +- .../cimtest/VirtualSystemManagementService/08_modifyresource.py | 4 ++-- .../VirtualSystemManagementService/13_refconfig_additional_devs.py | 4 ++-- .../cimtest/VirtualSystemManagementService/18_define_sys_bridge.py | 2 +- .../VirtualSystemManagementService/22_addmulti_brg_interface.py | 2 +- suites/libvirt-cim/lib/XenKvmLib/const.py | 6 +++--- suites/libvirt-cim/lib/XenKvmLib/test_xml.py | 2 +- 8 files changed, 12 insertions(+), 12 deletions(-) diff --git a/suites/libvirt-cim/cimtest/VSSD/06_duplicate_uuid.py b/suites/libvirt-cim/cimtest/VSSD/06_duplicate_uuid.py index c5d8a22..be82335 100644 --- a/suites/libvirt-cim/cimtest/VSSD/06_duplicate_uuid.py +++ b/suites/libvirt-cim/cimtest/VSSD/06_duplicate_uuid.py @@ -40,7 +40,7 @@ from XenKvmLib.const import get_provider_version sup_types = ['Xen', 'KVM', 'XenFV', 'LXC'] default_dom = 'uuid_domain' test_dom = 'test_domain' -nmac = '99:aa:bb:cc:ee:ff' +nmac = '88:aa:bb:cc:ee:ff' duplicate_uuid_support = 915 def get_vssd(ip, virt, dom): diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/06_addresource.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/06_addresource.py index fdcb84a..ac18c69 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/06_addresource.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/06_addresource.py @@ -38,7 +38,7 @@ from XenKvmLib.common_util import create_netpool_conf, destroy_netpool sup_types = ['Xen', 'KVM', 'XenFV'] default_dom = 'rstest_domain' -nmac = '99:aa:bb:cc:ee:ff' +nmac = '88:aa:bb:cc:ee:ff' ntype = 'network' npool_name = default_network_name + str(random.randint(1, 100)) diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/08_modifyresource.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/08_modifyresource.py index df58d1f..a7c3871 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/08_modifyresource.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/08_modifyresource.py @@ -43,9 +43,9 @@ cpu = 2 ncpu = 1 nmem = 131072 new_int = randint(10, 99) -new_mac1 = "11:%s:22:%s:33:%s" % (new_int, new_int, new_int) +new_mac1 = "88:%s:22:%s:33:%s" % (new_int, new_int, new_int) new_int += 1 -new_mac2 = "11:%s:22:%s:33:%s" % (new_int, new_int, new_int) +new_mac2 = "88:%s:22:%s:33:%s" % (new_int, new_int, new_int) def cleanup_env(ip, cxml): cxml.destroy(ip) diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/13_refconfig_additional_devs.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/13_refconfig_additional_devs.py index 6274527..a47db8f 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/13_refconfig_additional_devs.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/13_refconfig_additional_devs.py @@ -38,8 +38,8 @@ from XenKvmLib.vxml import get_class sup_types = ['Xen', 'XenFV', 'KVM'] test_dom = 'rstest_domain' test_dom2 = 'rstest_domain2' -mac1 = '99:aa:bb:cc:ee:ff' -mac2 = '99:aa:bb:cc:ee:aa' +mac1 = '88:aa:bb:cc:ee:ff' +mac2 = '88:aa:bb:cc:ee:aa' REQUESTED_STATE = 2 TIME = "00000000000000.000000:000" diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/18_define_sys_bridge.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/18_define_sys_bridge.py index e02de29..66b3e45 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/18_define_sys_bridge.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/18_define_sys_bridge.py @@ -32,7 +32,7 @@ from XenKvmLib.common_util import create_netpool_conf, destroy_netpool sup_types = ['Xen', 'KVM', 'XenFV'] default_dom = 'brgtest_domain' -nmac = '99:aa:bb:cc:ee:ff' +nmac = '88:aa:bb:cc:ee:ff' npool_name = default_network_name + str(random.randint(1, 100)) brg_name = "br" + str(random.randint(1, 100)) diff --git a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/22_addmulti_brg_interface.py b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/22_addmulti_brg_interface.py index 9abb21b..36d1873 100644 --- a/suites/libvirt-cim/cimtest/VirtualSystemManagementService/22_addmulti_brg_interface.py +++ b/suites/libvirt-cim/cimtest/VirtualSystemManagementService/22_addmulti_brg_interface.py @@ -36,7 +36,7 @@ sup_types = ['Xen', 'KVM', 'XenFV'] test_dom = "my_domain1" default_net = "my_network0" test_net = "my_network1" -test_mac = '99:aa:bb:cc:ee:ff' +test_mac = '88:aa:bb:cc:ee:ff' default_mac = "00:11:33:33:44:55" ntype = 'bridge' default_brg = 'mybr0' diff --git a/suites/libvirt-cim/lib/XenKvmLib/const.py b/suites/libvirt-cim/lib/XenKvmLib/const.py index 6dea644..6701d36 100755 --- a/suites/libvirt-cim/lib/XenKvmLib/const.py +++ b/suites/libvirt-cim/lib/XenKvmLib/const.py @@ -80,7 +80,7 @@ Xen_init_path = os.path.join(_image_dir, 'default-xen-initrd') Xen_disk_path = os.path.join(_image_dir, 'default-xen-dimage') Xen_secondary_disk_path = os.path.join(_image_dir, 'default-xen-dimage.2ND') Xen_default_disk_dev = 'xvda' -Xen_default_mac = '11:22:33:aa:bb:cc' +Xen_default_mac = '88:22:33:aa:bb:cc' # vxml.KVMXML KVM_default_emulator = '/usr/bin/qemu-system-x86_64' @@ -88,7 +88,7 @@ KVM_disk_path = os.path.join(_image_dir, 'default-kvm-dimage') KVM_secondary_disk_path = os.path.join(_image_dir, 'default-kvm-dimage.2ND') KVM_default_disk_dev = 'hda' KVM_default_cdrom_dev = 'hdc' -KVM_default_mac = '11:22:33:aa:bb:cc' +KVM_default_mac = '88:22:33:aa:bb:cc' # vxml.XenFVXML s, o = platform.architecture() @@ -109,7 +109,7 @@ LXC_default_emulator = '/usr/libexec/libvirt_lxc' LXC_default_tty = '/dev/ptmx' LXC_default_mp = '/tmp' LXC_default_source = '/var/lib/libvirt/images/lxc_files' -LXC_default_mac = '11:22:33:aa:bb:cc' +LXC_default_mac = '88:22:33:aa:bb:cc' LXC_netns_support = False parser = OptionParser() diff --git a/suites/libvirt-cim/lib/XenKvmLib/test_xml.py b/suites/libvirt-cim/lib/XenKvmLib/test_xml.py index b32ae4c..914dbb0 100755 --- a/suites/libvirt-cim/lib/XenKvmLib/test_xml.py +++ b/suites/libvirt-cim/lib/XenKvmLib/test_xml.py @@ -42,7 +42,7 @@ kernel_path = os.path.join(image_dir, 'default-xen-kernel') init_path = os.path.join(image_dir, 'default-xen-initrd') disk_path = os.path.join(image_dir, 'default-xen-dimage') -default_mac = '11:22:33:aa:bb:cc' +default_mac = '88:22:33:aa:bb:cc' def testxml(test_dom="domU1", mem = 128, vcpus = 1, mac = default_mac, disk_file_path = disk_path, disk = "xvda"): -- 1.8.1.4

