December 2016 - Libguestfs - libguestfs List Archives

[PATCH 1/3] generator: c: move internal functions
by Pino Toscano 07 Dec '16

07 Dec '16

Move the generate_all_structs and generate_all_headers functions, previously internal within the implementation of generate_guestfs_h, to be usable by other functions in the same "C" module (but not public). Only code motion. --- generator/c.ml | 163 +++++++++++++++++++++++++++++---------------------------- 1 file changed, 82 insertions(+), 81 deletions(-) diff --git a/generator/c.ml b/generator/c.ml index 6f5a517..f0df5ea 100644 --- a/generator/c.ml +++ b/generator/c.ml @@ -585,21 +585,85 @@ extern GUESTFS_DLL_PUBLIC void *guestfs_next_private (guestfs_h *g, const char * /* Structures. */ "; - (* The structures are carefully written to have exactly the same - * in-memory format as the XDR structures that we use on the wire to - * the daemon. The reason for creating copies of these structures - * here is just so we don't have to export the whole of - * guestfs_protocol.h (which includes much unrelated and - * XDR-dependent stuff that we don't want to be public, or required - * by clients). - * - * To reiterate, we will pass these structures to and from the client - * with a simple assignment or memcpy, so the format must be - * identical to what rpcgen / the RFC defines. - *) - (* Public structures. *) - let generate_all_structs = List.iter ( + generate_all_structs external_structs; + + pr "\ +/* Actions. */ +"; + + generate_all_headers public_functions_sorted; + + pr "\ +#if GUESTFS_PRIVATE +/* Symbols protected by GUESTFS_PRIVATE are NOT part of the public, + * stable API, and can change at any time! We export them because + * they are used by some of the language bindings. + */ + +/* Private functions. */ + +"; + + generate_all_headers private_functions_sorted; + + pr "\ +/* Private structures. */ + +"; + + generate_all_structs internal_structs; + +pr "\ + +#endif /* End of GUESTFS_PRIVATE. */ + +/* Deprecated macros. Use GUESTFS_HAVE_* instead. */ + +#define LIBGUESTFS_HAVE_CREATE_FLAGS 1 +#define LIBGUESTFS_HAVE_LAST_ERRNO 1 +#define LIBGUESTFS_HAVE_PUSH_ERROR_HANDLER 1 +#define LIBGUESTFS_HAVE_POP_ERROR_HANDLER 1 +#define LIBGUESTFS_HAVE_SET_EVENT_CALLBACK 1 +#define LIBGUESTFS_HAVE_DELETE_EVENT_CALLBACK 1 +#define LIBGUESTFS_HAVE_SET_CLOSE_CALLBACK 1 +#define LIBGUESTFS_HAVE_SET_PROGRESS_CALLBACK 1 +#define LIBGUESTFS_HAVE_SET_PRIVATE 1 +#define LIBGUESTFS_HAVE_GET_PRIVATE 1 +#define LIBGUESTFS_HAVE_FIRST_PRIVATE 1 +#define LIBGUESTFS_HAVE_NEXT_PRIVATE 1 + +"; + + List.iter ( + fun { name = shortname } -> + pr "#define LIBGUESTFS_HAVE_%s 1\n" (String.uppercase shortname); + ) public_functions_sorted; + + pr " +/* End of deprecated macros. */ + +#ifdef __cplusplus +} +#endif + +#endif /* GUESTFS_H_ */ +" + +(* The structures are carefully written to have exactly the same + * in-memory format as the XDR structures that we use on the wire to + * the daemon. The reason for creating copies of these structures + * here is just so we don't have to export the whole of + * guestfs_protocol.h (which includes much unrelated and + * XDR-dependent stuff that we don't want to be public, or required + * by clients). + * + * To reiterate, we will pass these structures to and from the client + * with a simple assignment or memcpy, so the format must be + * identical to what rpcgen / the RFC defines. + *) +and generate_all_structs structs = + List.iter ( fun { s_name = typ; s_cols = cols } -> pr "#define GUESTFS_HAVE_STRUCT_%s 1\n" (String.uppercase typ); pr "\n"; @@ -634,14 +698,9 @@ extern GUESTFS_DLL_PUBLIC void *guestfs_next_private (guestfs_h *g, const char * pr "extern GUESTFS_DLL_PUBLIC void guestfs_free_%s (struct guestfs_%s *);\n" typ typ; pr "extern GUESTFS_DLL_PUBLIC void guestfs_free_%s_list (struct guestfs_%s_list *);\n" typ typ; pr "\n" - ) in - - generate_all_structs external_structs; - - pr "\ -/* Actions. */ -"; + ) structs +and generate_all_headers actions = let generate_action_header { name = shortname; style = ret, args, optargs as style; deprecated_by = deprecated_by } = @@ -700,7 +759,7 @@ extern GUESTFS_DLL_PUBLIC void *guestfs_next_private (guestfs_h *g, const char * pr "\n" in - let generate_all_headers = List.iter ( + List.iter ( fun ({ name = name; style = ret, args, _ } as f) -> (* If once_had_no_optargs is set, then we need to generate a * <name>_opts variant, plus a backwards-compatible wrapper @@ -712,65 +771,7 @@ extern GUESTFS_DLL_PUBLIC void *guestfs_next_private (guestfs_h *g, const char * ) else generate_action_header f - ) in - - generate_all_headers public_functions_sorted; - - pr "\ -#if GUESTFS_PRIVATE -/* Symbols protected by GUESTFS_PRIVATE are NOT part of the public, - * stable API, and can change at any time! We export them because - * they are used by some of the language bindings. - */ - -/* Private functions. */ - -"; - - generate_all_headers private_functions_sorted; - - pr "\ -/* Private structures. */ - -"; - - generate_all_structs internal_structs; - -pr "\ - -#endif /* End of GUESTFS_PRIVATE. */ - -/* Deprecated macros. Use GUESTFS_HAVE_* instead. */ - -#define LIBGUESTFS_HAVE_CREATE_FLAGS 1 -#define LIBGUESTFS_HAVE_LAST_ERRNO 1 -#define LIBGUESTFS_HAVE_PUSH_ERROR_HANDLER 1 -#define LIBGUESTFS_HAVE_POP_ERROR_HANDLER 1 -#define LIBGUESTFS_HAVE_SET_EVENT_CALLBACK 1 -#define LIBGUESTFS_HAVE_DELETE_EVENT_CALLBACK 1 -#define LIBGUESTFS_HAVE_SET_CLOSE_CALLBACK 1 -#define LIBGUESTFS_HAVE_SET_PROGRESS_CALLBACK 1 -#define LIBGUESTFS_HAVE_SET_PRIVATE 1 -#define LIBGUESTFS_HAVE_GET_PRIVATE 1 -#define LIBGUESTFS_HAVE_FIRST_PRIVATE 1 -#define LIBGUESTFS_HAVE_NEXT_PRIVATE 1 - -"; - - List.iter ( - fun { name = shortname } -> - pr "#define LIBGUESTFS_HAVE_%s 1\n" (String.uppercase shortname); - ) public_functions_sorted; - - pr " -/* End of deprecated macros. */ - -#ifdef __cplusplus -} -#endif - -#endif /* GUESTFS_H_ */ -" + ) actions (* Generate the guestfs-internal-actions.h file. *) and generate_internal_actions_h () = -- 2.7.4

