int drbd_adm_add_minor(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_delete_minor(struct sk_buff *skb, struct genl_info *info);
-int drbd_adm_create_connection(struct sk_buff *skb, struct genl_info *info);
-int drbd_adm_delete_connection(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_new_resource(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_del_resource(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_down(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_set_role(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_attach(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_detach(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_connect(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_net_opts(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_resize(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_start_ov(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_new_c_uuid(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_suspend_io(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_resume_io(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_outdate(struct sk_buff *skb, struct genl_info *info);
-int drbd_adm_syncer(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_resource_opts(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_get_status(struct sk_buff *skb, struct genl_info *info);
int drbd_adm_get_timeout_type(struct sk_buff *skb, struct genl_info *info);
/* .dumpit */
int drbd_adm_get_status_all(struct sk_buff *skb, struct netlink_callback *cb);
#include <linux/drbd_genl_api.h>
+#include "drbd_nla.h"
#include <linux/genl_magic_func.h>
/* used blkdev_get_by_path, to claim our meta data device(s) */
#define VOLUME_UNSPECIFIED (-1U)
/* pointer into the request skb,
* limited lifetime! */
- char *conn_name;
+ char *resource_name;
+ struct nlattr *my_addr;
+ struct nlattr *peer_addr;
/* reply buffer */
struct sk_buff *reply_skb;
* If it returns successfully, adm_ctx members are valid.
*/
#define DRBD_ADM_NEED_MINOR 1
-#define DRBD_ADM_NEED_CONN 2
+#define DRBD_ADM_NEED_RESOURCE 2
+#define DRBD_ADM_NEED_CONNECTION 4
static int drbd_adm_prepare(struct sk_buff *skb, struct genl_info *info,
unsigned flags)
{
memset(&adm_ctx, 0, sizeof(adm_ctx));
/* genl_rcv_msg only checks for CAP_NET_ADMIN on "GENL_ADMIN_PERM" :( */
- if (cmd != DRBD_ADM_GET_STATUS
- && security_netlink_recv(skb, CAP_SYS_ADMIN))
+ if (cmd != DRBD_ADM_GET_STATUS && !capable(CAP_NET_ADMIN))
return -EPERM;
adm_ctx.reply_skb = genlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
- if (!adm_ctx.reply_skb)
+ if (!adm_ctx.reply_skb) {
+ err = -ENOMEM;
goto fail;
+ }
adm_ctx.reply_dh = genlmsg_put_reply(adm_ctx.reply_skb,
info, &drbd_genl_family, 0, cmd);
/* put of a few bytes into a fresh skb of >= 4k will always succeed.
* but anyways */
- if (!adm_ctx.reply_dh)
+ if (!adm_ctx.reply_dh) {
+ err = -ENOMEM;
goto fail;
+ }
adm_ctx.reply_dh->minor = d_in->minor;
adm_ctx.reply_dh->ret_code = NO_ERROR;
+ adm_ctx.volume = VOLUME_UNSPECIFIED;
if (info->attrs[DRBD_NLA_CFG_CONTEXT]) {
struct nlattr *nla;
/* parse and validate only */
- err = drbd_cfg_context_from_attrs(NULL, info->attrs);
+ err = drbd_cfg_context_from_attrs(NULL, info);
if (err)
goto fail;
/* and assign stuff to the global adm_ctx */
nla = nested_attr_tb[__nla_type(T_ctx_volume)];
- adm_ctx.volume = nla ? nla_get_u32(nla) : VOLUME_UNSPECIFIED;
- nla = nested_attr_tb[__nla_type(T_ctx_conn_name)];
if (nla)
- adm_ctx.conn_name = nla_data(nla);
- } else
- adm_ctx.volume = VOLUME_UNSPECIFIED;
+ adm_ctx.volume = nla_get_u32(nla);
+ nla = nested_attr_tb[__nla_type(T_ctx_resource_name)];
+ if (nla)
+ adm_ctx.resource_name = nla_data(nla);
+ adm_ctx.my_addr = nested_attr_tb[__nla_type(T_ctx_my_addr)];
+ adm_ctx.peer_addr = nested_attr_tb[__nla_type(T_ctx_peer_addr)];
+ if ((adm_ctx.my_addr &&
+ nla_len(adm_ctx.my_addr) > sizeof(adm_ctx.tconn->my_addr)) ||
+ (adm_ctx.peer_addr &&
+ nla_len(adm_ctx.peer_addr) > sizeof(adm_ctx.tconn->peer_addr))) {
+ err = -EINVAL;
+ goto fail;
+ }
+ }
adm_ctx.minor = d_in->minor;
adm_ctx.mdev = minor_to_mdev(d_in->minor);
- adm_ctx.tconn = conn_by_name(adm_ctx.conn_name);
+ adm_ctx.tconn = conn_get_by_name(adm_ctx.resource_name);
if (!adm_ctx.mdev && (flags & DRBD_ADM_NEED_MINOR)) {
drbd_msg_put_info("unknown minor");
return ERR_MINOR_INVALID;
}
- if (!adm_ctx.tconn && (flags & DRBD_ADM_NEED_CONN)) {
- drbd_msg_put_info("unknown connection");
+ if (!adm_ctx.tconn && (flags & DRBD_ADM_NEED_RESOURCE)) {
+ drbd_msg_put_info("unknown resource");
return ERR_INVALID_REQUEST;
}
+ if (flags & DRBD_ADM_NEED_CONNECTION) {