1 0

Re: [Libvirt-cim] [PATCH 1/9] Need to check "slp=true", not just "slp" since "slp=false" is possible
by WangXu 24 Apr '13

24 Apr '13

Reviewed-by: Xu Wang <cngesaint(a)gmail.com> > .../libvirt-cim/cimtest/Profile/04_verify_libvirt_cim_slp_profiles.py | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/suites/libvirt-cim/cimtest/Profile/04_verify_libvirt_cim_slp_profiles.py b/suites/libvirt-cim/cimtest/Profile/04_verify_libvirt_cim_slp_profiles.py > index f6cf5b1..f08d157 100644 > --- a/suites/libvirt-cim/cimtest/Profile/04_verify_libvirt_cim_slp_profiles.py > +++ b/suites/libvirt-cim/cimtest/Profile/04_verify_libvirt_cim_slp_profiles.py > @@ -64,7 +64,7 @@ def get_slp_info(server): > logger.info("Test not supported for sfcb yet ... hence skipping") > return SKIP > > - cmd = "cimconfig -l -p | grep slp" > + cmd = "cimconfig -l -p | grep slp=true" > rc, out = run_remote(server, cmd) > if rc != 0: > logger.error("SLP is not enabled for the cimserver on '%s'", server) > -- > 1.8.1.4

