March 2013 - CIM - libvirt List Archives

[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

[PATCH V5 00/15] Bug fix patches for 0.6.2
by Wenchao Xia 25 Mar '13

25 Mar '13

This serial fix a serial of issues. V5: General change: Remove script change patch, need John Ferlan's patch in following. 1/15-3/15: new patch make libvirt-cim conform to DSP more. 6/15: better commit message. 8/15: remove a strdup() to reduce a error path. 10/15: free *xml after a fail. 11/15: return fail when ind == NULL in trigger_mod_indiction(), check return value for calloc() in doms_to_xml(), better debug message, check strdup() return on raise_indication, better traverse in dom list. Reduce condition macros. 12/15: new patch to move CSI-libvirt/CSI-libvirt-cim together. 13/15: remove defensive code, only fix the problem of report fail when some VSSD succeed. 14/15: new separate patch from 15/15, to enhance config reading. 15/15: changed the interface, new method to copy/delete the key, rm the key before copy and check the key's status after copy. Wenchao Xia (15): 1 Remove property CreationClassName in some instance 2 VSSD: add missing property IsFullVirt in schema 3 SDC: use property BootDevices instead of BootDevice 4 do not deregister virt classes in yum upgrade 5 CSI, fix debug print crash 6 CSI, add lock to protect shared data in lifecycle_thread 7 DevicePool, fix debug print crash 8 DevicePool, reimplement get_diskpool_config with libvirt 9 RASDIndication, fix debug print crash 10 device parsing, add debug print 11 CSI Discard libvirt event by default 12 CSI: Move native CSI code together 13 VSSD: report success if not all VS fail in enum 14 misc_util: better way to read config 15 migration: allow ssh based migration with non root's key file libvirt-cim.conf | 19 + libvirt-cim.spec.in | 12 +- libxkutil/device_parsing.c | 16 +- libxkutil/misc_util.c | 135 +++- libxkutil/misc_util.h | 6 +- schema/Virt_VSSD.mof | 3 + 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 | 198 +++-- 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 | 10 +- src/Virt_SwitchService.c | 3 +- src/Virt_VSMigrationCapabilities.c | 3 +- src/Virt_VSMigrationService.c | 269 +++++- src/Virt_VSMigrationSettingData.c | 3 +- src/Virt_VSSD.c | 46 +- src/Virt_VirtualSystemManagementCapabilities.c | 3 +- src/Virt_VirtualSystemManagementService.c | 63 ++- src/Virt_VirtualSystemSnapshotService.c | 3 +- ...Virt_VirtualSystemSnapshotServiceCapabilities.c | 3 +- 34 files changed, 1652 insertions(+), 295 deletions(-)

2 47

[PATCHv2 0/8] Additional patches for 0.6.2
by John Ferlan 25 Mar '13

25 Mar '13

This series replaces my last set. It is a followup after rebase and rework to: https://www.redhat.com/archives/libvirt-cim/2013-March/msg00029.html Changes in v2: * Rework libxkutil/cs_util_instance.c to mirror functions as requested * Rework src/Virt_DevicePool.c to mirror functions as requested * Just merged the Makefile.am changes into one * Previous patch 4/10 to CSI is removed due to code change already posted Still to be reviewed is Makefile.am. John Ferlan (8): Makefile.am: Remove the $(top_srcdir) from subst command libvirt-cim.spec: Use systemctl for tog-pegasus restart Remove empty newline at bottom xmlgen: Only support script on bridge for xen domains libxkutil: Use virConnectListAllDomains() to fetch domains libxkutil: Adjust get_dominfo() logic DevicePool: Use the virConnectListAll interfaces register: Adjust the chatter output Makefile.am | 18 +-- libvirt-cim.spec.in | 12 +- libxkutil/cs_util_instance.c | 24 ++++ libxkutil/device_parsing.c | 7 +- libxkutil/xmlgen.c | 25 +++-- provider-register.sh | 9 +- schema/SwitchService.registration | 1 - src/Virt_DevicePool.c | 230 ++++++++++++++++++++++++++++++++++++++ 8 files changed, 296 insertions(+), 30 deletions(-) -- 1.8.1.4