2 4

[PATCH 1/2] libvirt: un-duplicate XPath code
by Pino Toscano 07 Dec '16

07 Dec '16

Move the checks for empty xmlXPathObjectPtr, and for extracting the result string out of it, to a new helper functions. This is just code motion, there should be no behaviour changes. --- src/libvirt-domain.c | 122 +++++++++++++++++++++------------------------------ 1 file changed, 50 insertions(+), 72 deletions(-) diff --git a/src/libvirt-domain.c b/src/libvirt-domain.c index 4d4142d..baab307 100644 --- a/src/libvirt-domain.c +++ b/src/libvirt-domain.c @@ -43,6 +43,8 @@ static xmlDocPtr get_domain_xml (guestfs_h *g, virDomainPtr dom); static ssize_t for_each_disk (guestfs_h *g, virConnectPtr conn, xmlDocPtr doc, int (*f) (guestfs_h *g, const char *filename, const char *format, int readonly, const char *protocol, char *const *server, const char *username, void *data), void *data); static int libvirt_selinux_label (guestfs_h *g, xmlDocPtr doc, char **label_rtn, char **imagelabel_rtn); static char *filename_from_pool (guestfs_h *g, virConnectPtr conn, const char *pool_nane, const char *volume_name); +static bool xPathObjectIsEmpty (xmlXPathObjectPtr obj); +static char *xPathObjectGetString (xmlDocPtr doc, xmlXPathObjectPtr obj); static void ignore_errors (void *ignore, virErrorPtr ignore2) @@ -513,7 +515,6 @@ for_each_disk (guestfs_h *g, CLEANUP_XMLXPATHFREEOBJECT xmlXPathObjectPtr xpusername = NULL; CLEANUP_XMLXPATHFREEOBJECT xmlXPathObjectPtr xppool = NULL; CLEANUP_XMLXPATHFREEOBJECT xmlXPathObjectPtr xpvolume = NULL; - xmlAttrPtr attr; int readonly; int t; @@ -527,31 +528,21 @@ for_each_disk (guestfs_h *g, * Check the <disk type=..> attribute first to find out which one. */ xptype = xmlXPathEvalExpression (BAD_CAST "./@type", xpathCtx); - if (xptype == NULL || - xptype->nodesetval == NULL || - xptype->nodesetval->nodeNr == 0) { + if (xPathObjectIsEmpty (xptype)) continue; /* no type attribute, skip it */ - } - assert (xptype->nodesetval->nodeTab[0]); - assert (xptype->nodesetval->nodeTab[0]->type == XML_ATTRIBUTE_NODE); - attr = (xmlAttrPtr) xptype->nodesetval->nodeTab[0]; - type = (char *) xmlNodeListGetString (doc, attr->children, 1); + type = xPathObjectGetString (doc, xptype); if (STREQ (type, "file")) { /* type = "file" so look at source/@file */ xpathCtx->node = nodes->nodeTab[i]; xpfilename = xmlXPathEvalExpression (BAD_CAST "./source/@file", xpathCtx); - if (xpfilename == NULL || - xpfilename->nodesetval == NULL || - xpfilename->nodesetval->nodeNr == 0) + if (xPathObjectIsEmpty (xpfilename)) continue; /* disk filename not found, skip this */ } else if (STREQ (type, "block")) { /* type = "block", use source/@dev */ xpathCtx->node = nodes->nodeTab[i]; xpfilename = xmlXPathEvalExpression (BAD_CAST "./source/@dev", xpathCtx); - if (xpfilename == NULL || - xpfilename->nodesetval == NULL || - xpfilename->nodesetval->nodeNr == 0) + if (xPathObjectIsEmpty (xpfilename)) continue; /* disk filename not found, skip this */ } else if (STREQ (type, "network")) { /* type = "network", use source/@name */ int hi; @@ -562,15 +553,9 @@ for_each_disk (guestfs_h *g, /* Get the protocol (e.g. "rbd"). Required. */ xpprotocol = xmlXPathEvalExpression (BAD_CAST "./source/@protocol", xpathCtx); - if (xpprotocol == NULL || - xpprotocol->nodesetval == NULL || - xpprotocol->nodesetval->nodeNr == 0) + if (xPathObjectIsEmpty (xpprotocol)) continue; - assert (xpprotocol->nodesetval->nodeTab[0]); - assert (xpprotocol->nodesetval->nodeTab[0]->type == - XML_ATTRIBUTE_NODE); - attr = (xmlAttrPtr) xpprotocol->nodesetval->nodeTab[0]; - protocol = (char *) xmlNodeListGetString (doc, attr->children, 1); + protocol = xPathObjectGetString (doc, xpprotocol); debug (g, "disk[%zu]: protocol: %s", i, protocol); /* <source name="..."> is the path/exportname. Optional. */ @@ -583,13 +568,8 @@ for_each_disk (guestfs_h *g, /* <auth username="...">. Optional. */ xpusername = xmlXPathEvalExpression (BAD_CAST "./auth/@username", xpathCtx); - if (xpusername != NULL && - xpusername->nodesetval != NULL && - xpusername->nodesetval->nodeNr != 0) { - assert (xpusername->nodesetval->nodeTab[0]); - assert (xpusername->nodesetval->nodeTab[0]->type == XML_ATTRIBUTE_NODE); - attr = (xmlAttrPtr) xpusername->nodesetval->nodeTab[0]; - username = (char *) xmlNodeListGetString (doc, attr->children, 1); + if (!xPathObjectIsEmpty (xpusername)) { + username = xPathObjectGetString (doc, xpusername); debug (g, "disk[%zu]: username: %s", i, username); } @@ -641,28 +621,16 @@ for_each_disk (guestfs_h *g, /* Get the source pool. Required. */ xppool = xmlXPathEvalExpression (BAD_CAST "./source/@pool", xpathCtx); - if (xppool == NULL || - xppool->nodesetval == NULL || - xppool->nodesetval->nodeNr == 0) + if (xPathObjectIsEmpty (xppool)) continue; - assert (xppool->nodesetval->nodeTab[0]); - assert (xppool->nodesetval->nodeTab[0]->type == - XML_ATTRIBUTE_NODE); - attr = (xmlAttrPtr) xppool->nodesetval->nodeTab[0]; - pool = (char *) xmlNodeListGetString (doc, attr->children, 1); + pool = xPathObjectGetString (doc, xppool); /* Get the source volume. Required. */ xpvolume = xmlXPathEvalExpression (BAD_CAST "./source/@volume", xpathCtx); - if (xpvolume == NULL || - xpvolume->nodesetval == NULL || - xpvolume->nodesetval->nodeNr == 0) + if (xPathObjectIsEmpty (xpvolume)) continue; - assert (xpvolume->nodesetval->nodeTab[0]); - assert (xpvolume->nodesetval->nodeTab[0]->type == - XML_ATTRIBUTE_NODE); - attr = (xmlAttrPtr) xpvolume->nodesetval->nodeTab[0]; - volume = (char *) xmlNodeListGetString (doc, attr->children, 1); + volume = xPathObjectGetString (doc, xpvolume); debug (g, "disk[%zu]: pool: %s; volume: %s", i, pool, volume); @@ -679,13 +647,8 @@ for_each_disk (guestfs_h *g, if (filename == NULL) { assert (xpfilename); assert (xpfilename->nodesetval); - if (xpfilename->nodesetval->nodeNr > 0) { - assert (xpfilename->nodesetval->nodeTab[0]); - assert (xpfilename->nodesetval->nodeTab[0]->type == - XML_ATTRIBUTE_NODE); - attr = (xmlAttrPtr) xpfilename->nodesetval->nodeTab[0]; - filename = (char *) xmlNodeListGetString (doc, attr->children, 1); - } + if (xpfilename->nodesetval->nodeNr > 0) + filename = xPathObjectGetString (doc, xpfilename); else /* For network protocols (eg. nbd), name may be omitted. */ filename = safe_strdup (g, ""); @@ -696,22 +659,14 @@ for_each_disk (guestfs_h *g, /* Get the disk format (may not be set). */ xpathCtx->node = nodes->nodeTab[i]; xpformat = xmlXPathEvalExpression (BAD_CAST "./driver/@type", xpathCtx); - if (xpformat != NULL && - xpformat->nodesetval && - xpformat->nodesetval->nodeNr > 0) { - assert (xpformat->nodesetval->nodeTab[0]); - assert (xpformat->nodesetval->nodeTab[0]->type == XML_ATTRIBUTE_NODE); - attr = (xmlAttrPtr) xpformat->nodesetval->nodeTab[0]; - format = (char *) xmlNodeListGetString (doc, attr->children, 1); - } + if (!xPathObjectIsEmpty (xpformat)) + format = xPathObjectGetString (doc, xpformat); /* Get the <readonly/> flag. */ xpathCtx->node = nodes->nodeTab[i]; xpreadonly = xmlXPathEvalExpression (BAD_CAST "./readonly", xpathCtx); readonly = 0; - if (xpreadonly != NULL && - xpreadonly->nodesetval && - xpreadonly->nodesetval->nodeNr > 0) + if (!xPathObjectIsEmpty (xpreadonly)) readonly = 1; if (f) @@ -774,23 +729,17 @@ connect_live (guestfs_h *g, virDomainPtr dom) if (nodes != NULL) { for (i = 0; i < nodes->nodeNr; ++i) { CLEANUP_XMLXPATHFREEOBJECT xmlXPathObjectPtr xppath = NULL; - xmlAttrPtr attr; /* See note in function above. */ xpathCtx->node = nodes->nodeTab[i]; /* The path is in <source path=..> attribute. */ xppath = xmlXPathEvalExpression (BAD_CAST "./source/@path", xpathCtx); - if (xppath == NULL || - xppath->nodesetval == NULL || - xppath->nodesetval->nodeNr == 0) { + if (xPathObjectIsEmpty (xppath)) { xmlXPathFreeObject (xppath); continue; /* no type attribute, skip it */ } - assert (xppath->nodesetval->nodeTab[0]); - assert (xppath->nodesetval->nodeTab[0]->type == XML_ATTRIBUTE_NODE); - attr = (xmlAttrPtr) xppath->nodesetval->nodeTab[0]; - path = (char *) xmlNodeListGetString (doc, attr->children, 1); + path = xPathObjectGetString (doc, xppath); break; } } @@ -888,6 +837,35 @@ filename_from_pool (guestfs_h *g, virConnectPtr conn, return filename; } +/* Check that C<obj> is not empty. + */ +static bool +xPathObjectIsEmpty (xmlXPathObjectPtr obj) +{ + return obj == NULL || + obj->nodesetval == NULL || + obj->nodesetval->nodeNr == 0; +} + +/* Get the string value from C<obj>. + * + * C<obj> is I<required> to not be empty, i.e. that C<xPathObjectIsEmpty> + * is C<false>. + */ +static char * +xPathObjectGetString (xmlDocPtr doc, xmlXPathObjectPtr obj) +{ + xmlAttrPtr attr; + char *value; + + assert (obj->nodesetval->nodeTab[0]); + assert (obj->nodesetval->nodeTab[0]->type == XML_ATTRIBUTE_NODE); + attr = (xmlAttrPtr) obj->nodesetval->nodeTab[0]; + value = (char *) xmlNodeListGetString (doc, attr->children, 1); + + return value; +} + #else /* no libvirt at compile time */ #define NOT_IMPL(r) \ -- 2.7.4