1 0

Release of libvirt-cim-0.6.2
by Daniel Veillard 15 Apr '13

15 Apr '13

Based on feedback last week, I also release libvirt-cim-0.6.2 it is tagged in git and pushed on the ftp: ftp://libvirt.org/libvirt-cim/ It includes the recent set of patches made by Wenchao Xia and John Ferlan plus of a few older patches from last year: libxkutil: set domain->dype to unknown by default (Wenchao Xia) register: Adjust the chatter output (John Ferlan) DevicePool: Use the virConnectListAll interfaces (John Ferlan) libxkutil: Use virConnectListAllDomains() to fetch domains (John Ferlan) xmlgen: Only support script on bridge for xen domains (John Ferlan) Remove empty newline at bottom (John Ferlan) libvirt-cim.spec: Use systemctl for tog-pegasus restart (John Ferlan) Makefile.am: Remove the $(top_srcdir) from subst command (John Ferlan) migration: allow ssh based migration with non root's key file (Wenchao Xia) misc_util: better way to read config (Wenchao Xia) VSSD: report success if not all VS fail in enum (Wenchao Xia) CSI: Move native CSI code together (Wenchao Xia) CSI Discard libvirt event by default (Wenchao Xia) device parsing, add debug print (Wenchao Xia) DevicePool, reimplement get_diskpool_config with libvirt (Wenchao Xia) CSI, add lock to protect shared data in lifecycle_thread (Wenchao Xia) CSI, DevicePool, RASDIndication: fix debug print crash (Wenchao Xia) do not deregister virt classes in yum upgrade (Wenchao Xia) SDC: use property BootDevices instead of BootDevice (Wenchao Xia) SDC: Fix spelling for property IsFullVirt (Wenchao Xia) Remove property CreationClassName in some instance (Wenchao Xia) fix id parsing with white space (Wenchao Xia) Fix an incomplete substitution in Makefile (Klaus Kaempf) distinguish running or inactive state (Wenchao Xia) Fix possible memory leaks (Eduardo Lima (Etrunko)) Pool Refresh Patch2 (Gareth S. Bestor) CSI: Only execute callback if indications are enabled (Eduardo Lima (Etrunko)) CSI: Fix log messages (Eduardo Lima (Etrunko)) CSI: Cleanup code (Eduardo Lima (Etrunko)) Plan9fs (aka 9pfs, VirtFS) support for QEMU/KVM environment (Deepak C Shetty) I also pushed an uncontroversial spec file update based on the Fedora spec file, but it is not included in the release in case some adjustments are needed, thanks everybody for your contributions ! Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

Release of libcmpiutil-0.5.7
by Daniel Veillard 15 Apr '13

15 Apr '13

I pushed on the FTP server and tagged in git that release it can be found at: ftp://libvirt.org/libvirt-cim/ It includes only the couple of patches from Wenchao Xia about log fix and improvement. I also pushed on git but after the release a patch which was applied to Fedora to allow compilation on arm, it was uncontroversial, Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[PATCH 0/3] cimtest follow patch
by Wenchao Xia 12 Apr '13

12 Apr '13

This is a patch follows John's 9 patches for cimtest, after it cimtest should only fail 3 case on RH6.4: HostSystem - 01_enum.py: FAIL HostSystem - 03_hs_to_settdefcap.py: FAIL VirtualSystemManagementService - 19_definenetwork_ers.py: FAIL This patch is only for review and test, it may need adjust and merge with John's patch, and change to author name(not root :|), please do not push directly. root (3): test: common_util, use number to check version test: rasd use int as comparation condtion for libvirt version test: RPCS fix nfs issue .../12_create_netfs_storagevolume_errs.py | 2 +- suites/libvirt-cim/lib/XenKvmLib/common_util.py | 32 ++++++++++++++++--- suites/libvirt-cim/lib/XenKvmLib/pool.py | 8 ++-- suites/libvirt-cim/lib/XenKvmLib/rasd.py | 7 ++-- 4 files changed, 36 insertions(+), 13 deletions(-)

