+ + free(pool->pool_info.net.ip_end); + pool->pool_info.net.ip_end = strdup(val);
If you plan to keep the duplicated strdups, you should free the alloc'ed memory here.
Yuck.. this was a pretty sloppy patch set. Thanks for catching the strdup() bit!
+ + return msg;
This will always return a NULL value. But you probably add this to be set when address validation fails, right?
Right - it's possible that in the future one of these attributes (or a new attribute) will be mandatory.
+ + s = get_pool_by_name(_BROKER, reference, full_id, &inst); + if (s.rc == CMPI_RC_OK) { + cu_statusf(_BROKER, &s, + CMPI_RC_ERR_FAILED, + "Pool with that name already exists"); + goto out; + } + + pool->id = strdup(name);
You can use name directly, since the create_child_pool_parse_args already allocs memory for the name var. Or, if you decide to keep that way, free memory from the name var after this.
There should a function to cleanup the virt_pool struct. I'll add that.
+ + xml = pool_to_xml(pool); + if (xml == NULL) { + cu_statusf(_BROKER, &s, + CMPI_RC_ERR_FAILED, + "Unable to generate XML for resource pool"); + goto out; + } + + CU_DEBUG("Pool XML:\n%s", xml); + + inst = connect_and_create(xml, reference, full_id, pool->type, &s); + if (inst == NULL) { + cu_statusf(_BROKER, &s, + CMPI_RC_ERR_FAILED, + "Unable to create resource pool"); + goto out; + } + + result = CMGetObjectPath(inst, &s); + if ((result != NULL) && (s.rc == CMPI_RC_OK)) { + CMSetNameSpace(result, NAMESPACE(reference)); + CMAddArg(argsout, "Pool", &result, CMPI_ref); + } + + /* FIXME: Trigger indication here */ + + cu_statusf(_BROKER, &s, CMPI_RC_OK, ""); + out: + free(xml); + free(full_id);
Missing free(msg), free(pool) How about free(inst)?
msg is a const - no need to free it. Good call on freeing the pool though - there should be a function that handles cleaning up the pool struct elements. -- Kaitlin Rupert IBM Linux Technology Center kaitlin@linux.vnet.ibm.com