2 3

[PATCH 1/2] daemon: allow to change the labels of swap partitions
by Pino Toscano 07 Dec '16

07 Dec '16

--- daemon/daemon.h | 1 + daemon/labels.c | 3 +++ daemon/swap.c | 21 +++++++++++++++++++++ generator/actions.ml | 4 ++++ 4 files changed, 29 insertions(+) diff --git a/daemon/daemon.h b/daemon/daemon.h index 79a5288..2379e31 100644 --- a/daemon/daemon.h +++ b/daemon/daemon.h @@ -254,6 +254,7 @@ extern int64_t ntfs_minimum_size (const char *device); /*-- in swap.c --*/ extern int swap_set_uuid (const char *device, const char *uuid); +extern int swap_set_label (const char *device, const char *label); /* ordinary daemon functions use these to indicate errors * NB: you don't need to prefix the string with the current command, diff --git a/daemon/labels.c b/daemon/labels.c index 20f27cb..aaa3eaf 100644 --- a/daemon/labels.c +++ b/daemon/labels.c @@ -85,6 +85,9 @@ do_set_label (const mountable_t *mountable, const char *label) else if (STREQ (vfs_type, "xfs")) r = xfslabel (mountable->device, label); + else if (STREQ (vfs_type, "swap")) + r = swap_set_label (mountable->device, label); + else NOT_SUPPORTED (-1, "don't know how to set the label for '%s' filesystems", vfs_type); diff --git a/daemon/swap.c b/daemon/swap.c index 9d7839e..028bc1e 100644 --- a/daemon/swap.c +++ b/daemon/swap.c @@ -239,3 +239,24 @@ swap_set_uuid (const char *device, const char *uuid) return 0; } + +int +swap_set_label (const char *device, const char *label) +{ + int r; + CLEANUP_FREE char *err = NULL; + + if (strlen (label) > SWAP_LABEL_MAX) { + reply_with_error ("%s: Linux swap labels are limited to %d bytes", + label, SWAP_LABEL_MAX); + return -1; + } + + r = command (NULL, &err, str_swaplabel, "-L", label, device, NULL); + if (r == -1) { + reply_with_error ("%s", err); + return -1; + } + + return 0; +} diff --git a/generator/actions.ml b/generator/actions.ml index 43de38b..5e0356f 100644 --- a/generator/actions.ml +++ b/generator/actions.ml @@ -10304,6 +10304,10 @@ when trying to set the label. The label is limited to 11 bytes. +=item swap + +The label is limited to 16 bytes. + =back If there is no support for changing the label -- 2.7.4

2 3

[PATCH 1/2] kernel: refactor build_kernel & find_kernel
by Pino Toscano 07 Dec '16

07 Dec '16

Move the handling of the SUPERMIN_* envvars, and the kernel copy/symlink out of find_kernel to build_kernel: this way find_kernel does only the search. Also, the modules path search is done in by find_kernel as well. This is mostly code motion, with no behaviour changes. --- src/kernel.ml | 89 +++++++++++++++++++++++++++++------------------------------ 1 file changed, 44 insertions(+), 45 deletions(-) diff --git a/src/kernel.ml b/src/kernel.ml index 9b0e8a2..2e061d8 100644 --- a/src/kernel.ml +++ b/src/kernel.ml @@ -40,28 +40,7 @@ let patt_of_cpu host_cpu = let rec build_kernel debug host_cpu dtb_wildcard copy_kernel kernel dtb = (* Locate the kernel. *) - let kernel_name, kernel_version = - find_kernel debug host_cpu copy_kernel kernel in - - (* If the user passed --dtb option, locate dtb. *) - (match dtb_wildcard with - | None -> () - | Some wildcard -> - find_dtb debug copy_kernel kernel_name wildcard dtb - ); - - (* Get the kernel modules. *) - let modpath = find_modpath debug kernel_version in - - if debug >= 1 then ( - printf "supermin: kernel: kernel_version %s\n" kernel_version; - printf "supermin: kernel: modules %s\n%!" modpath; - ); - - (kernel_version, modpath) - -and find_kernel debug host_cpu copy_kernel kernel = - let kernel_file, kernel_name, kernel_version = + let kernel_file, kernel_name, kernel_version, modpath = try let kernel_env = getenv "SUPERMIN_KERNEL" in if debug >= 1 then @@ -78,38 +57,58 @@ and find_kernel debug host_cpu copy_kernel kernel = printf "supermin: kernel: SUPERMIN_KERNEL version %s\n%!" kernel_version; let kernel_name = Filename.basename kernel_env in - kernel_env, kernel_name, kernel_version + let modpath = find_modpath debug kernel_version in + kernel_env, kernel_name, kernel_version, modpath with Not_found -> - let is_arm = - String.length host_cpu >= 3 && - host_cpu.[0] = 'a' && host_cpu.[1] = 'r' && host_cpu.[2] = 'm' in + find_kernel debug host_cpu kernel in - let all_files = Sys.readdir "/boot" in - let all_files = Array.to_list all_files in + (* If the user passed --dtb option, locate dtb. *) + (match dtb_wildcard with + | None -> () + | Some wildcard -> + find_dtb debug copy_kernel kernel_name wildcard dtb + ); - (* In original: ls -1dvr /boot/vmlinuz-*.$arch* 2>/dev/null | grep -v xen *) - let patterns = patt_of_cpu host_cpu in - let files = kernel_filter patterns is_arm all_files in + if debug >= 1 then ( + printf "supermin: kernel: kernel_version %s\n" kernel_version; + printf "supermin: kernel: modules %s\n%!" modpath; + ); - let files = - if files <> [] then files - else - (* In original: ls -1dvr /boot/vmlinuz-* 2>/dev/null | grep -v xen *) - kernel_filter ["vmlinu?-*"] is_arm all_files in + copy_or_symlink_file copy_kernel kernel_file kernel; - if files = [] then no_kernels host_cpu; + (kernel_version, modpath) - let files = List.sort (fun a b -> compare_version b a) files in - let kernel_name = List.hd files in - let kernel_version = get_kernel_version kernel_name in +and find_kernel debug host_cpu kernel = + let is_arm = + String.length host_cpu >= 3 && + host_cpu.[0] = 'a' && host_cpu.[1] = 'r' && host_cpu.[2] = 'm' in - if debug >= 1 then - printf "supermin: kernel: picked kernel %s\n%!" kernel_name; + let all_files = Sys.readdir "/boot" in + let all_files = Array.to_list all_files in - ("/boot" // kernel_name), kernel_name, kernel_version in + (* In original: ls -1dvr /boot/vmlinuz-*.$arch* 2>/dev/null | grep -v xen *) + let patterns = patt_of_cpu host_cpu in + let files = kernel_filter patterns is_arm all_files in - copy_or_symlink_file copy_kernel kernel_file kernel; - kernel_name, kernel_version + let files = + if files <> [] then files + else + (* In original: ls -1dvr /boot/vmlinuz-* 2>/dev/null | grep -v xen *) + kernel_filter ["vmlinu?-*"] is_arm all_files in + + if files = [] then no_kernels host_cpu; + + let files = List.sort (fun a b -> compare_version b a) files in + let kernel_name = List.hd files in + let kernel_version = get_kernel_version kernel_name in + + if debug >= 1 then + printf "supermin: kernel: picked kernel %s\n%!" kernel_name; + + (* Get the kernel modules. *) + let modpath = find_modpath debug kernel_version in + + ("/boot" // kernel_name), kernel_name, kernel_version, modpath and kernel_filter patterns is_arm all_files = let files = -- 2.7.4