2 13

Re: [Libvirt-cim] [PATCH] SwitchService: make command configurable
by WangXu 11 Apr '13

11 Apr '13

Dear John, What's your opinion about my solution? If no problem, we'll develop patch as described. ver. 0.6.2 has to be released as soon as possible:-) Sincerely yours, Xu Wang From: cngesaint(a)outlook.com To: libvirt-cim(a)redhat.com Date: Wed, 10 Apr 2013 10:20:34 +0000 Subject: Re: [Libvirt-cim] [PATCH] SwitchService: make command configurable I think the "vsi_search_string be an array" problem could be solved as following steps: 1. An item such as (vsi_support_key_string="{supported forwarding mode: (0x40) reflective relay, supported capabilities: (0x7) RTE ECP VDP}";) is configured well in the .conf file. 2. The application read this value from .conf file as a long string. 3. The application executes supporting function about VSI and check whether the output returned is a substring of vsi_support_key_string. If the output contains in the vsi_support_key_string, supporting of VSI is proved. Sincerely yours, Xu Wang >On 04/09/2013 06:40 AM, Wenchao Xia wrote: >> Hi, John >> Sorry to interrupt but still need you to review this patch, which >>make it work more flex on different version. >> By the way, I got only 3 fail and they seems small problems in test >> suit's code, do you think it is ready to release 0.6.2(with this patch >> pushed)? >> I don't have 'lldptool' installed on my system, so I cannot verify the output or the fix. > >What concerns me though is the reliance on something that could (and does) change between versions. >What you have is just "two" versions of differences. What if there's a 3rd version that had/has >something different? Is the output different on different archs/OS's? > >Is there no way to make the vsi_search_string be an array in the configuration file? Rather than two >constant strings. > >I suppose what you have works, but it seems there has to be a better way to do this. I also think the >comments in the .conf file could be beefed up so someone would know what to look for. Is that a >specific field in the output or is that string what is returned. Essentially how would someone know >what to look for in order to add a new/different string. > >Call this a "luke-warm" ACK. It works, but I think it's a shortcut. > >John > > > >>> In RH6.4 lldptool query command and output changed a bit, ...... _______________________________________________ Libvirt-cim mailing list Libvirt-cim(a)redhat.com https://www.redhat.com/mailman/listinfo/libvirt-cim

2 1

Tyrel Datwyler is out of the office.
by Tyrel Datwyler 11 Apr '13

11 Apr '13

I will be out of the office starting 04/11/2013 and will not return until 04/15/2013. During this time I will not have access to email.

1 0

Re: [Libvirt-cim] [PATCH] SwitchService: make command configurable
by WangXu 10 Apr '13

10 Apr '13

I think the "vsi_search_string be an array" problem could be solved as following steps: 1. An item such as (vsi_support_key_string="{supported forwarding mode: (0x40) reflective relay, supported capabilities: (0x7) RTE ECP VDP}";) is configured well in the .conf file. 2. The application read this value from .conf file as a long string. 3. The application executes supporting function about VSI and check whether the output returned is a substring of vsi_support_key_string. If the output contains in the vsi_support_key_string, supporting of VSI is proved. Sincerely yours, Xu Wang >On 04/09/2013 06:40 AM, Wenchao Xia wrote: >> Hi, John >> Sorry to interrupt but still need you to review this patch, which >>make it work more flex on different version. >> By the way, I got only 3 fail and they seems small problems in test >> suit's code, do you think it is ready to release 0.6.2(with this patch >> pushed)? >> I don't have 'lldptool' installed on my system, so I cannot verify the output or the fix. > >What concerns me though is the reliance on something that could (and does) change between versions. >What you have is just "two" versions of differences. What if there's a 3rd version that had/has >something different? Is the output different on different archs/OS's? > >Is there no way to make the vsi_search_string be an array in the configuration file? Rather than two >constant strings. > >I suppose what you have works, but it seems there has to be a better way to do this. I also think the >comments in the .conf file could be beefed up so someone would know what to look for. Is that a >specific field in the output or is that string what is returned. Essentially how would someone know >what to look for in order to add a new/different string. > >Call this a "luke-warm" ACK. It works, but I think it's a shortcut. > >John > > > >>> In RH6.4 lldptool query command and output changed a bit, ......

