Qemu accepts gluster protocol as supported storage backend beside others. This patch allows users to specify disks on gluster backends like this: <disk type='network' device='disk'> <driver name='qemu' type='raw'/> <source protocol='gluster' name='volume/image'> <host name='example.org' port='6000' transport='socket'/> </source> <target dev='vda' bus='virtio'/> </disk> Note: In the <host> element above, transport is an optional attribute. Valid transport types for a network based disk can be socket, unix or rdma. TODO: - Add support for IPv6 format based server addr - Support for transport types other than socket. Signed-off-by: Harsh Prateek Bora <harsh(a)linux.vnet.ibm.com&gt; --- docs/schemas/domaincommon.rng | 8 +++ src/conf/domain_conf.c | 14 ++++- src/conf/domain_conf.h | 3 +- src/qemu/qemu_command.c | 123 ++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 145 insertions(+), 3 deletions(-) diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 145caf7..30c0d8c 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -1029,6 +1029,7 @@ <value>nbd</value> <value>rbd</value> <value>sheepdog</value> + <value>gluster</value> </choice> </attribute> <optional> @@ -1042,6 +1043,13 @@ <attribute name="port"> <ref name="unsignedInt"/> </attribute> + <attribute name="transport"> + <choice> + <value>socket</value> + <value>unix</value> + <value>rdma</value> + </choice> + </attribute> </element> </zeroOrMore> <empty/> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 419088c..c89035e 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -200,7 +200,8 @@ VIR_ENUM_IMPL(virDomainDiskErrorPolicy, VIR_DOMAIN_DISK_ERROR_POLICY_LAST, VIR_ENUM_IMPL(virDomainDiskProtocol, VIR_DOMAIN_DISK_PROTOCOL_LAST, "nbd", "rbd", - "sheepdog") + "sheepdog", + "gluster") VIR_ENUM_IMPL(virDomainDiskSecretType, VIR_DOMAIN_DISK_SECRET_TYPE_LAST, "none", @@ -994,6 +995,7 @@ void virDomainDiskHostDefFree(virDomainDiskHostDefPtr def) VIR_FREE(def->name); VIR_FREE(def->port); + VIR_FREE(def->transport); } void virDomainControllerDefFree(virDomainControllerDefPtr def) @@ -3489,6 +3491,7 @@ virDomainDiskDefParseXML(virCapsPtr caps, } hosts[nhosts].name = NULL; hosts[nhosts].port = NULL; + hosts[nhosts].transport = NULL; nhosts++; hosts[nhosts - 1].name = virXMLPropString(child, "name"); @@ -3503,6 +3506,8 @@ virDomainDiskDefParseXML(virCapsPtr caps, "%s", _("missing port for host")); goto error; } + /* transport can be socket, unix, rdma, etc. */ + hosts[nhosts - 1].transport = virXMLPropString(child, "transport"); } child = child->next; } @@ -11479,8 +11484,13 @@ virDomainDiskDefFormat(virBufferPtr buf, for (i = 0; i < def->nhosts; i++) { virBufferEscapeString(buf, " <host name='%s'", def->hosts[i].name); - virBufferEscapeString(buf, " port='%s'/>\n", + virBufferEscapeString(buf, " port='%s'", def->hosts[i].port); + if (def->hosts[i].transport) { + virBufferEscapeString(buf, " transport='%s'", + def->hosts[i].transport); + } + virBufferAddLit(buf, "/>\n"); } virBufferAddLit(buf, " </source>\n"); } diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 0c3824e..67e023f 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -442,7 +442,7 @@ enum virDomainDiskProtocol { VIR_DOMAIN_DISK_PROTOCOL_NBD, VIR_DOMAIN_DISK_PROTOCOL_RBD, VIR_DOMAIN_DISK_PROTOCOL_SHEEPDOG, - + VIR_DOMAIN_DISK_PROTOCOL_GLUSTER, VIR_DOMAIN_DISK_PROTOCOL_LAST }; @@ -467,6 +467,7 @@ typedef virDomainDiskHostDef *virDomainDiskHostDefPtr; struct _virDomainDiskHostDef { char *name; char *port; + char *transport; }; enum virDomainDiskIo { diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index ca62f0c..c8a0f27 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -2068,6 +2068,86 @@ no_memory: return -1; } +static int qemuParseGlusterString(virDomainDiskDefPtr def) +{ + char *port, *volimg, *transp, *marker; + + marker = strchr(def->src, ':'); + if (marker) { + /* port found */ + port = marker; + *port++ = '\0'; + marker = port; + } else { + /* port not given, assume port = 0 */ + port = NULL; + marker = def->src; + } + + volimg = strchr(marker, '/'); + if (!volimg) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("cannot parse gluster filename '%s'"), def->src); + return -1; + } + *volimg++ = '\0'; + transp = strchr(volimg, '?'); + if (transp) { + *transp++ = '\0'; + transp = strchr(transp, '='); + transp++; + } + if (VIR_ALLOC(def->hosts) < 0) { + virReportOOMError(); + return -1; + } + def->nhosts = 1; + def->hosts->name = def->src; + if (port) { + def->hosts->port = strdup(port); + } else { + def->hosts->port = strdup("0"); + } + if (transp) { + def->hosts->transport = strdup(transp); + if (!def->hosts->transport) { + virReportOOMError(); + return -1; + } + } else { + def->hosts->transport = NULL; + } + if (!def->hosts->port) { + virReportOOMError(); + return -1; + } + def->src = strdup(volimg); + if (!def->src) { + virReportOOMError(); + return -1; + } + + return 0; +} + +static int +qemuBuildGlusterString(virDomainDiskDefPtr disk, virBufferPtr opt) +{ + int ret = 0; + virBufferAddLit(opt, "file="); + if (disk->nhosts != 1) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("gluster accepts only one host")); + ret = -1; + } else { + virBufferAsprintf(opt, "gluster://%s:%s/%s", + disk->hosts->name, disk->hosts->port, disk->src); + if (disk->hosts->transport) + virBufferAsprintf(opt, "?transport=%s", disk->hosts->transport); + } + return ret; +} + char * qemuBuildDriveStr(virConnectPtr conn ATTRIBUTE_UNUSED, virDomainDiskDefPtr disk, @@ -2209,6 +2289,12 @@ qemuBuildDriveStr(virConnectPtr conn ATTRIBUTE_UNUSED, goto error; virBufferAddChar(&opt, ','); break; + case VIR_DOMAIN_DISK_PROTOCOL_GLUSTER: + if (qemuBuildGlusterString(disk, &opt) < 0) + goto error; + virBufferAddChar(&opt, ','); + break; + case VIR_DOMAIN_DISK_PROTOCOL_SHEEPDOG: if (disk->nhosts == 0) { virBufferEscape(&opt, ',', ",", "file=sheepdog:%s,", @@ -5135,6 +5221,18 @@ qemuBuildCommandLine(virConnectPtr conn, file = virBufferContentAndReset(&opt); } break; + case VIR_DOMAIN_DISK_PROTOCOL_GLUSTER: + { + virBuffer opt = VIR_BUFFER_INITIALIZER; + if (qemuBuildGlusterString(disk, &opt) < 0) + goto error; + if (virBufferError(&opt)) { + virReportOOMError(); + goto error; + } + file = virBufferContentAndReset(&opt); + } + break; case VIR_DOMAIN_DISK_PROTOCOL_SHEEPDOG: if (disk->nhosts == 0) { if (virAsprintf(&file, "sheepdog:%s,", disk->src) < 0) { @@ -6811,6 +6909,21 @@ qemuParseCommandLineDisk(virCapsPtr caps, goto cleanup; VIR_FREE(p); + } else if (STRPREFIX(def->src, "gluster://")) { + char *p = def->src; + + def->type = VIR_DOMAIN_DISK_TYPE_NETWORK; + def->protocol = VIR_DOMAIN_DISK_PROTOCOL_GLUSTER; + def->src = strdup(p + strlen("gluster://")); + if (!def->src) { + virReportOOMError(); + goto cleanup; + } + + if (qemuParseGlusterString(def) < 0) + goto cleanup; + + VIR_FREE(p); } else if (STRPREFIX(def->src, "sheepdog:")) { char *p = def->src; char *port, *vdi; @@ -7976,6 +8089,10 @@ virDomainDefPtr qemuParseCommandLine(virCapsPtr caps, disk->type = VIR_DOMAIN_DISK_TYPE_NETWORK; disk->protocol = VIR_DOMAIN_DISK_PROTOCOL_RBD; val += strlen("rbd:"); + } else if (STRPREFIX(val, "gluster://")) { + disk->type = VIR_DOMAIN_DISK_TYPE_NETWORK; + disk->protocol = VIR_DOMAIN_DISK_PROTOCOL_GLUSTER; + val += strlen("gluster://"); } else if (STRPREFIX(val, "sheepdog:")) { disk->type = VIR_DOMAIN_DISK_TYPE_NETWORK; disk->protocol = VIR_DOMAIN_DISK_PROTOCOL_SHEEPDOG; @@ -8061,6 +8178,12 @@ virDomainDefPtr qemuParseCommandLine(virCapsPtr caps, goto no_memory; } break; + case VIR_DOMAIN_DISK_PROTOCOL_GLUSTER: + + if (qemuParseGlusterString(disk) < 0) + goto error; + + break; } } -- 1.7.11.2

On Wed, Sep 5, 2012 at 7:03 PM, Jiri Denemark <jdenemar(a)redhat.com&gt; wrote: That is how gluster calls it and hence I am using the same in QEMU and the same is true here too. This is something for gluster developers to decide if they want to change socket to something more specific like tcp as you suggest. Regards, Bharata.

On Wed, Sep 5, 2012 at 8:45 PM, Eric Blake <eblake(a)redhat.com&gt; wrote: Vijay - would really like to have your inputs here... - While the transport-type for a volume is shown as tcp in "gluster volume info", libgfapi forces me to use transport=socket to access the same volume from QEMU. So does "socket" mean "tcp" really ? If so, should I just switch over to using transport=tcp from QEMU ? If not, can you explain a bit about the difference b/n socket and tcp transport types ? - Also apart from socket (or tcp ?), rdma and unix, are there any other transport options that QEMU should care about ? - Are rdma and unix transport types operational at the moment ? If not, do you see them being used in gluster any time in the future ? The reason behind asking this is to check if we are spending effort in defining semantics in QEMU for a transport type that is never going to be used in gluster. Also I see that "gluster volume create" supports tcp and rdma but doesn't list unix as an option. Regards, Bharata.