2 2

[PATCH] v2v: Rename RHEV to RHV throughout.
by Richard W.M. Jones 07 Dec '16

07 Dec '16

You can now use -o rhv (-o rhev is supported for compatibility). --- v2v/Makefile.am | 8 +- v2v/OVF.ml | 30 +++---- v2v/OVF.mli | 2 +- v2v/changeuid.mli | 4 +- v2v/cmdline.ml | 12 +-- v2v/convert_windows.ml | 2 +- v2v/{output_rhev.ml => output_rhv.ml} | 26 +++--- v2v/{output_rhev.mli => output_rhv.mli} | 8 +- v2v/output_vdsm.ml | 4 +- v2v/test-harness/virt-v2v-test-harness.pod | 4 +- v2v/test-v2v-no-copy.sh | 6 +- v2v/{test-v2v-o-rhev.sh => test-v2v-o-rhv.sh} | 10 +-- v2v/types.mli | 2 +- v2v/virt-v2v.pod | 110 +++++++++++++------------- 14 files changed, 116 insertions(+), 112 deletions(-) rename v2v/{output_rhev.ml => output_rhv.ml} (90%) rename v2v/{output_rhev.mli => output_rhv.mli} (83%) rename v2v/{test-v2v-o-rhev.sh => test-v2v-o-rhv.sh} (95%) diff --git a/v2v/Makefile.am b/v2v/Makefile.am index 5e3c3eb..2802056 100644 --- a/v2v/Makefile.am +++ b/v2v/Makefile.am @@ -49,7 +49,7 @@ SOURCES_MLI = \ output_local.mli \ output_null.mli \ output_qemu.mli \ - output_rhev.mli \ + output_rhv.mli \ output_vdsm.mli \ OVF.mli \ target_bus_assignment.mli \ @@ -94,7 +94,7 @@ SOURCES_ML = \ output_libvirt.ml \ output_local.ml \ output_qemu.ml \ - output_rhev.ml \ + output_rhv.ml \ output_vdsm.ml \ inspect_source.ml \ target_bus_assignment.ml \ @@ -283,7 +283,7 @@ TESTS += \ test-v2v-o-libvirt.sh \ test-v2v-o-null.sh \ test-v2v-o-qemu.sh \ - test-v2v-o-rhev.sh \ + test-v2v-o-rhv.sh \ test-v2v-o-vdsm-options.sh \ test-v2v-oa-option.sh \ test-v2v-of-option.sh \ @@ -373,7 +373,7 @@ EXTRA_DIST += \ test-v2v-o-libvirt.sh \ test-v2v-o-null.sh \ test-v2v-o-qemu.sh \ - test-v2v-o-rhev.sh \ + test-v2v-o-rhv.sh \ test-v2v-o-vdsm-options.sh \ test-v2v-oa-option.sh \ test-v2v-of-option.sh \ diff --git a/v2v/OVF.ml b/v2v/OVF.ml index eea1e63..4c93c12 100644 --- a/v2v/OVF.ml +++ b/v2v/OVF.ml @@ -108,11 +108,11 @@ and get_ostype = function | { i_type = "linux" } -> "OtherLinux" | { i_type = "windows"; i_major_version = 5; i_minor_version = 1 } -> - "WindowsXP" (* no architecture differentiation of XP on RHEV *) + "WindowsXP" (* no architecture differentiation of XP on RHV *) | { i_type = "windows"; i_major_version = 5; i_minor_version = 2; i_product_name = product } when String.find product "XP" >= 0 -> - "WindowsXP" (* no architecture differentiation of XP on RHEV *) + "WindowsXP" (* no architecture differentiation of XP on RHV *) | { i_type = "windows"; i_major_version = 5; i_minor_version = 2; i_arch = "i386" } -> @@ -206,7 +206,7 @@ let origin_of_source_hypervisor = function (* Generate the .meta file associated with each volume. *) let create_meta_files output_alloc sd_uuid image_uuids targets = (* Note: Upper case in the .meta, mixed case in the OVF. *) - let output_alloc_for_rhev = + let output_alloc_for_rhv = match output_alloc with | Sparse -> "SPARSE" | Preallocated -> "PREALLOCATED" in @@ -220,12 +220,12 @@ let create_meta_files output_alloc sd_uuid image_uuids targets = ov.ov_virtual_size; ov.ov_virtual_size /^ 512L in - let format_for_rhev = + let format_for_rhv = match t.target_format with | "raw" -> "RAW" | "qcow2" -> "COW" | _ -> - error (f_"RHEV does not support the output format '%s', only raw or qcow2") t.target_format in + error (f_"RHV does not support the output format '%s', only raw or qcow2") t.target_format in let buf = Buffer.create 256 in let bpf fs = bprintf buf fs in @@ -239,8 +239,8 @@ let create_meta_files output_alloc sd_uuid image_uuids targets = bpf "LEGALITY=LEGAL\n"; bpf "POOL_UUID=\n"; bpf "SIZE=%Ld\n" size_in_sectors; - bpf "FORMAT=%s\n" format_for_rhev; - bpf "TYPE=%s\n" output_alloc_for_rhev; + bpf "FORMAT=%s\n" format_for_rhv; + bpf "TYPE=%s\n" output_alloc_for_rhv; bpf "DESCRIPTION=%s\n" (String.replace generated_by "=" "_"); bpf "EOF\n"; Buffer.contents buf @@ -329,7 +329,7 @@ let rec create_ovf source targets guestcaps inspect e "rasd:ResourceType" [] [PCData "23"]; e "rasd:UsbPolicy" [] [PCData "Disabled"]; ]; - (* We always add a qxl device when outputting to RHEV. + (* We always add a qxl device when outputting to RHV. * See RHBZ#1213701 and RHBZ#1211231 for the reasoning * behind that. *) @@ -366,7 +366,7 @@ let rec create_ovf source targets guestcaps inspect *) (match source with | { s_display = Some { s_password = Some _ } } -> - warning (f_"This guest required a password for connection to its display, but this is not supported by RHEV. Therefore the converted guest's display will not require a separate password to connect."); + warning (f_"This guest required a password for connection to its display, but this is not supported by RHV. Therefore the converted guest's display will not require a separate password to connect."); | _ -> ()); if verbose () then ( @@ -408,7 +408,7 @@ and add_disks targets guestcaps output_alloc sd_uuid image_uuids vol_uuids ovf = let fileref = sprintf "%s/%s" image_uuid vol_uuid in (* ovf:size and ovf:actual_size fields are integer GBs. If you - * use floating point numbers then RHEV will fail to parse them. + * use floating point numbers then RHV will fail to parse them. * In case the size is just below a gigabyte boundary, round up. *) let bytes_to_gb b = @@ -425,15 +425,15 @@ and add_disks targets guestcaps output_alloc sd_uuid image_uuids vol_uuids ovf = | None, Some estimated_size -> Some (bytes_to_gb estimated_size), true | None, None -> None, false in - let format_for_rhev = + let format_for_rhv = match t.target_format with | "raw" -> "RAW" | "qcow2" -> "COW" | _ -> - error (f_"RHEV does not support the output format '%s', only raw or qcow2") t.target_format in + error (f_"RHV does not support the output format '%s', only raw or qcow2") t.target_format in (* Note: Upper case in the .meta, mixed case in the OVF. *) - let output_alloc_for_rhev = + let output_alloc_for_rhv = match output_alloc with | Sparse -> "Sparse" | Preallocated -> "Preallocated" in @@ -456,8 +456,8 @@ and add_disks targets guestcaps output_alloc sd_uuid image_uuids vol_uuids ovf = "ovf:fileRef", fileref; "ovf:parentRef", ""; "ovf:vm_snapshot_id", uuidgen (); - "ovf:volume-format", format_for_rhev; - "ovf:volume-type", output_alloc_for_rhev; + "ovf:volume-format", format_for_rhv; + "ovf:volume-type", output_alloc_for_rhv; "ovf:format", "http://en.wikipedia.org/wiki/Byte"; (* wtf? *) "ovf:disk-interface", (match guestcaps.gcaps_block_bus with diff --git a/v2v/OVF.mli b/v2v/OVF.mli index 3b260a5..89b96da 100644 --- a/v2v/OVF.mli +++ b/v2v/OVF.mli @@ -21,7 +21,7 @@ val create_meta_files : Types.output_allocation -> string -> string list -> Types.target list -> string list (** Create the .meta file associated with each target. - Note this does not write them, since output_rhev has to do a + Note this does not write them, since output_rhv has to do a permissions dance when writing files. Instead the contents of each file is returned (one per target), and they must be written to [target_file ^ ".meta"]. *) diff --git a/v2v/changeuid.mli b/v2v/changeuid.mli index 7838f0f..52f5359 100644 --- a/v2v/changeuid.mli +++ b/v2v/changeuid.mli @@ -18,8 +18,8 @@ (** Functions for making files and directories as another user. - [-o rhev] output mode has to write files as UID:GID 36:36, - otherwise RHEV cannot read them. Because the files are located on + [-o rhv] output mode has to write files as UID:GID 36:36, + otherwise RHV cannot read them. Because the files are located on NFS (and hence might be root-squashed) we also cannot chown the files. We cannot setuid the whole process to 36:36 because it needs to do other root things like mounting and unmounting the NFS diff --git a/v2v/cmdline.ml b/v2v/cmdline.ml index 9a56d60..7b57a6f 100644 --- a/v2v/cmdline.ml +++ b/v2v/cmdline.ml @@ -123,7 +123,7 @@ let parse_cmdline () = | "libvirt" -> output_mode := `Libvirt | "disk" | "local" -> output_mode := `Local | "null" -> output_mode := `Null - | "ovirt" | "rhev" -> output_mode := `RHEV + | "ovirt" | "rhv" | "rhev" -> output_mode := `RHV | "qemu" -> output_mode := `QEmu | "vdsm" -> output_mode := `VDSM | s -> @@ -218,7 +218,7 @@ let parse_cmdline () = virt-v2v -ic vpx://vcenter.example.com/Datacenter/esxi -os imported esx_guest virt-v2v -ic vpx://vcenter.example.com/Datacenter/esxi esx_guest \ - -o rhev -os rhev.nfs:/export_domain --network rhevm + -o rhv -os rhv.nfs:/export_domain --network rhvm virt-v2v -i libvirtxml guest-domain.xml -o local -os /var/tmp @@ -389,15 +389,15 @@ read the man page virt-v2v(1). | Some d -> d in Output_qemu.output_qemu os qemu_boot - | `RHEV -> + | `RHV -> let os = match output_storage with | None -> - error (f_"-o rhev: output storage was not specified, use '-os'"); + error (f_"-o rhv: output storage was not specified, use '-os'"); | Some d -> d in if qemu_boot then - error (f_"-o rhev: --qemu-boot option cannot be used in this output mode"); - Output_rhev.output_rhev os output_alloc + error (f_"-o rhv: --qemu-boot option cannot be used in this output mode"); + Output_rhv.output_rhv os output_alloc | `VDSM -> let os = diff --git a/v2v/convert_windows.ml b/v2v/convert_windows.ml index 25acdcc..f8337a0 100644 --- a/v2v/convert_windows.ml +++ b/v2v/convert_windows.ml @@ -319,7 +319,7 @@ echo Wait for PnP to complete (String.concat "/" pnp_wait_path)) and configure_rhev_apt tool_path = - (* Configure RHEV-APT (the RHEV guest agent). However if it doesn't + (* Configure RHEV-APT (the RHV guest agent). However if it doesn't * exist just warn about it and continue. *) g#upload tool_path "/rhev-apt.exe"; (* XXX *) diff --git a/v2v/output_rhev.ml b/v2v/output_rhv.ml similarity index 90% rename from v2v/output_rhev.ml rename to v2v/output_rhv.ml index 3280150..5ddcefa 100644 --- a/v2v/output_rhev.ml +++ b/v2v/output_rhv.ml @@ -66,7 +66,7 @@ and check_storage_domain domain_class os mp = let entries = try Sys.readdir mp with Sys_error msg -> - error (f_"could not read the %s specified by the '-os %s' parameter on the command line. Is it really an OVirt or RHEV-M %s? The original error is: %s") domain_class os domain_class msg in + error (f_"could not read the %s specified by the '-os %s' parameter on the command line. Is it really an OVirt or RHV-M %s? The original error is: %s") domain_class os domain_class msg in let entries = Array.to_list entries in let uuids = List.filter ( fun entry -> @@ -78,7 +78,7 @@ and check_storage_domain domain_class os mp = match uuids with | [uuid] -> uuid | [] -> - error (f_"there are no UUIDs in the %s (%s). Is it really an OVirt or RHEV-M %s?") domain_class os domain_class + error (f_"there are no UUIDs in the %s (%s). Is it really an OVirt or RHV-M %s?") domain_class os domain_class | _::_ -> error (f_"there are multiple UUIDs in the %s (%s). This is unexpected, and may be a bug in virt-v2v or OVirt.") domain_class os in @@ -88,7 +88,7 @@ and check_storage_domain domain_class os mp = let () = let master_vms_dir = mp // uuid // "master" // "vms" in if not (is_directory master_vms_dir) then - error (f_"%s does not exist or is not a directory.