+ if (adm_ctx.tconn && !(flags & DRBD_ADM_NEED_RESOURCE)) {
+ drbd_msg_put_info("no resource name expected");
+ return ERR_INVALID_REQUEST;
+ }
+ if (adm_ctx.mdev) {
+ drbd_msg_put_info("no minor number expected");
+ return ERR_INVALID_REQUEST;
+ }
+ if (adm_ctx.my_addr && adm_ctx.peer_addr)
+ adm_ctx.tconn = conn_get_by_addrs(nla_data(adm_ctx.my_addr),
+ nla_len(adm_ctx.my_addr),
+ nla_data(adm_ctx.peer_addr),
+ nla_len(adm_ctx.peer_addr));
+ if (!adm_ctx.tconn) {
+ drbd_msg_put_info("unknown connection");
+ return ERR_INVALID_REQUEST;
+ }
+ }
+
/* some more paranoia, if the request was over-determined */
+ if (adm_ctx.mdev && adm_ctx.tconn &&
+ adm_ctx.mdev->tconn != adm_ctx.tconn) {
+ pr_warning("request: minor=%u, resource=%s; but that minor belongs to connection %s\n",
+ adm_ctx.minor, adm_ctx.resource_name,
+ adm_ctx.mdev->tconn->name);
+ drbd_msg_put_info("minor exists in different resource");
+ return ERR_INVALID_REQUEST;
+ }
if (adm_ctx.mdev &&
adm_ctx.volume != VOLUME_UNSPECIFIED &&
adm_ctx.volume != adm_ctx.mdev->vnr) {
pr_warning("request: minor=%u, volume=%u; but that minor is volume %u in %s\n",
adm_ctx.minor, adm_ctx.volume,
adm_ctx.mdev->vnr, adm_ctx.mdev->tconn->name);
- drbd_msg_put_info("over-determined configuration context mismatch");
- return ERR_INVALID_REQUEST;
- }
- if (adm_ctx.mdev && adm_ctx.tconn &&
- adm_ctx.mdev->tconn != adm_ctx.tconn) {
- pr_warning("request: minor=%u, conn=%s; but that minor belongs to connection %s\n",
- adm_ctx.minor, adm_ctx.conn_name, adm_ctx.mdev->tconn->name);
- drbd_msg_put_info("over-determined configuration context mismatch");
+ drbd_msg_put_info("minor exists as different volume");
return ERR_INVALID_REQUEST;
}
+
return NO_ERROR;
fail:
nlmsg_free(adm_ctx.reply_skb);
adm_ctx.reply_skb = NULL;
- return -ENOMEM;
+ return err;
}
static int drbd_adm_finish(struct genl_info *info, int retcode)
{
- struct nlattr *nla;
- const char *conn_name = NULL;
+ if (adm_ctx.tconn) {
+ kref_put(&adm_ctx.tconn->kref, &conn_destroy);
+ adm_ctx.tconn = NULL;
+ }
if (!adm_ctx.reply_skb)
return -ENOMEM;
adm_ctx.reply_dh->ret_code = retcode;
-
- nla = info->attrs[DRBD_NLA_CFG_CONTEXT];
- if (nla) {
- nla = nla_find_nested(nla, __nla_type(T_ctx_conn_name));
- if (nla)
- conn_name = nla_data(nla);
- }
-
drbd_adm_send_reply(adm_ctx.reply_skb, info);
return 0;
}
+static void setup_khelper_env(struct drbd_tconn *tconn, char **envp)
+{
+ char *afs;
+
+ /* FIXME: A future version will not allow this case. */
+ if (tconn->my_addr_len == 0 || tconn->peer_addr_len == 0)
+ return;
+
+ switch (((struct sockaddr *)&tconn->peer_addr)->sa_family) {
+ case AF_INET6:
+ afs = "ipv6";
+ snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI6",
+ &((struct sockaddr_in6 *)&tconn->peer_addr)->sin6_addr);
+ break;
+ case AF_INET:
+ afs = "ipv4";
+ snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI4",
+ &((struct sockaddr_in *)&tconn->peer_addr)->sin_addr);
+ break;
+ default:
+ afs = "ssocks";
+ snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI4",
+ &((struct sockaddr_in *)&tconn->peer_addr)->sin_addr);
+ }
+ snprintf(envp[3], 20, "DRBD_PEER_AF=%s", afs);
+}
+
int drbd_khelper(struct drbd_conf *mdev, char *cmd)
{
char *envp[] = { "HOME=/",
"TERM=linux",
"PATH=/sbin:/usr/sbin:/bin:/usr/bin",
- NULL, /* Will be set to address family */
- NULL, /* Will be set to address */
+ (char[20]) { }, /* address family */
+ (char[60]) { }, /* address */
NULL };
- char mb[12], af[20], ad[60], *afs;
+ char mb[12];
char *argv[] = {usermode_helper, cmd, mb, NULL };
+ struct drbd_tconn *tconn = mdev->tconn;
struct sib_info sib;
int ret;
- snprintf(mb, 12, "minor-%d", mdev_to_minor(mdev));
+ if (current == tconn->worker.task)
+ set_bit(CALLBACK_PENDING, &tconn->flags);
- if (get_net_conf(mdev->tconn)) {
- switch (((struct sockaddr *)mdev->tconn->net_conf->peer_addr)->sa_family) {
- case AF_INET6:
- afs = "ipv6";
- snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI6",
- &((struct sockaddr_in6 *)mdev->tconn->net_conf->peer_addr)->sin6_addr);
- break;
- case AF_INET:
- afs = "ipv4";
- snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI4",
- &((struct sockaddr_in *)mdev->tconn->net_conf->peer_addr)->sin_addr);
- break;
- default:
- afs = "ssocks";
- snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI4",
- &((struct sockaddr_in *)mdev->tconn->net_conf->peer_addr)->sin_addr);
- }
- snprintf(af, 20, "DRBD_PEER_AF=%s", afs);
- envp[3]=af;
- envp[4]=ad;
- put_net_conf(mdev->tconn);
- }
+ snprintf(mb, 12, "minor-%d", mdev_to_minor(mdev));
+ setup_khelper_env(tconn, envp);
/* The helper may take some time.