1 0

[PATCH 0/2] Add version comparison method for virsh_version output
by John Ferlan 10 Apr '13

10 Apr '13

I believe this set of patches should be applied instead of those from https://www.redhat.com/archives/libvirt-cim/2013-April/msg00006.html I suppose we could consider combining the two functions; however, there are a couple place where the return version string is compared more than once, so I kept both interfaces. The vxml change is just an informative message that showed me the name of the image that was failing - I had created 'default-kvm-dimage.img' and tests were failing to find 'default-kvm-dimage' to place into the cimtest storage pool. John Ferlan (2): vxml: Add which volume could not be found to error message Add and utilize virsh_version_cmp .../08_CreateDiskResourcePool.py | 4 ++-- .../10_create_storagevolume.py | 6 ++++-- .../11_create_dir_storagevolume_errs.py | 5 +++-- .../12_create_netfs_storagevolume_errs.py | 5 +++-- .../13_delete_storagevolume.py | 5 +++-- .../14_delete_storagevolume_errs.py | 5 +++-- .../ResourcePoolConfigurationService/15_DiskPoolAutostart.py | 4 ++-- .../cimtest/SettingsDefineCapabilities/01_forward.py | 1 - .../VirtualSystemManagementService/19_definenetwork_ers.py | 8 ++++---- suites/libvirt-cim/lib/XenKvmLib/common_util.py | 9 +++++---- suites/libvirt-cim/lib/XenKvmLib/pool.py | 6 +++--- suites/libvirt-cim/lib/XenKvmLib/rasd.py | 10 ++++++---- suites/libvirt-cim/lib/XenKvmLib/vxml.py | 4 ++-- suites/libvirt-cim/lib/XenKvmLib/xm_virt_util.py | 6 ++++++ 14 files changed, 46 insertions(+), 32 deletions(-) -- 1.8.1.4