\n\nMost likely cause: Either the %s (%s) has not been attached to any Data Center, or the path %s is not an %s at all.\n\nYou have to attach the %s to a Data Center using the RHEV-M / OVirt user interface first.\n\nIf you don't know what the %s mount point should be then you can also find this out through the RHEV-M user interface.") + error (f_"%s does not exist or is not a directory.\n\nMost likely cause: Either the %s (%s) has not been attached to any Data Center, or the path %s is not an %s at all.\n\nYou have to attach the %s to a Data Center using the RHV-M / OVirt user interface first.\n\nIf you don't know what the %s mount point should be then you can also find this out through the RHV-M user interface.") master_vms_dir domain_class os os domain_class domain_class domain_class in @@ -98,7 +98,7 @@ and check_storage_domain domain_class os mp = (* UID:GID required for files and directories when writing to ESD. *) let uid = 36 and gid = 36 -class output_rhev os output_alloc = +class output_rhv os output_alloc = (* Create a UID-switching handle. If we're not root, create a dummy * one because we cannot switch UIDs. *) @@ -111,11 +111,11 @@ class output_rhev os output_alloc = object inherit output - method as_options = sprintf "-o rhev -os %s" os + method as_options = sprintf "-o rhv -os %s" os method supported_firmware = [ TargetBIOS ] - (* RHEV doesn't support serial consoles. This causes the conversion + (* RHV doesn't support serial consoles. This causes the conversion * step to remove it. *) method keep_serial_console = false @@ -157,7 +157,7 @@ object mount_and_check_storage_domain (s_"Export Storage Domain") os in esd_mp <- mp; esd_uuid <- uuid; - debug "RHEV: ESD mountpoint: %s\nRHEV: ESD UUID: %s" esd_mp esd_uuid; + debug "RHV: ESD mountpoint: %s\nRHV: ESD UUID: %s" esd_mp esd_uuid; (* See if we can write files as UID:GID 36:36. *) let () = @@ -166,10 +166,10 @@ object let stat = stat testfile in Changeuid.unlink changeuid_t testfile; let actual_uid = stat.st_uid and actual_gid = stat.st_gid in - debug "RHEV: actual UID:GID of new files is %d:%d" actual_uid actual_gid; + debug "RHV: actual UID:GID of new files is %d:%d" actual_uid actual_gid; if uid <> actual_uid || gid <> actual_gid then ( if running_as_root then - warning (f_"cannot write files to the NFS server as %d:%d, even though we appear to be running as root. This probably means the NFS client or idmapd is not configured properly.\n\nYou will have to chown the files that virt-v2v creates after the run, otherwise RHEV-M will not be able to import the VM.") uid gid + warning (f_"cannot write files to the NFS server as %d:%d, even though we appear to be running as root. This probably means the NFS client or idmapd is not configured properly.\n\nYou will have to chown the files that virt-v2v creates after the run, otherwise RHV-M will not be able to import the VM.") uid gid else warning (f_"cannot write files to the NFS server as %d:%d. You might want to stop virt-v2v (^C) and rerun it as root.") uid gid ) in @@ -225,7 +225,7 @@ object fun ({ target_overlay = ov } as t, image_uuid, vol_uuid) -> let ov_sd = ov.ov_sd in let target_file = images_dir // image_uuid // vol_uuid in - debug "RHEV: will export %s to %s" ov_sd target_file; + debug "RHV: will export %s to %s" ov_sd target_file; { t with target_file = target_file } ) (combine3 targets image_uuids vol_uuids) in @@ -247,7 +247,7 @@ object ?clustersize path format size = Changeuid.func changeuid_t ( fun () -> - let g = open_guestfs ~identifier:"rhev_disk_create" () in + let g = open_guestfs ~identifier:"rhv_disk_create" () in g#disk_create ?backingfile ?backingformat ?preallocation ?compat ?clustersize path format size; (* Make it sufficiently writable so that possibly root, or @@ -277,5 +277,5 @@ object delete_target_directory <- false end -let output_rhev = new output_rhev -let () = Modules_list.register_output_module "rhev" +let output_rhv = new output_rhv +let () = Modules_list.register_output_module "rhv" diff --git a/v2v/output_rhev.mli b/v2v/output_rhv.mli similarity index 83% rename from v2v/output_rhev.mli rename to v2v/output_rhv.mli index 27df737..71ddd0a 100644 --- a/v2v/output_rhev.mli +++ b/v2v/output_rhv.mli @@ -16,12 +16,12 @@ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. *) -(** [-o rhev] target. *) +(** [-o rhv] target. *) val mount_and_check_storage_domain : string -> string -> (string * string) (** This helper function is also used by the VDSM target. *) -val output_rhev : string -> Types.output_allocation -> Types.output -(** [output_rhev os output_alloc] creates and +val output_rhv : string -> Types.output_allocation -> Types.output +(** [output_rhv os output_alloc] creates and returns a new {!Types.output} object specialized for writing - output to RHEV-M or oVirt Export Storage Domain. *) + output to RHV-M or oVirt Export Storage Domain. *) diff --git a/v2v/output_vdsm.ml b/v2v/output_vdsm.ml index fbddee8..ee7cc4e 100644 --- a/v2v/output_vdsm.ml +++ b/v2v/output_vdsm.ml @@ -49,7 +49,7 @@ object method supported_firmware = [ TargetBIOS ] - (* RHEV doesn't support serial consoles. This causes the conversion + (* RHV doesn't support serial consoles. This causes the conversion * step to remove it. *) method keep_serial_console = false @@ -62,7 +62,7 @@ object * name of the target files that eventually get written by the main * code. * - * 'os' is the output storage domain (-os /rhev/data/<data center>/<data domain>) + * 'os' is the output storage domain (-os /rhv/data/<data center>/<data domain>) * this is already mounted path. * * Note it's good to fail here (early) if there are any problems, since diff --git a/v2v/test-harness/virt-v2v-test-harness.pod b/v2v/test-harness/virt-v2v-test-harness.pod index 4bbaca8..a323549 100644 --- a/v2v/test-harness/virt-v2v-test-harness.pod +++ b/v2v/test-harness/virt-v2v-test-harness.pod @@ -17,8 +17,8 @@ virt-v2v-test-harness - Used to test virt-v2v against real test cases =head1 DESCRIPTION L<virt-v2v(1)> converts guests from a foreign hypervisor to run on -KVM, managed by libvirt, OpenStack, oVirt, Red Hat Enterprise -Virtualisation (RHEV) or several other targets. +KVM, managed by libvirt, OpenStack, oVirt, Red Hat Virtualisation +(RHV) or several other targets. Virt-v2v-test-harness is a small library (module name: C<V2v_test_harness>) used to run virt-v2v against a set of test cases diff --git a/v2v/test-v2v-no-copy.sh b/v2v/test-v2v-no-copy.sh index 6309c92..23109ae 100755 --- a/v2v/test-v2v-no-copy.sh +++ b/v2v/test-v2v-no-copy.sh @@ -58,17 +58,17 @@ test -f $d/windows.xml # Test the disk was NOT created. ! test -f $d/windows-sda -# --no-copy with -o rhev. +# --no-copy with -o rhv. mkdir $d/12345678-1234-1234-1234-123456789abc mkdir $d/12345678-1234-1234-1234-123456789abc/images mkdir $d/12345678-1234-1234-1234-123456789abc/master mkdir $d/12345678-1234-1234-1234-123456789abc/master/vms -# $VG - XXX Disabled because the forking used to write files in -o rhev +# $VG - XXX Disabled because the forking used to write files in -o rhv # mode confuses valgrind. virt-v2v --debug-gc --no-copy \ -i libvirt -ic "$libvirt_uri" windows \ - -o rhev -os $d + -o rhv -os $d # Test the OVF metadata was created. test -f $d/12345678-1234-1234-1234-123456789abc/master/vms/*/*.ovf diff --git a/v2v/test-v2v-o-rhev.sh b/v2v/test-v2v-o-rhv.sh similarity index 95% rename from v2v/test-v2v-o-rhev.sh rename to v2v/test-v2v-o-rhv.sh index 43e2d42..ce43b27 100755 --- a/v2v/test-v2v-o-rhev.sh +++ b/v2v/test-v2v-o-rhv.sh @@ -16,13 +16,13 @@ # along with this program; if not, write to the Free Software # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. -# Test -o rhev. +# Test -o rhv. unset CDPATH export LANG=C set -e -if [ -n "$SKIP_TEST_V2V_O_RHEV_SH" ]; then +if [ -n "$SKIP_TEST_V2V_O_RHV_SH" ]; then echo "$0: test skipped because environment variable is set" exit 77 fi @@ -43,7 +43,7 @@ fi export VIRT_TOOLS_DATA_DIR="$srcdir/../test-data/fake-virt-tools" -d=test-v2v-o-rhev.d +d=test-v2v-o-rhv.d rm -rf $d mkdir $d @@ -53,11 +53,11 @@ mkdir $d/12345678-1234-1234-1234-123456789abc/images mkdir $d/12345678-1234-1234-1234-123456789abc/master mkdir $d/12345678-1234-1234-1234-123456789abc/master/vms -# $VG - XXX Disabled because the forking used to write files in -o rhev +# $VG - XXX Disabled because the forking used to write files in -o rhv # mode confuses valgrind. virt-v2v --debug-gc \ -i libvirt -ic "$libvirt_uri" windows \ - -o rhev -os $d + -o rhv -os $d # Test the OVF metadata was created. test -f $d/12345678-1234-1234-1234-123456789abc/master/vms/*/*.ovf diff --git a/v2v/types.mli b/v2v/types.mli index 514565c..b007bf4 100644 --- a/v2v/types.mli +++ b/v2v/types.mli @@ -366,6 +366,6 @@ class virtual output : object (** Called in order to create disks on the target. The method has the same signature as Guestfs#disk_create. *) method keep_serial_console : bool - (** Whether this output supports serial consoles (RHEV does not). *) + (** Whether this output supports serial consoles (RHV does not). *) end (** Encapsulates all [-o], etc output arguments as an object. *) diff --git a/v2v/virt-v2v.pod b/v2v/virt-v2v.pod index 1654b35..0382c95 100644 --- a/v2v/virt-v2v.pod +++ b/v2v/virt-v2v.pod @@ -7,7 +7,7 @@ virt-v2v - Convert a guest to use KVM virt-v2v -ic vpx://vcenter.example.com/Datacenter/esxi vmware_guest virt-v2v -ic vpx://vcenter.example.com/Datacenter/esxi vmware_guest \ - -o rhev -os rhev.nfs:/export_domain --network rhevm + -o rhv -os rhv.nfs:/export_domain --network rhvm virt-v2v -i libvirtxml guest-domain.xml -o local -os /var/tmp @@ -22,8 +22,8 @@ virt-v2v - Convert a guest to use KVM Virt-v2v converts guests from a foreign hypervisor to run on KVM. It can read Linux and Windows guests running on VMware, Xen, Hyper-V and some other hypervisors, and convert them to KVM managed by libvirt, -OpenStack, oVirt, Red Hat Enterprise Virtualisation (RHEV) or several -other targets. +OpenStack, oVirt, Red Hat Virtualisation (RHV) or several other +targets. There is also a companion front-end called L<virt-p2v(1)> which comes as an ISO, CD or PXE image that can be booted on physical machines to @@ -42,7 +42,7 @@ libguestfs E<ge> 1.28. Xen ───▶│ -i libvirt ──▶ │ │ │ (default) │ ... ───▶│ (default) │ │ │ ──┐ └────────────┘ └────────────┘ │ │ ─┐└──────▶ -o glance - -i libvirtxml ─────────▶ │ │ ┐└─────────▶ -o rhev + -i libvirtxml ─────────▶ │ │ ┐└─────────▶ -o rhv └────────────┘ └──────────▶ -o vdsm