2 10

Re: [Libvirt-cim] [PATCH V4] DevicePool, reimplement get_diskpool_config with libvirt
by John Ferlan 19 Mar '13

19 Mar '13

Since I don't have an email to reply-to, here is the link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00039.html In get_diskpool_config(): * Rather than use the "racy" NumOfStoragePools and ListStoragePools, why not use virConnectListAllStoragePools() passing the "Active" flag for all active pools only? * You may even want to consider keeping the returned virStoragePoolPtr structures around.. I also imagine the other objects (networks and domains) could use the similar calls. I guess the answer somewhat depends on what is the minimum version of libvirt that needs to be supported. But if you "have" to stay with the current model... * The return 'names[i]' is something you'd have to free() anyway, so rather than strdup(names[i]), just take it when setting pools[i].tag and set names[i] = NULL; That avoids an error path. John

2 2

Re: [Libvirt-cim] [PATCH V3 04/11] CSI, add lock to protect shared data
by John Ferlan 19 Mar '13

19 Mar '13

Since I don't have the email to reply-to, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00026.html Why was the decision to reuse the lifecycle_mutex rather than generating a data specific mutex? Seems like that mutex is used for other operations to protect filters and indications. Thus unless the two are somehow tied, I really think it'd be better to create/use a data specific mutex. Furthermore, there's a few mutex adds around thread mgmt that probably should be a separate patch. That is - the non "dom_list" protection additions appear to be different. John

2 2

[PATCH 0/9] cimtest changes
by John Ferlan 18 Mar '13

18 Mar '13

Here's my set of cimtest changes. Quite a few of the changes deal with the illegal MAC address that starts with 99-* - this is a multicast MAC which is not allowed. There are changes to handle systemd/nfs, a couple dealing with fedora vs. rhel default network device name, and a couple of fixes that I just dug into to resolve on my f18 box. I get the following results: ================================================= FAIL : 4 XFAIL : 3 SKIP : 11 PASS : 174 ----------------- Total : 192 ================================================= FAIL Test Summary: ComputerSystemIndication - 01_created_indication.py: FAIL RASDIndications - 01_guest_states_rasd_ind.py: FAIL RASDIndications - 02_guest_add_mod_rem_rasd_ind.py: FAIL VirtualSystemManagementService - 22_addmulti_brg_interface.py: FAIL ================================================= XFAIL Test Summary: SwitchService - 01_enum.py: XFAIL VirtualSystemManagementService - 28_definesystem_with_vsi_profile.py: XFAIL VirtualSystemManagementService - 30_dynamic_disk_mod.py: XFAIL ================================================= SKIP Test Summary: ComputerSystem - 02_nosystems.py: SKIP ComputerSystemMigrationJobIndication - 01_csmig_ind_for_offline_mig.py: SKIP LogicalDisk - 02_nodevs.py: SKIP Profile - 04_verify_libvirt_cim_slp_profiles.py: SKIP VirtualSystemMigrationService - 01_migratable_host.py: SKIP VirtualSystemMigrationService - 02_host_migrate_type.py: SKIP VirtualSystemMigrationService - 05_migratable_host_errs.py: SKIP VirtualSystemMigrationService - 06_remote_live_migration.py: SKIP VirtualSystemMigrationService - 07_remote_offline_migration.py: SKIP VirtualSystemMigrationService - 08_remote_restart_resume_migration.py: SKIP VSSD - 02_bootldr.py: SKIP John Ferlan (9): 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 .../Profile/04_verify_libvirt_cim_slp_profiles.py | 2 +- .../08_CreateDiskResourcePool.py | 26 ++++++-- .../09_DeleteDiskPool.py | 19 ++++-- .../15_DiskPoolAutostart.py | 11 +++- .../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 | 17 +++-- .../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 | 77 +++++++++++++++++----- suites/libvirt-cim/lib/XenKvmLib/const.py | 6 +- suites/libvirt-cim/lib/XenKvmLib/test_xml.py | 2 +- suites/libvirt-cim/main.py | 15 +++++ 18 files changed, 183 insertions(+), 53 deletions(-) -- 1.8.1.4