* write out any unsynced meta data changes now */
sib.sib_reason = SIB_HELPER_PRE;
sib.helper_name = cmd;
drbd_bcast_event(mdev, &sib);
- ret = call_usermodehelper(usermode_helper, argv, envp, 1);
+ ret = call_usermodehelper(usermode_helper, argv, envp, UMH_WAIT_PROC);
if (ret)
dev_warn(DEV, "helper command: %s %s %s exit code %u (0x%x)\n",
usermode_helper, cmd, mb,
sib.helper_exit_code = ret;
drbd_bcast_event(mdev, &sib);
+ if (current == tconn->worker.task)
+ clear_bit(CALLBACK_PENDING, &tconn->flags);
+
+ if (ret < 0) /* Ignore any ERRNOs we got. */
+ ret = 0;
+
+ return ret;
+}
+
+int conn_khelper(struct drbd_tconn *tconn, char *cmd)
+{
+ char *envp[] = { "HOME=/",
+ "TERM=linux",
+ "PATH=/sbin:/usr/sbin:/bin:/usr/bin",
+ (char[20]) { }, /* address family */
+ (char[60]) { }, /* address */
+ NULL };
+ char *argv[] = {usermode_helper, cmd, tconn->name, NULL };
+ int ret;
+
+ setup_khelper_env(tconn, envp);
+ conn_md_sync(tconn);
+
+ conn_info(tconn, "helper command: %s %s %s\n", usermode_helper, cmd, tconn->name);
+ /* TODO: conn_bcast_event() ?? */
+
+ ret = call_usermodehelper(usermode_helper, argv, envp, UMH_WAIT_PROC);
+ if (ret)
+ conn_warn(tconn, "helper command: %s %s %s exit code %u (0x%x)\n",
+ usermode_helper, cmd, tconn->name,
+ (ret >> 8) & 0xff, ret);
+ else
+ conn_info(tconn, "helper command: %s %s %s exit code %u (0x%x)\n",
+ usermode_helper, cmd, tconn->name,
+ (ret >> 8) & 0xff, ret);
+ /* TODO: conn_bcast_event() ?? */
+
if (ret < 0) /* Ignore any ERRNOs we got. */
ret = 0;
return ret;
}
-enum drbd_disk_state drbd_try_outdate_peer(struct drbd_conf *mdev)
+static enum drbd_fencing_p highest_fencing_policy(struct drbd_tconn *tconn)
+{
+ enum drbd_fencing_p fp = FP_NOT_AVAIL;
+ struct drbd_conf *mdev;
+ int vnr;
+
+ rcu_read_lock();
+ idr_for_each_entry(&tconn->volumes, mdev, vnr) {
+ if (get_ldev_if_state(mdev, D_CONSISTENT)) {
+ fp = max_t(enum drbd_fencing_p, fp,
+ rcu_dereference(mdev->ldev->disk_conf)->fencing);
+ put_ldev(mdev);
+ }
+ }
+ rcu_read_unlock();
+
+ return fp;
+}
+
+bool conn_try_outdate_peer(struct drbd_tconn *tconn)
{
+ union drbd_state mask = { };
+ union drbd_state val = { };
+ enum drbd_fencing_p fp;
char *ex_to_string;
int r;
- enum drbd_disk_state nps;
- enum drbd_fencing_p fp;
- D_ASSERT(mdev->state.pdsk == D_UNKNOWN);
+ if (tconn->cstate >= C_WF_REPORT_PARAMS) {
+ conn_err(tconn, "Expected cstate < C_WF_REPORT_PARAMS\n");
+ return false;
+ }
- if (get_ldev_if_state(mdev, D_CONSISTENT)) {
- fp = mdev->ldev->dc.fencing;
- put_ldev(mdev);
- } else {
- dev_warn(DEV, "Not fencing peer, I'm not even Consistent myself.\n");
- nps = mdev->state.pdsk;
+ fp = highest_fencing_policy(tconn);
+ switch (fp) {
+ case FP_NOT_AVAIL:
+ conn_warn(tconn, "Not fencing peer, I'm not even Consistent myself.\n");
goto out;
+ case FP_DONT_CARE:
+ return true;
+ default: ;
}
- r = drbd_khelper(mdev, "fence-peer");
+ r = conn_khelper(tconn, "fence-peer");
switch ((r>>8) & 0xff) {
case 3: /* peer is inconsistent */
ex_to_string = "peer is inconsistent or worse";
- nps = D_INCONSISTENT;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_INCONSISTENT;
break;
case 4: /* peer got outdated, or was already outdated */
ex_to_string = "peer was fenced";
- nps = D_OUTDATED;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_OUTDATED;
break;
case 5: /* peer was down */
- if (mdev->state.disk == D_UP_TO_DATE) {
+ if (conn_highest_disk(tconn) == D_UP_TO_DATE) {
/* we will(have) create(d) a new UUID anyways... */
ex_to_string = "peer is unreachable, assumed to be dead";
- nps = D_OUTDATED;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_OUTDATED;
} else {
ex_to_string = "peer unreachable, doing nothing since disk != UpToDate";
- nps = mdev->state.pdsk;
}
break;
case 6: /* Peer is primary, voluntarily outdate myself.