Virt-v2v has a number of possible input and output modes, selected @@ -74,7 +74,7 @@ libvirt configuration file (mainly for testing). I<-o qemu> writes to a local disk image with a shell script for booting the guest directly in qemu (mainly for testing). -I<-o rhev> is used to write to a RHEV-M / oVirt target. I<-o vdsm> +I<-o rhv> is used to write to a RHV / oVirt target. I<-o vdsm> is only used when virt-v2v runs under VDSM control. I<--in-place> instructs virt-v2v to customize the guest OS in the input @@ -97,23 +97,23 @@ disks to F</var/lib/libvirt/images>. For more information see L</INPUT FROM VMWARE VCENTER SERVER> below. -=head2 Convert from VMware to RHEV-M/oVirt +=head2 Convert from VMware to RHV/oVirt This is the same as the previous example, except you want to send the -guest to a RHEV-M Export Storage Domain which is located remotely -(over NFS) at C<rhev.nfs:/export_domain>. If you are unclear about +guest to a RHV-M Export Storage Domain which is located remotely +(over NFS) at C<rhv.nfs:/export_domain>. If you are unclear about the location of the Export Storage Domain you should check the -settings on your RHEV-M management console. Guest network -interface(s) are connected to the target network called C<rhevm>. +settings on your RHV-M management console. Guest network +interface(s) are connected to the target network called C<rhvm>. virt-v2v -ic vpx://vcenter.example.com/Datacenter/esxi vmware_guest \ - -o rhev -os rhev.nfs:/export_domain --network rhevm + -o rhv -os rhv.nfs:/export_domain --network rhvm In this case the host running virt-v2v acts as a B<conversion server>. -Note that after conversion, the guest will appear in the RHEV-M Export -Storage Domain, from where you will need to import it using the RHEV-M -user interface. (See L</OUTPUT TO RHEV>). +Note that after conversion, the guest will appear in the RHV-M Export +Storage Domain, from where you will need to import it using the RHV-M +user interface. (See L</OUTPUT TO RHV>). =head2 Convert disk image to OpenStack glance @@ -199,7 +199,7 @@ QEMU and KVM only. =item OpenStack Glance -=item Red Hat Enterprise Virtualization (RHEV) 4.1 and up +=item Red Hat Virtualization (RHV) 4.1 and up =item Local libvirt @@ -452,7 +452,7 @@ written. =item B<-o> B<ovirt> -This is the same as I<-o rhev>. +This is the same as I<-o rhv>. =item B<-o> B<qemu> @@ -467,22 +467,26 @@ option which boots the guest under qemu immediately. =item B<-o> B<rhev> -Set the output method to I<rhev>. +This is the same as I<-o rhv>. -The converted guest is written to a RHEV Export Storage Domain. The +=item B<-o> B<rhv> + +Set the output method to I<rhv>. + +The converted guest is written to a RHV Export Storage Domain. The I<-os> parameter must also be used to specify the location of the Export Storage Domain. Note this does not actually import the guest -into RHEV. You have to do that manually later using the UI. +into RHV. You have to do that manually later using the UI. -See L</OUTPUT TO RHEV> below. +See L</OUTPUT TO RHV> below. =item B<-o> B<vdsm> Set the output method to I<vdsm>. -This mode is similar to I<-o rhev>, but the full path to the +This mode is similar to I<-o rhv>, but the full path to the data domain must be given: -F</rhev/data-center/E<lt>data-center-uuidE<gt>/E<lt>data-domain-uuidE<gt>>. +F</rhv/data-center/E<lt>data-center-uuidE<gt>/E<lt>data-domain-uuidE<gt>>. This mode is only used when virt-v2v runs under VDSM control. =item B<-oa> B<sparse> @@ -520,10 +524,10 @@ For I<-o libvirt>, this is a libvirt directory pool For I<-o local> and I<-o qemu>, this is a directory name. The directory must exist. -For I<-o rhev>, this can be an NFS path of the Export Storage Domain +For I<-o rhv>, this can be an NFS path of the Export Storage Domain of the form C<E<lt>hostE<gt>:E<lt>pathE<gt>>, eg: - rhev-storage.example.com:/rhev/export + rhv-storage.example.com:/rhv/export The NFS export must be mountable and writable by the user and host running virt-v2v, since the virt-v2v program has to actually mount it @@ -621,7 +625,7 @@ default in a future version of virt-v2v. =item B<--vdsm-ovf-output> -Normally the RHEV output mode chooses random UUIDs for the target +Normally the RHV output mode chooses random UUIDs for the target guest. However VDSM needs to control the UUIDs and passes these parameters when virt-v2v runs under VDSM control. The parameters control: @@ -898,7 +902,7 @@ On RHEL E<ge> 7.3, only qemu-kvm-rhev (not qemu-kvm) is supported. Not supported. -=item UEFI on RHEV +=item UEFI on RHV Not supported. @@ -929,13 +933,13 @@ This is typical of a libvirt guest: It has a single network interface connected to a network called C<default>. To map a specific network to a target network, for example C<default> -on the source to C<rhevm> on the target, use: +on the source to C<rhvm> on the target, use: - virt-v2v [...] --network default:rhevm + virt-v2v [...] --network default:rhvm To map every network to a target network, use: - virt-v2v [...] --network rhevm + virt-v2v [...] --network rhvm Bridges are handled in the same way, but you have to use the I<--bridge> option instead. For example: @@ -1471,29 +1475,29 @@ Define the final guest in libvirt: =back -=head1 OUTPUT TO RHEV +=head1 OUTPUT TO RHV -This section only applies to the I<-o rhev> output mode. If you use -virt-v2v from the RHEV-M user interface, then behind the scenes the +This section only applies to the I<-o rhv> output mode. If you use +virt-v2v from the RHV-M user interface, then behind the scenes the import is managed by VDSM using the I<-o vdsm> output mode (which end users should not try to use directly). -You have to specify I<-o rhev> and an I<-os> option that points to the -RHEV-M Export Storage Domain. You can either specify the NFS server -and mountpoint, eg. S<C<-os rhev-storage:/rhev/export>>, or you can +You have to specify I<-o rhv> and an I<-os> option that points to the +RHV-M Export Storage Domain. You can either specify the NFS server +and mountpoint, eg. S<C<-os rhv-storage:/rhv/export>>, or you can mount that first and point to the directory where it is mounted, eg. S<C<-os /tmp/mnt>>. Be careful not to point to the Data Storage Domain by accident as that will not work. On successful completion virt-v2v will have written the new guest to the Export Storage Domain, but it will not yet be ready to run. It -must be imported into RHEV using the UI before it can be used. +must be imported into RHV using the UI before it can be used. -In RHEV E<ge> 2.2 this is done from the Storage tab. Select the +In RHV E<ge> 2.2 this is done from the Storage tab. Select the export domain the guest was written to. A pane will appear underneath the storage domain list displaying several tabs, one of which is "VM Import". The converted guest will be listed here. Select the -appropriate guest an click "Import". See the RHEV documentation for +appropriate guest an click "Import". See the RHV documentation for additional details. If you export several guests, then you can import them all at the same @@ -1735,7 +1739,7 @@ require either root or a special user: =item Mounting the Export Storage Domain -When using I<-o rhev -os server:/esd> virt-v2v has to have sufficient +When using I<-o rhv -os server:/esd> virt-v2v has to have sufficient privileges to NFS mount the Export Storage Domain from C<server>. You can avoid needing root here by mounting it yourself before running @@ -1744,11 +1748,11 @@ read the next S<section ...> =item Writing to the Export Storage Domain as 36:36 -RHEV-M cannot read files and directories from the Export Storage +RHV-M cannot read files and directories from the Export Storage Domain unless they have UID:GID 36:36. You will see VM import problems if the UID:GID is not correct. -When you run virt-v2v I<-o rhev> as root, virt-v2v attempts to create +When you run virt-v2v I<-o rhv> as root, virt-v2v attempts to create files and directories with the correct ownership. If you run virt-v2v as non-root, it will probably still work, but you will need to manually change ownership after virt-v2v has finished. @@ -1774,19 +1778,19 @@ documentation. =back -=head1 DEBUGGING RHEV-M IMPORT FAILURES +=head1 DEBUGGING RHV-M IMPORT FAILURES -When you export to the RHEV-M Export Storage Domain, and then import -that guest through the RHEV-M UI, you may encounter an import failure. +When you export to the RHV-M Export Storage Domain, and then import +that guest through the RHV-M UI, you may encounter an import failure. Diagnosing these failures is infuriatingly difficult as the UI generally hides the true reason for the failure. There are two log files of interest. The first is stored on the -RHEV-M server itself, and is called +RHV-M server itself, and is called F</var/log/ovirt-engine/engine.log> The second file, which is the most useful, is found on the SPM host -(SPM stands for "Storage Pool Manager"). This is a RHEV node that is +(SPM stands for "Storage Pool Manager"). This is a RHV node that is elected to do all metadata modifications in the data center, such as image or snapshot creation. You can find out which host is the current SPM from the "Hosts" tab "Spm Status" column. Once you have @@ -1972,14 +1976,14 @@ I<--firstboot> or I<--firstboot-command> options with Windows guests. (Optional) -The RHEV Application Provisioning Tool (RHEV APT). If this file is +The RHV Application Provisioning Tool (RHEV APT). If this file is present, then RHEV APT will be installed in the Windows guest during conversion. This tool is a guest agent which ensures that the virtio drivers remain up to date when the guest is running on Red Hat -Enterprise Virtualization (RHEV). +Virtualization (RHV). -This file comes from Red Hat Enterprise Virtualization (RHEV), and is -not distributed with virt-v2v. +This file comes from Red Hat Virtualization (RHV), and is not +distributed with virt-v2v. =back @@ -2011,14 +2015,14 @@ conversion. Variously called C<engine-image-uploader>, C<ovirt-image-uploader> or C<rhevm-image-uploader>, this tool allows you to copy a guest from one -oVirt or RHEV Export Storage Domain to another. It only permits -importing a guest that was previously exported from another oVirt/RHEV +oVirt or RHV Export Storage Domain to another. It only permits +importing a guest that was previously exported from another oVirt/RHV instance. =item L<import-to-ovirt.pl|http://git.annexia.org/?p=import-to-ovirt.git> This script can be used to import guests that already run on KVM to -oVirt or RHEV. For more information, see this blog posting by the +oVirt or RHV. For more information, see this blog posting by the author of virt-v2v: L<https://rwmj.wordpress.com/2015/09/18/importing-kvm-guests-to-ovirt-or-rhev…> -- 2.10.2