2 10

[PATCH 00/10] Adjustments based on recent review
by John Ferlan 15 Mar '13

15 Mar '13

As requested here is a series of patches to address a couple of issues I discovered during my initial foray into using libvirt-cim as well as from my code review. These are all *based on* the code I reviewed. Feel free to pick, choose, and apply. I have run them on my f18 system and things seem to work, but it's possible I've made a configuration error. I also will have a series of cimtest fixes, but I need to tidy them up a bit before sending. I did figure out that rather than uninstalling and reinstalling the package, I can use make preuninstall, make preinstall, restart the CIMOM, make install, and make postinsall which seems to replace the mofs "mostly correctly". There is an error in the postinstall phase which I haven't had the cycles to chase yet. John Ferlan (10): Remove empty newline at bottom Makefile.am: Remove the $(top_srcdir) from subst command xmlgen: Only support script on bridge for xen domains CSI: Fix bug found during 'make distcheck' Makefile.am: Use the top_srcdir rather than direct path in subst libxkutil: Use virConnectListAllDomains() to fetch domains libvirt-cim.spec: Use systemctl for tog-pegasus restart libxkutil: Adjust get_dominfo() logic DevicePool: Use the virConnectListAll interfaces register: Adjust the chatter output Makefile.am | 18 +++--- libvirt-cim.spec.in | 10 ++- libxkutil/cs_util_instance.c | 23 +++++++ libxkutil/device_parsing.c | 10 +-- libxkutil/xmlgen.c | 25 +++++--- provider-register.sh | 9 +-- schema/SwitchService.registration | 1 - src/Virt_ComputerSystemIndication.c | 4 ++ src/Virt_DevicePool.c | 121 +++++++++++++++++++++++++++++++++++- 9 files changed, 190 insertions(+), 31 deletions(-) -- 1.8.1.4

2 21

Tyrel Datwyler is out of the office.
by Tyrel Datwyler 15 Mar '13

15 Mar '13

I will be out of the office starting 03/13/2013 and will not return until 03/18/2013. During this time I will not have access to email. In the case of any urgent issues please contact my back up David Heller.

1 0

Re: [Libvirt-cim] [PATCH V3 11/11] allow ssh based migration with non root's key file
by John Ferlan 13 Mar '13

13 Mar '13

Since I don't have the original email to reply-to, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00033.html libvirt-cim.conf Consider changing the following: +# Defines a temp key file which would be used as ssh key in ssh migration, +# Only valid when SSH_Key is set in migration call. +# If SSH_Key is not in a directory owned by root, set this value to a path +# owned by root, tells libvirt-cim copy it there before use it. The directory +# in the path must exist and totally owned by root. +# If SSH_Key is already in a valid place, don't set it to to tell libvirt-cim +# use SSH_Key directly. # Only valid when SSH_Key is set for the migration call. # # In order to allow using a non-root owned SSH Key during migration, set this variable # to a location to copy the SSH_Key from the SSH_Key path and file not owned by root # to a directory path and file owned by root. The target path must exist and it must # be owned by root. If it is not, the migration will fail. # # If the SSH_Key is already in a directory owned by root there is no need to set, thus # allowing libvirt-cim to use the SSH_Key directly. libxkutil/misc_util.c In libvirt_cim_config_get() * The 'default:' case should probably also which prop->name was being used for the prop->value_type In is_read_only() & get_mig_ssh_tmp_key() * Probably should initialize prop.value_string = NULL just to be clear * Call to libvirt_cim_config_get() does not check return status - perhaps make setting prop.have_read conditional on return value... src/Virt_VSMigrationService.c In ssh_key_cp(): * Do you need a path to 'cp' (eg /sbin/cp)? * Is it worth a stat() afterwards to ensure the copy occurred since your fgets() isn't returning anything? * Other thoughts - should there be a "stat" and "rm" before? Perhaps another safety ensure we copy something? In get_msd_values() * Since you already print the source ssh_key file, change the the copy src to dest message to just indicate the destination (e.g. "Copying ssh key to '%s'.", tmp_keyfile) So let's say this is successful, now we leave the 'migrate_ssh_temp_key' around? Is that a good idea? Shouldn't there be a corresponding 'unlink()' after we're done? We're really not a "temporary" file if we keep it around. In fact, after the connection is made, shouldn't we be able to safely delete the file? John Finally, this one seems to require/need an update to documentation "somewhere" to describe that new conf variable... That is how does one know how to use this unless they read the conf file?