* This is useful when an unconnected R_SECONDARY is asked to
* become R_PRIMARY, but finds the other peer being active. */
ex_to_string = "peer is active";
- dev_warn(DEV, "Peer is primary, outdating myself.\n");
- nps = D_UNKNOWN;
- _drbd_request_state(mdev, NS(disk, D_OUTDATED), CS_WAIT_COMPLETE);
+ conn_warn(tconn, "Peer is primary, outdating myself.\n");
+ mask.disk = D_MASK;
+ val.disk = D_OUTDATED;
break;
case 7:
if (fp != FP_STONITH)
- dev_err(DEV, "fence-peer() = 7 && fencing != Stonith !!!\n");
+ conn_err(tconn, "fence-peer() = 7 && fencing != Stonith !!!\n");
ex_to_string = "peer was stonithed";
- nps = D_OUTDATED;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_OUTDATED;
break;
default:
/* The script is broken ... */
- nps = D_UNKNOWN;
- dev_err(DEV, "fence-peer helper broken, returned %d\n", (r>>8)&0xff);
- return nps;
+ conn_err(tconn, "fence-peer helper broken, returned %d\n", (r>>8)&0xff);
+ return false; /* Eventually leave IO frozen */
}
- dev_info(DEV, "fence-peer helper returned %d (%s)\n",
- (r>>8) & 0xff, ex_to_string);
+ conn_info(tconn, "fence-peer helper returned %d (%s)\n",
+ (r>>8) & 0xff, ex_to_string);
-out:
- if (mdev->state.susp_fen && nps >= D_UNKNOWN) {
- /* The handler was not successful... unfreeze here, the
- state engine can not unfreeze... */
- _drbd_request_state(mdev, NS(susp_fen, 0), CS_VERBOSE);
- }
+ out:
+
+ /* Not using
+ conn_request_state(tconn, mask, val, CS_VERBOSE);
+ here, because we might were able to re-establish the connection in the
+ meantime. */
+ spin_lock_irq(&tconn->req_lock);
+ if (tconn->cstate < C_WF_REPORT_PARAMS && !test_bit(STATE_SENT, &tconn->flags))
+ _conn_request_state(tconn, mask, val, CS_VERBOSE);
+ spin_unlock_irq(&tconn->req_lock);
- return nps;
+ return conn_highest_pdsk(tconn) <= D_OUTDATED;
}
static int _try_outdate_peer_async(void *data)
{
- struct drbd_conf *mdev = (struct drbd_conf *)data;
- enum drbd_disk_state nps;
- union drbd_state ns;
+ struct drbd_tconn *tconn = (struct drbd_tconn *)data;
- nps = drbd_try_outdate_peer(mdev);
-
- /* Not using
- drbd_request_state(mdev, NS(pdsk, nps));
- here, because we might were able to re-establish the connection
- in the meantime. This can only partially be solved in the state's
- engine is_valid_state() and is_valid_state_transition()
- functions.
-
- nps can be D_INCONSISTENT, D_OUTDATED or D_UNKNOWN.
- pdsk == D_INCONSISTENT while conn >= C_CONNECTED is valid,
- therefore we have to have the pre state change check here.
- */
- spin_lock_irq(&mdev->tconn->req_lock);
- ns = mdev->state;
- if (ns.conn < C_WF_REPORT_PARAMS) {
- ns.pdsk = nps;
- _drbd_set_state(mdev, ns, CS_VERBOSE, NULL);
- }
- spin_unlock_irq(&mdev->tconn->req_lock);
+ conn_try_outdate_peer(tconn);
+ kref_put(&tconn->kref, &conn_destroy);
return 0;
}
-void drbd_try_outdate_peer_async(struct drbd_conf *mdev)
+void conn_try_outdate_peer_async(struct drbd_tconn *tconn)
{
struct task_struct *opa;
- opa = kthread_run(_try_outdate_peer_async, mdev, "drbd%d_a_helper", mdev_to_minor(mdev));
- if (IS_ERR(opa))
- dev_err(DEV, "out of mem, failed to invoke fence-peer helper\n");
+ kref_get(&tconn->kref);
+ opa = kthread_run(_try_outdate_peer_async, tconn, "drbd_async_h");
+ if (IS_ERR(opa)) {
+ conn_err(tconn, "out of mem, failed to invoke fence-peer helper\n");
+ kref_put(&tconn->kref, &conn_destroy);
+ }
}
enum drbd_state_rv
{
const int max_tries = 4;
enum drbd_state_rv rv = SS_UNKNOWN_ERROR;
+ struct net_conf *nc;
int try = 0;
int forced = 0;
union drbd_state mask, val;
- enum drbd_disk_state nps;
if (new_role == R_PRIMARY)
request_ping(mdev->tconn); /* Detect a dead peer ASAP */
if (rv == SS_NO_UP_TO_DATE_DISK &&
mdev->state.disk == D_CONSISTENT && mask.pdsk == 0) {
D_ASSERT(mdev->state.pdsk == D_UNKNOWN);
- nps = drbd_try_outdate_peer(mdev);
- if (nps == D_OUTDATED || nps == D_INCONSISTENT) {
+ if (conn_try_outdate_peer(mdev->tconn)) {
val.disk = D_UP_TO_DATE;
mask.disk = D_MASK;
}
-
- val.pdsk = nps;
- mask.pdsk = D_MASK;
-
continue;
}
if (rv == SS_NOTHING_TO_DO)
goto out;
if (rv == SS_PRIMARY_NOP && mask.pdsk == 0) {
- nps = drbd_try_outdate_peer(mdev);
-
- if (force && nps > D_OUTDATED) {
+ if (!conn_try_outdate_peer(mdev->tconn) && force) {
dev_warn(DEV, "Forced into split brain situation!\n");
- nps = D_OUTDATED;
- }
-
- mask.pdsk = D_MASK;
- val.pdsk = nps;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_OUTDATED;
+ }
continue;
}
if (rv == SS_TWO_PRIMARIES) {
/* Maybe the peer is detected as dead very soon...
