initial mgmt/monitoring service skeleton

diff --git a/lustre/mgmt/Makefile.am b/lustre/mgmt/Makefile.am
new file mode 100644 (file)
index 0000000..a170791
--- /dev/null
+++ b/lustre/mgmt/Makefile.am
@@ -0,0 +1,11 @@
+# Copyright (C) 2001  Cluster File Systems, Inc.
+#
+# This code is issued under the GNU General Public License.
+# See the file COPYING in this distribution
+
+DEFS=
+
+modulefs_DATA = mgmt_svc.o
+EXTRA_PROGRAMS = mgmt_svc
+
+include $(top_srcdir)/Rules

diff --git a/lustre/mgmt/mgmt_svc.c b/lustre/mgmt/mgmt_svc.c
new file mode 100644 (file)
index 0000000..fd59ba1
--- /dev/null
+++ b/lustre/mgmt/mgmt_svc.c
@@ -0,0 +1,156 @@
+/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*-
+ * vim:expandtab:shiftwidth=8:tabstop=8:
+ *
+ * Implementation of the management/health monitoring service.
+ *
+ *  Copyright (c) 2003 Cluster File Systems, Inc.
+ *   Author: Mike Shaver <shaver@clusterfs.com>
+ *
+ *   This file is part of Lustre, http://www.lustre.org.
+ *
+ *   Lustre is free software; you can redistribute it and/or
+ *   modify it under the terms of version 2 of the GNU General Public
+ *   License as published by the Free Software Foundation.
+ *
+ *   Lustre is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *   GNU General Public License for more details.
+ *
+ *   You should have received a copy of the GNU General Public License
+ *   along with Lustre; if not, write to the Free Software
+ *   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#define EXPORT_SYMTAB
+#define DEBUG_SUBSYSTEM S_RPC /* S_MGMT */
+#include <linux/module.h>
+#include <linux/init.h>
+
+#include <linux/obd_class.h>
+#include <linux/lustre_net.h>
+
+#define MGMT_NEVENTS     1024UL
+#define MGMT_NBUFS       128UL
+#define MGMT_BUFSIZE     8192
+#define MGMT_MAXREQSIZE  128
+#define MGMT_NUM_THREADS 4
+#define MGMT_DEVICE_NAME "mgmt"
+
+static int mgmt_initialized;
+static struct ptlrpc_service *mgmt_service;
+
+static int mgmt_connect(struct ptlrpc_request *req)
+{
+        int rc;
+        ENTRY;
+
+        rc = -EINVAL;
+        
+        RETURN(rc);
+}
+
+static int mgmt_ping(struct ptlrpc_request *req)
+{
+        return lustre_pack_msg(0, NULL, NULL, &req->rq_replen, &req->rq_repmsg);
+}
+
+static int mgmt_handler(struct ptlrpc_request *req)
+{
+        int rc;
+        ENTRY;
+
+        switch (req->rq_reqmsg->opc) {
+        case OBD_PING:
+                DEBUG_REQ(D_RPCTRACE, req, "ping");
+                rc = mgmt_ping(req);
+                break;
+        case MGMT_CONNECT:
+                DEBUG_REQ(D_RPCTRACE, req, "connect");
+                rc = mgmt_connect(req);
+                break;
+        default:
+                DEBUG_REQ(D_RPCTRACE, req, "UNKNOWN OP");
+                rc = -EINVAL;
+        }
+
+        if (rc)
+                ptlrpc_error(req);
+        else
+                ptlrpc_reply(req);
+
+        RETURN(0);
+}
+
+static int mgmt_setup(struct obd_device *obd, obd_count len, void *buf)
+{
+        int i, rc;
+        ENTRY;
+
+        if (mgmt_initialized)
+                RETURN(-EALREADY);
+        
+        mgmt_service = ptlrpc_init_svc(MGMT_NEVENTS, MGMT_NBUFS, MGMT_BUFSIZE,
+                                       MGMT_MAXREQSIZE, MGMT_REQUEST_PORTAL,
+                                       MGMT_REPLY_PORTAL, mgmt_handler,
+                                       "mgmt", obd);
+        if (!mgmt_service) {
+                CERROR("Failed to start mgmt service\n");
+                RETURN(-ENOMEM);
+        }
+
+        for (i = 0; i < MGMT_NUM_THREADS; i++) {
+                char name[32];
+                sprintf(name, "mgmt_%02d", i);
+                rc = ptlrpc_start_thread(obd, mgmt_service, name);
+                if (rc) {
+                        CERROR("failed to start mgmt thread %d: %d\n", i, rc);
+                        LBUG();
+                }
+        }
+
+        mgmt_initialized = 1;
+        
+        RETURN(0);
+}
+
+static int mgmt_cleanup(struct obd_device *obd, int force, int failover)
+{
+        ENTRY;
+        
+        if (!mgmt_initialized)
+                RETURN(-ENOENT);
+
+        ptlrpc_stop_all_threads(mgmt_service);
+        ptlrpc_unregister_service(mgmt_service);
+        
+        mgmt_initialized = 0;
+        RETURN(0);
+}
+
+static struct obd_ops mgmt_obd_ops = {
+        o_owner:   THIS_MODULE,
+        o_setup:   mgmt_setup,
+        o_cleanup: mgmt_cleanup
+};
+
+static int __init mgmt_init(void)
+{
+        int rc = class_register_type(&mgmt_obd_ops, 0, MGMT_DEVICE_NAME);
+
+        return rc;
+}
+
+static void __exit mgmt_exit(void)
+{
+        class_unregister_type(MGMT_DEVICE_NAME);
+}
+
+#ifdef __KERNEL__
+MODULE_AUTHOR("Cluster File Systems, Inc. <info@clusterfs.com>");
+MODULE_DESCRIPTION("Lustre monitoring service v0.1");
+MODULE_LICENSE("GPL");
+
+module_init(mgmt_init);
+module_exit(mgmt_exit);
+#endif