2 2

Re: [Libvirt-cim] [PATCH V3 00/11] bug fix and enhancement patch rebase for 6.2
by John Ferlan 12 Mar '13

12 Mar '13

Now that I've gone through the series - hopefully that gives you enough to put out another series. Once we get through it, then we can work with DV in order to go through the next steps. I'll look at the libcmpiutil changes next. John

2 1

Re: [Libvirt-cim] [PATCH V4] VSSD, fix a error report issue in VSSD enum
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00038.html In instance_from_dom(): * virDomainIsActive returns: Returns 1 if running, 0 if inactive, -1 on error. You are testing for < 0 (error condition) and an else with no determination about whether it's really active or not. Your comments in the code and the CU_DEBUG statements don't seem to sync. How is VSSD_RECOVERABLE true when active = -1? NOTE: In looking at get_dominfo() again - and thinking about this as a caller - why does it matter if autostart is set or not? I really think the "get_domain_list()" should change to use virConnectListAllDomains() - it may solve more timing issues. John

2 1

Re: [Libvirt-cim] [PATCH V3 09/11] CSI Discard libvirt event by default
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to, here is the link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00032.html Seems this was "pulled back" from some other commit - might be nice to reference the commit id where it was removed and now replaced - at least for historical purposes. src/Virt_ComputerSystem.c In trigger_mod_indication() * The "goto out" if/when "(ind == NULL)" will perhaps trigger a false positive this succeeded since "s" is initialized to CMPI_RC_OK. src/Virt_ComputerSystemIndication.c In doms_to_xml(), * the calloc() has no check for successful return of memory leading to possible NULL pointer deref of *dom_xml_list. In async_ind_native(): * The CU_DEBUG() message "...indication diliv..." should be "...indication delivery..." * The CU_DEBUG("Unrecognized indication type"); should report what 'ind_type' is (btw: it's unrecognised) In lifecycle_thread_native(): * Seems inefficient to traverse lists so many times. In the for loop running through prev_xml, it would seem that if 'dom_in_list()' returns 1, then we could call 'dom_changed()' and possibly 'async_ind_native()' if it has changed. * Since this routine keeps lifecycle_mutex for the duration - I believe it makes it even more important to separate it from 'dom_list' in your previous change... NOTE: the get_domain_list() call could be made better by getting all "Active and Defined" domains in one virConnectListAllDomains() call, but that's mostly unrelated to this code. Getting the list of All domains at one time is "preferred" since you won't run into odd timing issues by making two calls. In raise_indication() * The returns for the strdup()'s are not checked. John

2 1

Re: [Libvirt-cim] [PATCH V3 08/11] device parsing, add debug print
by John Ferlan 12 Mar '13

12 Mar '13

Since I didn't have the original email to reply-to, here is the link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00029.html Should the following return 0 as well?: + if (ret != 1) { + CU_DEBUG("Failed to translate xml into struct domain"); + } Especially since get_dominfo_from_xml() will free and initialize dominfo to NULL. You'll also need a "free(xml);" if you return. That's a memory leak. What probably should be done is (btw: 8 space tabs are hard to do!) ret = get_dominfo_from_xml(xml, dominfo); if (ret == 0) { CU_DEBUG("Failed to translate xml into struct domain"); goto out; } ret = virDomainGetAutostart(dom, &start); if (ret != 0) { CU_DEBUG("Failed to get dom autostart with libvirt API."); goto out; } (*dominfo)->autostrt = start; out: free(xml); return ret; John