retry at most once more in this case. */
- schedule_timeout_interruptible((mdev->tconn->net_conf->ping_timeo+1)*HZ/10);
+ int timeo;
+ rcu_read_lock();
+ nc = rcu_dereference(mdev->tconn->net_conf);
+ timeo = nc ? (nc->ping_timeo + 1) * HZ / 10 : 1;
+ rcu_read_unlock();
+ schedule_timeout_interruptible(timeo);
if (try < max_tries)
try = max_tries - 1;
continue;
/* Wait until nothing is on the fly :) */
wait_event(mdev->misc_wait, atomic_read(&mdev->ap_pending_cnt) == 0);
+ /* FIXME also wait for all pending P_BARRIER_ACK? */
+
if (new_role == R_SECONDARY) {
set_disk_ro(mdev->vdisk, true);
if (get_ldev(mdev)) {
put_ldev(mdev);
}
} else {
- if (get_net_conf(mdev->tconn)) {
- mdev->tconn->net_conf->want_lose = 0;
- put_net_conf(mdev->tconn);
- }
+ mutex_lock(&mdev->tconn->conf_update);
+ nc = mdev->tconn->net_conf;
+ if (nc)
+ nc->discard_my_data = 0; /* without copy; single bit op is atomic */
+ mutex_unlock(&mdev->tconn->conf_update);
+
set_disk_ro(mdev->vdisk, false);
if (get_ldev(mdev)) {
if (((mdev->state.conn < C_CONNECTED ||
/* if this was forced, we should consider sync */
if (forced)
drbd_send_uuids(mdev);
- drbd_send_state(mdev);
+ drbd_send_current_state(mdev);
}
drbd_md_sync(mdev);
{
return err == -ENOMSG ? "required attribute missing" :
err == -EOPNOTSUPP ? "unknown mandatory attribute" :
+ err == -EEXIST ? "can not change invariant setting" :
"invalid attribute value";
}
memset(&parms, 0, sizeof(parms));
if (info->attrs[DRBD_NLA_SET_ROLE_PARMS]) {
- err = set_role_parms_from_attrs(&parms, info->attrs);
+ err = set_role_parms_from_attrs(&parms, info);
if (err) {
retcode = ERR_MANDATORY_TAG;
drbd_msg_put_info(from_attrs_err_to_txt(err));
struct drbd_backing_dev *bdev)
{
sector_t md_size_sect = 0;
- switch (bdev->dc.meta_dev_idx) {
+ int meta_dev_idx;
+
+ rcu_read_lock();
+ meta_dev_idx = rcu_dereference(bdev->disk_conf)->meta_dev_idx;
+
+ switch (meta_dev_idx) {
default:
/* v07 style fixed size indexed meta data */
bdev->md.md_size_sect = MD_RESERVED_SECT;
bdev->md.bm_offset = -md_size_sect + MD_AL_OFFSET;
break;
}
+ rcu_read_unlock();
}
/* input size is expected to be in KB */
void drbd_suspend_io(struct drbd_conf *mdev)
{
set_bit(SUSPEND_IO, &mdev->flags);
- if (is_susp(mdev->state))
+ if (drbd_suspended(mdev))
return;
wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_bio_cnt));
}
enum determine_dev_size drbd_determine_dev_size(struct drbd_conf *mdev, enum dds_flags flags) __must_hold(local)
{
sector_t prev_first_sect, prev_size; /* previous meta location */
- sector_t la_size;
+ sector_t la_size, u_size;
sector_t size;
char ppb[10];
/* TODO: should only be some assert here, not (re)init... */
drbd_md_set_sector_offsets(mdev, mdev->ldev);
- size = drbd_new_dev_size(mdev, mdev->ldev, flags & DDSF_FORCED);
+ rcu_read_lock();
+ u_size = rcu_dereference(mdev->ldev->disk_conf)->disk_size;
+ rcu_read_unlock();
+ size = drbd_new_dev_size(mdev, mdev->ldev, u_size, flags & DDSF_FORCED);
if (drbd_get_capacity(mdev->this_bdev) != size ||
drbd_bm_capacity(mdev) != size) {
la_size_changed && md_moved ? "size changed and md moved" :
la_size_changed ? "size changed" : "md moved");
/* next line implicitly does drbd_suspend_io()+drbd_resume_io() */
- err = drbd_bitmap_io(mdev, &drbd_bm_write,
- "size changed", BM_LOCKED_MASK);
+ err = drbd_bitmap_io(mdev, md_moved ? &drbd_bm_write_all : &drbd_bm_write,
+ "size changed", BM_LOCKED_MASK);
if (err) {
rv = dev_size_error;
goto out;
}
sector_t
-drbd_new_dev_size(struct drbd_conf *mdev, struct drbd_backing_dev *bdev, int assume_peer_has_space)
+drbd_new_dev_size(struct drbd_conf *mdev, struct drbd_backing_dev *bdev,
+ sector_t u_size, int assume_peer_has_space)
{
sector_t p_size = mdev->p_size; /* partner's disk size. */
sector_t la_size = bdev->md.la_size_sect; /* last agreed size. */
sector_t m_size; /* my size */
- sector_t u_size = bdev->dc.disk_size; /* size requested by user. */
sector_t size = 0;
m_size = drbd_get_max_capacity(bdev);
* failed, and 0 on success. You should call drbd_md_sync() after you called
* this function.