2 6

[PATCH v2] v2v: Rename RHEV to RHV throughout.
by Richard W.M. Jones 07 Dec '16

07 Dec '16

v2: - Fix virt-p2v messages too. Rich.

2 2

[PATCH 0/5] Improve inspection of /usr filesystems
by Pino Toscano 07 Dec '16

07 Dec '16

Hi, this patch series improves the way /usr filesystems are handled: tag them appropriately, so later on we can find them and merge results they contain directly back for the root filesystem. The series includes also a new private debug API, and its usage to fix the resolution of /dev/mapper/.. devices found in fstab; without it, LVM /usr filesystems are not recognized as belonging to their roots. Maybe a better API for this could be added, but since it's something only related to the appliance then can stay internal for now. (Better suggestions always welcome, of course.) Thanks, Pino Toscano (5): inspect: change is_root flag into enum inspect: mark CoreOS /usr partitions with own USR role daemon: debug: new "exists" subcommand inspect: fix existance check of /dev/mapper devices inspect: gather info from /usr filesystems as well (RHBZ#1401474) daemon/debug.c | 39 ++++++++++++++++++++++ src/guestfs-internal.h | 13 ++++++-- src/inspect-fs-cd.c | 2 +- src/inspect-fs-unix.c | 42 ++++++++++++++++++++++- src/inspect-fs.c | 26 ++++++++------- src/inspect.c | 90 +++++++++++++++++++++++++++++++++++++++++++++----- 6 files changed, 187 insertions(+), 25 deletions(-) -- 2.7.4

2 10

libguestfs and exporting to OVA/OVF
by Emmanuel Kasper 06 Dec '16

06 Dec '16

Hi I've been looking for a standalone tool to create OVA/OVF VM files based on a disk image and found none. So I was thinking to write my own. Would you be interested in having such a tool in the libguestfs umbrella ? My aim is too have a free toolchain to build VirtualBox images for Debian (for Vagrant to be more exact, see https://wiki.debian.org/Teams/Cloud/VagrantBaseBoxes) Interface for such a tool would be something like disk2ova --memory 2G --controller sata --network E1000 myDiskImage.{raw,qcow2} based on the command line switches, a XML ovf would be created and packed with the disk image in a OVA file. There is some python code around which does that, for example the ganeti export code and imagefactory https://github.com/redhat-imaging/imagefactory/blob/master/imagefactory_plu… but nothing available as a standalone tool. I am afraid I can only contribute Perl code. Emmanuel