2 3

[PATCH] SwitchService: make command configurable
by Wenchao Xia 10 Apr '13

10 Apr '13

In RH6.4 lldptool query command and output changed a bit, to make it work on different version, the string used in search and options used in lldptool command should be set in configuration file. Signed-off-by: Wenchao Xia <xiawenc(a)linux.vnet.ibm.com> --- libvirt-cim.conf | 12 ++++++++++++ libxkutil/misc_util.c | 27 +++++++++++++++++++++++++++ libxkutil/misc_util.h | 4 +++- src/Virt_SwitchService.c | 31 ++++++++++++++++++++++++------- 4 files changed, 66 insertions(+), 8 deletions(-) diff --git a/libvirt-cim.conf b/libvirt-cim.conf index 37d7b0f..40fc94a 100644 --- a/libvirt-cim.conf +++ b/libvirt-cim.conf @@ -30,3 +30,15 @@ # Default value: NULL, that is not set. # # migrate_ssh_temp_key = "/root/vm_migrate_tmp_id_rsa"; + +# lldptool_query_options (string) +# Defines the command used in SwitchService to query VEPA support, will be +# used as "lldptool -i [INTERFACE] [OPTIONS]" +# +# lldptool_query_options = "-t -g ncb -V evbcfg"; + +# vsi_search_string (string) +# Defines the string used in SwitchService to search in lldptool's output +# +# vsi_search_string0 = "supported forwarding mode: (0x40) reflective relay"; +# vsi_search_string1 = "supported capabilities: (0x7) RTE ECP VDP"; diff --git a/libxkutil/misc_util.c b/libxkutil/misc_util.c index 00eb4b1..1a9a1b4 100644 --- a/libxkutil/misc_util.c +++ b/libxkutil/misc_util.c @@ -236,6 +236,33 @@ const char *get_mig_ssh_tmp_key(void) return prop.value_string; } +const char *get_lldptool_query_options(void) +{ + static LibvirtcimConfigProperty prop = { + "lldptool_query_options", CONFIG_STRING, {0}, 0}; + + libvirt_cim_config_get(&prop); + return prop.value_string; +} + +const char *get_vsi_search_string0(void) +{ + static LibvirtcimConfigProperty prop = { + "vsi_search_string0", CONFIG_STRING, {0}, 0}; + + libvirt_cim_config_get(&prop); + return prop.value_string; +} + +const char *get_vsi_search_string1(void) +{ + static LibvirtcimConfigProperty prop = { + "vsi_search_string1", CONFIG_STRING, {0}, 0}; + + libvirt_cim_config_get(&prop); + return prop.value_string; +} + virConnectPtr connect_by_classname(const CMPIBroker *broker, const char *classname, CMPIStatus *s) diff --git a/libxkutil/misc_util.h b/libxkutil/misc_util.h index 0f52290..06a37cc 100644 --- a/libxkutil/misc_util.h +++ b/libxkutil/misc_util.h @@ -154,7 +154,9 @@ int virt_set_status(const CMPIBroker *broker, /* get libvirt-cim config */ const char *get_mig_ssh_tmp_key(void); - +const char *get_lldptool_query_options(void); +const char *get_vsi_search_string0(void); +const char *get_vsi_search_string1(void); /* * Local Variables: * mode: C diff --git a/src/Virt_SwitchService.c b/src/Virt_SwitchService.c index 8991426..103f125 100644 --- a/src/Virt_SwitchService.c +++ b/src/Virt_SwitchService.c @@ -46,13 +46,24 @@ static CMPIStatus check_vsi_support(char *command) CMPIStatus s = {CMPI_RC_OK, NULL}; char buff[MAX_LEN]; FILE *stream = NULL; - const char *searchStr[] = {" supported forwarding mode: " - "(0x40) reflective relay", - " supported capabilities: " - "(0x07) RTE ECP VDP", - NULL}; + const char *string0 = get_vsi_search_string0(); + const char *string1 = get_vsi_search_string1(); + const char *searchStr[3]; int matched = 0; + if (!string0) { + string0 = " supported forwarding mode: " + "(0x40) reflective relay"; + } + if (!string1) { + string1 = " supported capabilities: " + "(0x7) RTE ECP VDP"; + } + CU_DEBUG("vsi searching for string [%s] [%s]", string0, string1); + searchStr[0] = string0; + searchStr[1] = string1; + searchStr[2] = NULL; + // Run lldptool command to find vsi support. stream = popen(command, "r"); if (stream == NULL) { @@ -214,6 +225,7 @@ static CMPIStatus get_switchservice(const CMPIObjectPath *reference, int i; char **if_list; char cmd[MAX_LEN]; + const char *lldptool_query_options = NULL; *_inst = NULL; conn = connect_by_classname(broker, CLASSNAME(reference), &s); @@ -257,10 +269,15 @@ static CMPIStatus get_switchservice(const CMPIObjectPath *reference, CU_DEBUG("Found %d interfaces", count); + lldptool_query_options = get_lldptool_query_options(); + if (!lldptool_query_options) { + lldptool_query_options = "-t -g ncb -V evbcfg"; + } for (i=0; i<count; i++) { - sprintf(cmd, "lldptool -i %s -t -V evbcfg", if_list[i]); - CU_DEBUG("running command %s ...", cmd); + sprintf(cmd, "lldptool -i %s %s", + if_list[i], lldptool_query_options); + CU_DEBUG("running command [%s]", cmd); s = check_vsi_support(cmd); if (s.rc == CMPI_RC_OK) { vsi = true; -- 1.7.1