*/
-static int drbd_check_al_size(struct drbd_conf *mdev)
+static int drbd_check_al_size(struct drbd_conf *mdev, struct disk_conf *dc)
{
struct lru_cache *n, *t;
struct lc_element *e;
unsigned int in_use;
int i;
- if (!expect(mdev->sync_conf.al_extents >= DRBD_AL_EXTENTS_MIN))
- mdev->sync_conf.al_extents = DRBD_AL_EXTENTS_MIN;
-
if (mdev->act_log &&
- mdev->act_log->nr_elements == mdev->sync_conf.al_extents)
+ mdev->act_log->nr_elements == dc->al_extents)
return 0;
in_use = 0;
t = mdev->act_log;
n = lc_create("act_log", drbd_al_ext_cache, AL_UPDATES_PER_TRANSACTION,
- mdev->sync_conf.al_extents, sizeof(struct lc_element), 0);
+ dc->al_extents, sizeof(struct lc_element), 0);
if (n == NULL) {
dev_err(DEV, "Cannot allocate act_log lru!\n");
static void drbd_setup_queue_param(struct drbd_conf *mdev, unsigned int max_bio_size)
{
struct request_queue * const q = mdev->rq_queue;
- int max_hw_sectors = max_bio_size >> 9;
- int max_segments = 0;
+ unsigned int max_hw_sectors = max_bio_size >> 9;
+ unsigned int max_segments = 0;
if (get_ldev_if_state(mdev, D_ATTACHING)) {
struct request_queue * const b = mdev->ldev->backing_bdev->bd_disk->queue;
max_hw_sectors = min(queue_max_hw_sectors(b), max_bio_size >> 9);
- max_segments = mdev->ldev->dc.max_bio_bvecs;
+ rcu_read_lock();
+ max_segments = rcu_dereference(mdev->ldev->disk_conf)->max_bio_bvecs;
+ rcu_read_unlock();
put_ldev(mdev);
}
void drbd_reconsider_max_bio_size(struct drbd_conf *mdev)
{
- int now, new, local, peer;
+ unsigned int now, new, local, peer;
now = queue_max_hw_sectors(mdev->rq_queue) << 9;
local = mdev->local_max_bio_size; /* Eventually last known value, from volatile memory */
mdev->local_max_bio_size = local;
put_ldev(mdev);
}
+ local = min(local, DRBD_MAX_BIO_SIZE);
/* We may ignore peer limits if the peer is modern enough.
Because new from 8.3.8 onwards the peer can use multiple
BIOs for a single peer_request */
if (mdev->state.conn >= C_CONNECTED) {
if (mdev->tconn->agreed_pro_version < 94)
- peer = mdev->peer_max_bio_size;
+ peer = min( mdev->peer_max_bio_size, DRBD_MAX_SIZE_H80_PACKET);
+ /* Correct old drbd (up to 8.3.7) if it believes it can do more than 32KiB */
else if (mdev->tconn->agreed_pro_version == 94)
peer = DRBD_MAX_SIZE_H80_PACKET;
- else /* drbd 8.3.8 onwards */
+ else if (mdev->tconn->agreed_pro_version < 100)
+ peer = DRBD_MAX_BIO_SIZE_P95; /* drbd 8.3.8 onwards, before 8.4.0 */
+ else
peer = DRBD_MAX_BIO_SIZE;
}
- new = min_t(int, local, peer);
+ new = min(local, peer);
if (mdev->state.role == R_PRIMARY && new < now)
- dev_err(DEV, "ASSERT FAILED new < now; (%d < %d)\n", new, now);
+ dev_err(DEV, "ASSERT FAILED new < now; (%u < %u)\n", new, now);
if (new != now)
dev_info(DEV, "max BIO size = %u\n", new);
drbd_setup_queue_param(mdev, new);
}
-/* serialize deconfig (worker exiting, doing cleanup)
- * and reconfig (drbdsetup disk, drbdsetup net)
- *
- * Wait for a potentially exiting worker, then restart it,
- * or start a new one. Flush any pending work, there may still be an
- * after_state_change queued.
- */
+/* Starts the worker thread */
static void conn_reconfig_start(struct drbd_tconn *tconn)
{
- wait_event(tconn->ping_wait, !test_and_set_bit(CONFIG_PENDING, &tconn->flags));
- wait_event(tconn->ping_wait, !test_bit(OBJECT_DYING, &tconn->flags));
drbd_thread_start(&tconn->worker);
conn_flush_workqueue(tconn);
}
-/* if still unconfigured, stops worker again.
- * if configured now, clears CONFIG_PENDING.
- * wakes potential waiters */
+/* if still unconfigured, stops worker again. */
static void conn_reconfig_done(struct drbd_tconn *tconn)
{
+ bool stop_threads;
spin_lock_irq(&tconn->req_lock);
- if (conn_all_vols_unconf(tconn)) {
- set_bit(OBJECT_DYING, &tconn->flags);
- drbd_thread_stop_nowait(&tconn->worker);
- } else
- clear_bit(CONFIG_PENDING, &tconn->flags);
+ stop_threads = conn_all_vols_unconf(tconn) &&
+ tconn->cstate == C_STANDALONE;
spin_unlock_irq(&tconn->req_lock);
- wake_up(&tconn->ping_wait);
+ if (stop_threads) {
+ /* asender is implicitly stopped by receiver
+ * in conn_disconnect() */
+ drbd_thread_stop(&tconn->receiver);
+ drbd_thread_stop(&tconn->worker);
+ }
}
/* Make sure IO is suspended before calling this function(). */
dev_info(DEV, "Suspended AL updates\n");
}
+
+static bool should_set_defaults(struct genl_info *info)
+{