2 1

Re: [Libvirt-cim] [PATCH V3 01/11] remove script for bridge network
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to for a code review, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00030.html NACK As previously noted, a script is supported (and perhaps required) in the Xen guest. Consider the following, which adds a parameter to the call in order to allow for Xen: -static const char *bridge_net_to_xml(xmlNodePtr nic, struct net_device *dev) +static const char *bridge_net_to_xml(xmlNodePtr nic, struct net_device *dev, + int domtype) { const char *script = "vif-bridge"; xmlNodePtr tmp; const char *msg = NULL; - tmp = xmlNewChild(nic, NULL, BAD_CAST "script", NULL); - if (tmp == NULL) - return XML_ERROR; - xmlNewProp(tmp, BAD_CAST "path", BAD_CAST script); + /* Scripts only supported on Xen guests see 'libvirt' + * commit id 1734cdb99 (since 0.9.10) */ + if (domtype == DOMAIN_XENPV || domtype == DOMAIN_XENFV) { + tmp = xmlNewChild(nic, NULL, BAD_CAST "script", NULL); + if (tmp == NULL) + return XML_ERROR; + xmlNewProp(tmp, BAD_CAST "path", BAD_CAST script); + } msg = set_net_source(nic, dev, "bridge"); @@ -375,13 +380,13 @@ static const char *net_xml(xmlNodePtr root, struct domain } #endif - if (STREQ(dev->dev.net.type, "network")) + if (STREQ(dev->dev.net.type, "network")) { msg = set_net_source(nic, net, "network"); - else if (STREQ(dev->dev.net.type, "bridge")) - msg = bridge_net_to_xml(nic, net); - else if (STREQ(dev->dev.net.type, "user")) + } else if (STREQ(dev->dev.net.type, "bridge")) { + msg = bridge_net_to_xml(nic, net, dominfo->type); + } else if (STREQ(dev->dev.net.type, "user")) { continue; - else if (STREQ(dev->dev.net.type, "direct")) { + } else if (STREQ(dev->dev.net.type, "direct")) { msg = set_net_source(nic, net, "direct"); if (net->vsi.vsi_type != NULL) { struct vsi_device *vsi = &dev->dev.net.vsi;

2 1

Re: [Libvirt-cim] [PATCH V3 2/2] libcmpiutil: add time and thread info in debug log
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to, here's the link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00020.html ACK

1 0

Re: [Libvirt-cim] [PATCH V3 1/2] libcmpiutil: fix potential debug print crash
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to, here's a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00019.html ACK John

1 0

Re: [Libvirt-cim] [PATCH V3 10/11] VSSD, fix a error report issue in VSSD enum
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00031.html NACK since this has been replaced by: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00038.html Which I will review separately. John

1 0

Re: [Libvirt-cim] [PATCH V3 07/11] RASDIndication, fix debug print crash
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00027.html ACK Although the 3 of these could be "one patch" realistically since they fix the same problem. I know there are a lot of calls to CU_DEBUG, perhaps it'd be best to go through all of them to make sure there are no other similar issues lurking. John

1 0

Re: [Libvirt-cim] [PATCH V3 06/11] DevicePool, reimplement get_diskpool_config with libvirt
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00023.html NACK this since it's replaced by: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00039.html I will review that one separately. John

1 0

Re: [Libvirt-cim] [PATCH V3 05/11] DevicePool, fix debug print crash
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the original email to reply-to, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00024.html ACK

1 0

Re: [Libvirt-cim] [PATCH V3 02/11] do not deregister virt classes in yum upgrade
by John Ferlan 12 Mar '13

12 Mar '13

Since I don't have the email to reply-to for a review, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00021.html I'm not sure what "$1" is in this context, let's consider setting "$1" to some variable that is more understandable in the context. Furthermore, is "0" a special number/string? How would anyone know reading the code without having knowledge of the code. John

2 1

Re: [Libvirt-cim] [PATCH V3 03/11] CSI, fix debug print crash
by John Ferlan 11 Mar '13

11 Mar '13

Since I don't have the email to reply-to, here is a link: https://www.redhat.com/archives/libvirt-cim/2012-December/msg00028.html ACK

1 0

[RFC] Patches need no testcase
by Wenchao Xia 05 Mar '13

05 Mar '13

Hi, Danial Happy new year! Could u have a look v3 serial of libcmpituil patches? It is simple and need no test case, hope it can be merged if no problem found. -- Best Regards Wenchao Xia

2 2