2 3

[PATCH] appliance: mount also /dev/pts
by Pino Toscano 06 Dec '16

06 Dec '16

Tools could require the use of pseudo-terminals, so make sure we have /dev/pts available in the appliance. The "command" API already bind-mounts it when running commands, so this is the only bit needed. --- appliance/init | 2 ++ 1 file changed, 2 insertions(+) diff --git a/appliance/init b/appliance/init index e678e42..8be27a2 100755 --- a/appliance/init +++ b/appliance/init @@ -72,6 +72,8 @@ fi # devtmpfs is required since udev 176 mount -t devtmpfs /dev /dev +mkdir -p /dev/pts +mount -t devpts /dev/pts /dev/pts if [[ $cmdline == *selinux=1* ]]; then mount -t selinuxfs none /sys/fs/selinux -- 2.7.4

2 1

[PATCH v2] v2v: -o vdsm: Add --vdsm-compat-11 flag, -o rhev: Drop support for RHV < 4.1 (RHBZ#1400205).
by Richard W.M. Jones 06 Dec '16

06 Dec '16

Support for RHEV with RHEL 6 nodes required us to output the old style qcow2 compat=0.10 images. Since RHEV 3.6 GA, RHEL 6 has not been supported as a RHEV node type. Since RHV 4.1, compat=1.1 is supported. Support for compat=1.1 is uncertain in RHV 4.0 even on RHEL 7 nodes. There are significant downsides to using qcow2 compat=0.10 instead of the modern default (compat=1.1). Therefore this patch does two things: For -o rhev, it drops support for compat=0.10 completely. You now must use RHV 4.1. For -o vdsm, it adds an interim flag (--vdsm-compat-11) which turns off the previous behaviour. We can drop this flag later when all RHV instances have moved to 4.1. It also adds: vdsm-compat-11-option to the `virt-v2v --machine-readable' output to indicate that this flag can be used. Thanks: Yaniv Kaul, Michal Skrivanek. --- v2v/cmdline.ml | 5 +++++ v2v/output_rhev.ml | 4 ---- v2v/output_vdsm.ml | 11 ++++++++--- v2v/output_vdsm.mli | 1 + v2v/virt-v2v.pod | 11 ++++++++++- 5 files changed, 24 insertions(+), 8 deletions(-) diff --git a/v2v/cmdline.ml b/v2v/cmdline.ml index 2d0a10a..9a56d60 100644 --- a/v2v/cmdline.ml +++ b/v2v/cmdline.ml @@ -65,6 +65,7 @@ let parse_cmdline () = let password_file = ref None in let vdsm_vm_uuid = ref None in let vdsm_ovf_output = ref None in (* default "." *) + let vdsm_compat_11 = ref false in let set_string_option_once optname optref arg = match !optref with @@ -198,6 +199,7 @@ let parse_cmdline () = [ L"print-source" ], Getopt.Set print_source, s_"Print source and stop"; [ L"qemu-boot" ], Getopt.Set qemu_boot, s_"Boot in qemu (-o qemu only)"; [ L"root" ], Getopt.String ("ask|... ", set_root_choice), s_"How to choose root filesystem"; + [ L"vdsm-compat-11" ], Getopt.Set vdsm_compat_11, s_"Write qcow2 with compat=1.1"; [ L"vdsm-image-uuid" ], Getopt.String ("uuid", add_vdsm_image_uuid), s_"Output image UUID(s)"; [ L"vdsm-vol-uuid" ], Getopt.String ("uuid", add_vdsm_vol_uuid), s_"Output vol UUID(s)"; [ L"vdsm-vm-uuid" ], Getopt.String ("uuid", set_string_option_once "--vdsm-vm-uuid" vdsm_vm_uuid), @@ -259,6 +261,7 @@ read the man page virt-v2v(1). let print_source = !print_source in let qemu_boot = !qemu_boot in let root_choice = !root_choice in + let vdsm_compat_11 = !vdsm_compat_11 in let vdsm_image_uuids = List.rev !vdsm_image_uuids in let vdsm_vol_uuids = List.rev !vdsm_vol_uuids in let vdsm_vm_uuid = !vdsm_vm_uuid in @@ -272,6 +275,7 @@ read the man page virt-v2v(1). printf "virt-v2v\n"; printf "libguestfs-rewrite\n"; printf "colours-option\n"; + printf "vdsm-compat-11-option\n"; List.iter (printf "input:%s\n") (Modules_list.input_modules ()); List.iter (printf "output:%s\n") (Modules_list.output_modules ()); List.iter (printf "convert:%s\n") (Modules_list.convert_modules ()); @@ -415,6 +419,7 @@ read the man page virt-v2v(1). vol_uuids = vdsm_vol_uuids; vm_uuid = vdsm_vm_uuid; ovf_output = vdsm_ovf_output; + compat_11 = vdsm_compat_11; } in Output_vdsm.output_vdsm os vdsm_params output_alloc in diff --git a/v2v/output_rhev.ml b/v2v/output_rhev.ml index e45043b..3280150 100644 --- a/v2v/output_rhev.ml +++ b/v2v/output_rhev.ml @@ -248,10 +248,6 @@ object Changeuid.func changeuid_t ( fun () -> let g = open_guestfs ~identifier:"rhev_disk_create" () in - (* For qcow2, override v2v-supplied compat option, because RHEL 6 - * nodes cannot handle qcow2 v3 (RHBZ#1145582). - *) - let compat = if format <> "qcow2" then compat else Some "0.10" in g#disk_create ?backingfile ?backingformat ?preallocation ?compat ?clustersize path format size; (* Make it sufficiently writable so that possibly root, or diff --git a/v2v/output_vdsm.ml b/v2v/output_vdsm.ml index 7cd94c0..fbddee8 100644 --- a/v2v/output_vdsm.ml +++ b/v2v/output_vdsm.ml @@ -30,6 +30,7 @@ type vdsm_params = { vol_uuids : string list; vm_uuid : string; ovf_output : string; + compat_11 : bool; } class output_vdsm os vdsm_params output_alloc = @@ -37,13 +38,14 @@ object inherit output method as_options = - sprintf "-o vdsm -os %s%s%s --vdsm-vm-uuid %s --vdsm-ovf-output %s" os + sprintf "-o vdsm -os %s%s%s --vdsm-vm-uuid %s --vdsm-ovf-output %s%s" os (String.concat "" (List.map (sprintf " --vdsm-image-uuid %s") vdsm_params.image_uuids)) (String.concat "" (List.map (sprintf " --vdsm-vol-uuid %s") vdsm_params.vol_uuids)) vdsm_params.vm_uuid vdsm_params.ovf_output + (if vdsm_params.compat_11 then " --vdsm-compat-11" else "") method supported_firmware = [ TargetBIOS ] @@ -149,9 +151,12 @@ object ?clustersize path format size = let g = open_guestfs ~identifier:"vdsm_disk_create" () in (* For qcow2, override v2v-supplied compat option, because RHEL 6 - * nodes cannot handle qcow2 v3 (RHBZ#1145582). + * nodes cannot handle qcow2 v3 (RHBZ#1145582). Using + * --vdsm-compat-11 disables the override. *) - let compat = if format <> "qcow2" then compat else Some "0.10" in + let compat = + if vdsm_params.compat_11 || format <> "qcow2" then compat + else Some "0.10" in g#disk_create ?backingfile ?backingformat ?preallocation ?compat ?clustersize path format size diff --git a/v2v/output_vdsm.mli b/v2v/output_vdsm.mli index 532227a..3b089c7 100644 --- a/v2v/output_vdsm.mli +++ b/v2v/output_vdsm.mli @@ -23,6 +23,7 @@ type vdsm_params = { vol_uuids : string list; (* --vdsm-vol-uuid (multiple) *) vm_uuid : string; (* --vdsm-vm-uuid *) ovf_output : string; (* --vdsm-ovf-output *) + compat_11 : bool; (* --vdsm-compat-11 *) } (** Miscellaneous extra command line parameters used by VDSM. *) diff --git a/v2v/virt-v2v.pod b/v2v/virt-v2v.pod index 4e0c65a..1654b35 100644 --- a/v2v/virt-v2v.pod +++ b/v2v/virt-v2v.pod @@ -199,7 +199,7 @@ QEMU and KVM only. =item OpenStack Glance -=item Red Hat Enterprise Virtualization (RHEV) 2.2 and up +=item Red Hat Enterprise Virtualization (RHEV) 4.1 and up =item Local libvirt @@ -604,6 +604,15 @@ boot an operating system from the first VirtIO disk. Specifically, F</boot> must be on the first VirtIO disk, and it cannot chainload an OS which is not in the first VirtIO disk. +=item B<--vdsm-compat-11> + +If I<-o vdsm> and the output format is qcow2, then by default the +I<compat=0.10> option is added to the output file for compatibility +with RHEL 6 (see L<https://bugzilla.redhat.com/1145582>). However if +this flag is given, then I<compat=1.1> is used instead (which is the +default for all other output modes). This flag will become the +default in a future version of virt-v2v. + =item B<--vdsm-image-uuid> UUID =item B<--vdsm-vol-uuid> UUID -- 2.10.2

2 1