+ unsigned flags = ((struct drbd_genlmsghdr*)info->userhdr)->flags;
+ return 0 != (flags & DRBD_GENL_F_SET_DEFAULTS);
+}
+
+static void enforce_disk_conf_limits(struct disk_conf *dc)
+{
+ if (dc->al_extents < DRBD_AL_EXTENTS_MIN)
+ dc->al_extents = DRBD_AL_EXTENTS_MIN;
+ if (dc->al_extents > DRBD_AL_EXTENTS_MAX)
+ dc->al_extents = DRBD_AL_EXTENTS_MAX;
+
+ if (dc->c_plan_ahead > DRBD_C_PLAN_AHEAD_MAX)
+ dc->c_plan_ahead = DRBD_C_PLAN_AHEAD_MAX;
+}
+
+int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info)
+{
+ enum drbd_ret_code retcode;
+ struct drbd_conf *mdev;
+ struct disk_conf *new_disk_conf, *old_disk_conf;
+ struct fifo_buffer *old_plan = NULL, *new_plan = NULL;
+ int err, fifo_size;
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ mdev = adm_ctx.mdev;
+
+ /* we also need a disk
+ * to change the options on */
+ if (!get_ldev(mdev)) {
+ retcode = ERR_NO_DISK;
+ goto out;
+ }
+
+ new_disk_conf = kmalloc(sizeof(struct disk_conf), GFP_KERNEL);
+ if (!new_disk_conf) {
+ retcode = ERR_NOMEM;
+ goto fail;
+ }
+
+ mutex_lock(&mdev->tconn->conf_update);
+ old_disk_conf = mdev->ldev->disk_conf;
+ *new_disk_conf = *old_disk_conf;
+ if (should_set_defaults(info))
+ set_disk_conf_defaults(new_disk_conf);
+
+ err = disk_conf_from_attrs_for_change(new_disk_conf, info);
+ if (err && err != -ENOMSG) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ }
+
+ if (!expect(new_disk_conf->resync_rate >= 1))
+ new_disk_conf->resync_rate = 1;
+
+ enforce_disk_conf_limits(new_disk_conf);
+
+ fifo_size = (new_disk_conf->c_plan_ahead * 10 * SLEEP_TIME) / HZ;
+ if (fifo_size != mdev->rs_plan_s->size) {
+ new_plan = fifo_alloc(fifo_size);
+ if (!new_plan) {
+ dev_err(DEV, "kmalloc of fifo_buffer failed");
+ retcode = ERR_NOMEM;
+ goto fail_unlock;
+ }
+ }
+
+ drbd_suspend_io(mdev);
+ wait_event(mdev->al_wait, lc_try_lock(mdev->act_log));
+ drbd_al_shrink(mdev);
+ err = drbd_check_al_size(mdev, new_disk_conf);
+ lc_unlock(mdev->act_log);
+ wake_up(&mdev->al_wait);
+ drbd_resume_io(mdev);
+
+ if (err) {
+ retcode = ERR_NOMEM;
+ goto fail_unlock;
+ }
+
+ write_lock_irq(&global_state_lock);
+ retcode = drbd_resync_after_valid(mdev, new_disk_conf->resync_after);
+ if (retcode == NO_ERROR) {
+ rcu_assign_pointer(mdev->ldev->disk_conf, new_disk_conf);
+ drbd_resync_after_changed(mdev);
+ }
+ write_unlock_irq(&global_state_lock);
+
+ if (retcode != NO_ERROR)
+ goto fail_unlock;
+
+ if (new_plan) {
+ old_plan = mdev->rs_plan_s;
+ rcu_assign_pointer(mdev->rs_plan_s, new_plan);
+ }
+
+ mutex_unlock(&mdev->tconn->conf_update);
+
+ if (new_disk_conf->al_updates)
+ mdev->ldev->md.flags &= ~MDF_AL_DISABLED;
+ else
+ mdev->ldev->md.flags |= MDF_AL_DISABLED;
+
+ if (new_disk_conf->md_flushes)
+ clear_bit(MD_NO_FUA, &mdev->flags);
+ else
+ set_bit(MD_NO_FUA, &mdev->flags);
+
+ drbd_bump_write_ordering(mdev->tconn, WO_bdev_flush);
+
+ drbd_md_sync(mdev);
+
+ if (mdev->state.conn >= C_CONNECTED)
+ drbd_send_sync_param(mdev);
+
+ synchronize_rcu();
+ kfree(old_disk_conf);
+ kfree(old_plan);
+ mod_timer(&mdev->request_timer, jiffies + HZ);
+ goto success;
+
+fail_unlock:
+ mutex_unlock(&mdev->tconn->conf_update);
+ fail:
+ kfree(new_disk_conf);
+ kfree(new_plan);
+success:
+ put_ldev(mdev);
+ out:
+ drbd_adm_finish(info, retcode);
+ return 0;
+}
+
int drbd_adm_attach(struct sk_buff *skb, struct genl_info *info)
{
struct drbd_conf *mdev;
sector_t max_possible_sectors;
sector_t min_md_device_sectors;
struct drbd_backing_dev *nbc = NULL; /* new_backing_conf */
+ struct disk_conf *new_disk_conf = NULL;
struct block_device *bdev;
struct lru_cache *resync_lru = NULL;
+ struct fifo_buffer *new_plan = NULL;
union drbd_state ns, os;
enum drbd_state_rv rv;
- int cp_discovered = 0;
+ struct net_conf *nc;
retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
if (!adm_ctx.reply_skb)
return retcode;
if (retcode != NO_ERROR)
- goto fail;
+ goto finish;
mdev = adm_ctx.mdev;
conn_reconfig_start(mdev->tconn);
* to realize a "hot spare" feature (not that I'd recommend that) */
wait_event(mdev->misc_wait, !atomic_read(&mdev->local_cnt));
+ /* make sure there is no leftover from previous force-detach attempts */
+ clear_bit(FORCE_DETACH, &mdev->flags);
+ clear_bit(WAS_IO_ERROR, &mdev->flags);
+ clear_bit(WAS_READ_ERROR, &mdev->flags);
+
+ /* and no leftover from previously aborted resync or verify, either */