3 5

[PATCH V6 00/20] Bug fix patches for 0.6.2
by Wenchao Xia 08 Apr '13

08 Apr '13

This serial fix a batch of issues. v6: general change: rebase and add John Ferlan's patches after the tail. 2/20: do not add this property now, just fixed the spelling in code, since it is only set when Xen is used in libvirt-cim, so it is OK. 3/20: also fixed debug print spelling. 5/20: squashed patch for debug print crash issues. 6/20: spelling fix in commit message. 7/20: skipped free_diskpool on fail of setting parent pool, refined the code in get_diskpool_config() when libvirt pool was not used, removed the duplicated macro in .c file, set member to NULL in parse_diskpool_line when fail. 8/20: squashed the patch with John Ferlan's one about "adjust logic" for it. 11/20: free the domain ptr after domain name retrieving. 13/20: rename no_root_ssh_key to use_non_root_ssh_key, fail when file can't be deleted before copy in ssh_key_copy(). 17/20: small code style change for {} in if condition. 18/20: added a macro to activate it. 19/20: added three macro to activate them, added {} in if statement when only one sentence follow. John Ferlan (7): 14 Makefile.am: Remove the $(top_srcdir) from subst command 15 libvirt-cim.spec: Use systemctl for tog-pegasus restart 16 Remove empty newline at bottom 17 xmlgen: Only support script on bridge for xen domains 18 libxkutil: Use virConnectListAllDomains() to fetch domains 19 DevicePool: Use the virConnectListAll interfaces 20 register: Adjust the chatter output Wenchao Xia (13): 1 Remove property CreationClassName in some instance 2 SDC: Fix spelling for property IsFullVirt 3 SDC: use property BootDevices instead of BootDevice 4 do not deregister virt classes in yum upgrade 5 CSI, DevicePool, RASDIndication: fix debug print crash 6 CSI, add lock to protect shared data in lifecycle_thread 7 DevicePool, reimplement get_diskpool_config with libvirt 8 device parsing, add debug print 9 CSI Discard libvirt event by default 10 CSI: Move native CSI code together 11 VSSD: report success if not all VS fail in enum 12 misc_util: better way to read config 13 migration: allow ssh based migration with non root's key file Makefile.am | 18 +- libvirt-cim.conf | 19 + libvirt-cim.spec.in | 24 +- libxkutil/cs_util_instance.c | 26 + libxkutil/device_parsing.c | 19 +- libxkutil/misc_util.c | 135 +++- libxkutil/misc_util.h | 6 +- libxkutil/xmlgen.c | 26 +- provider-register.sh | 9 +- schema/SwitchService.registration | 1 - src/Virt_AllocationCapabilities.c | 3 +- src/Virt_ComputerSystem.c | 66 ++- src/Virt_ComputerSystemIndication.c | 1018 +++++++++++++++++--- src/Virt_ConsoleRedirectionService.c | 3 +- src/Virt_ConsoleRedirectionServiceCapabilities.c | 3 +- src/Virt_Device.c | 18 +- src/Virt_DevicePool.c | 480 ++++++++-- src/Virt_DevicePool.h | 4 + src/Virt_EnabledLogicalElementCapabilities.c | 3 +- src/Virt_FilterEntry.c | 3 +- src/Virt_FilterList.c | 3 +- src/Virt_HostSystem.c | 3 +- src/Virt_KVMRedirectionSAP.c | 3 +- src/Virt_RASD.c | 3 +- src/Virt_ReferencedProfile.c | 3 +- src/Virt_RegisteredProfile.c | 3 +- src/Virt_ResourceAllocationSettingDataIndication.c | 6 +- src/Virt_ResourcePoolConfigurationCapabilities.c | 3 +- src/Virt_ResourcePoolConfigurationService.c | 8 +- src/Virt_SettingsDefineCapabilities.c | 14 +- src/Virt_SwitchService.c | 3 +- src/Virt_VSMigrationCapabilities.c | 3 +- src/Virt_VSMigrationService.c | 270 +++++- src/Virt_VSMigrationSettingData.c | 3 +- src/Virt_VSSD.c | 50 +- src/Virt_VirtualSystemManagementCapabilities.c | 3 +- src/Virt_VirtualSystemManagementService.c | 67 ++- src/Virt_VirtualSystemSnapshotService.c | 3 +- ...Virt_VirtualSystemSnapshotServiceCapabilities.c | 3 +- 39 files changed, 1990 insertions(+), 348 deletions(-)

4 30

[PATCH V3 0/2] libcmpiutil: patch rebase for libvirt-cim 6.2
by Wenchao Xia 03 Apr '13

03 Apr '13

Hi, Danial Sorry for have sent confusing patches, please use this serial for libcmpiutil. Wenchao Xia (2): libcmpiutil: fix potential debug print crash libcmpiutil: add time and thread info in debug log debug_util.c | 19 ++++++++++++++++++- std_indication.c | 14 ++++++++++++-- 2 files changed, 30 insertions(+), 3 deletions(-)

2 3