+ mdev->rs_total = 0;
+ mdev->rs_failed = 0;
+ atomic_set(&mdev->rs_pending_cnt, 0);
+
/* allocation not in the IO path, drbdsetup context */
nbc = kzalloc(sizeof(struct drbd_backing_dev), GFP_KERNEL);
if (!nbc) {
retcode = ERR_NOMEM;
goto fail;
}
+ spin_lock_init(&nbc->md.uuid_lock);
- nbc->dc.disk_size = DRBD_DISK_SIZE_SECT_DEF;
- nbc->dc.on_io_error = DRBD_ON_IO_ERROR_DEF;
- nbc->dc.fencing = DRBD_FENCING_DEF;
- nbc->dc.max_bio_bvecs = DRBD_MAX_BIO_BVECS_DEF;
+ new_disk_conf = kzalloc(sizeof(struct disk_conf), GFP_KERNEL);
+ if (!new_disk_conf) {
+ retcode = ERR_NOMEM;
+ goto fail;
+ }
+ nbc->disk_conf = new_disk_conf;
- err = disk_conf_from_attrs(&nbc->dc, info->attrs);
+ set_disk_conf_defaults(new_disk_conf);
+ err = disk_conf_from_attrs(new_disk_conf, info);
if (err) {
retcode = ERR_MANDATORY_TAG;
drbd_msg_put_info(from_attrs_err_to_txt(err));
goto fail;
}
- if ((int)nbc->dc.meta_dev_idx < DRBD_MD_INDEX_FLEX_INT) {
+ enforce_disk_conf_limits(new_disk_conf);
+
+ new_plan = fifo_alloc((new_disk_conf->c_plan_ahead * 10 * SLEEP_TIME) / HZ);
+ if (!new_plan) {
+ retcode = ERR_NOMEM;
+ goto fail;
+ }
+
+ if (new_disk_conf->meta_dev_idx < DRBD_MD_INDEX_FLEX_INT) {
retcode = ERR_MD_IDX_INVALID;
goto fail;
}
- if (get_net_conf(mdev->tconn)) {
- int prot = mdev->tconn->net_conf->wire_protocol;
- put_net_conf(mdev->tconn);
- if (nbc->dc.fencing == FP_STONITH && prot == DRBD_PROT_A) {
+ rcu_read_lock();
+ nc = rcu_dereference(mdev->tconn->net_conf);
+ if (nc) {
+ if (new_disk_conf->fencing == FP_STONITH && nc->wire_protocol == DRBD_PROT_A) {
+ rcu_read_unlock();
retcode = ERR_STONITH_AND_PROT_A;
goto fail;
}
}
+ rcu_read_unlock();
- bdev = blkdev_get_by_path(nbc->dc.backing_dev,
+ bdev = blkdev_get_by_path(new_disk_conf->backing_dev,
FMODE_READ | FMODE_WRITE | FMODE_EXCL, mdev);
if (IS_ERR(bdev)) {
- dev_err(DEV, "open(\"%s\") failed with %ld\n", nbc->dc.backing_dev,
+ dev_err(DEV, "open(\"%s\") failed with %ld\n", new_disk_conf->backing_dev,
PTR_ERR(bdev));
retcode = ERR_OPEN_DISK;
goto fail;
* should check it for you already; but if you don't, or
* someone fooled it, we need to double check here)
*/
- bdev = blkdev_get_by_path(nbc->dc.meta_dev,
+ bdev = blkdev_get_by_path(new_disk_conf->meta_dev,
FMODE_READ | FMODE_WRITE | FMODE_EXCL,
- ((int)nbc->dc.meta_dev_idx < 0) ?
+ (new_disk_conf->meta_dev_idx < 0) ?
(void *)mdev : (void *)drbd_m_holder);
if (IS_ERR(bdev)) {
- dev_err(DEV, "open(\"%s\") failed with %ld\n", nbc->dc.meta_dev,
+ dev_err(DEV, "open(\"%s\") failed with %ld\n", new_disk_conf->meta_dev,
PTR_ERR(bdev));
retcode = ERR_OPEN_MD_DISK;
goto fail;
nbc->md_bdev = bdev;
if ((nbc->backing_bdev == nbc->md_bdev) !=
- (nbc->dc.meta_dev_idx == DRBD_MD_INDEX_INTERNAL ||
- nbc->dc.meta_dev_idx == DRBD_MD_INDEX_FLEX_INT)) {
+ (new_disk_conf->meta_dev_idx == DRBD_MD_INDEX_INTERNAL ||
+ new_disk_conf->meta_dev_idx == DRBD_MD_INDEX_FLEX_INT)) {
retcode = ERR_MD_IDX_INVALID;
goto fail;
}
/* RT - for drbd_get_max_capacity() DRBD_MD_INDEX_FLEX_INT */
drbd_md_set_sector_offsets(mdev, nbc);
- if (drbd_get_max_capacity(nbc) < nbc->dc.disk_size) {
+ if (drbd_get_max_capacity(nbc) < new_disk_conf->disk_size) {
dev_err(DEV, "max capacity %llu smaller than disk size %llu\n",
(unsigned long long) drbd_get_max_capacity(nbc),
- (unsigned long long) nbc->dc.disk_size);
- retcode = ERR_DISK_TO_SMALL;
+ (unsigned long long) new_disk_conf->disk_size);
+ retcode = ERR_DISK_TOO_SMALL;
goto fail;
}
- if ((int)nbc->dc.meta_dev_idx < 0) {
+ if (new_disk_conf->meta_dev_idx < 0) {
max_possible_sectors = DRBD_MAX_SECTORS_FLEX;
/* at least one MB, otherwise it does not make sense */
min_md_device_sectors = (2<<10);
} else {
max_possible_sectors = DRBD_MAX_SECTORS;
- min_md_device_sectors = MD_RESERVED_SECT * (nbc->dc.meta_dev_idx + 1);
+ min_md_device_sectors = MD_RESERVED_SECT * (new_disk_conf->meta_dev_idx + 1);
}
if (drbd_get_capacity(nbc->md_bdev) < min_md_device_sectors) {
- retcode = ERR_MD_DISK_TO_SMALL;
+ retcode = ERR_MD_DISK_TOO_SMALL;
dev_warn(DEV, "refusing attach: md-device too small, "
"at least %llu sectors needed for this meta-disk type\n",
(unsigned long long) min_md_device_sectors);
* (we may currently be R_PRIMARY with no local disk...) */
if (drbd_get_max_capacity(nbc) <
drbd_get_capacity(mdev->this_bdev)) {
- retcode = ERR_DISK_TO_SMALL;
+ retcode = ERR_DISK_TOO_SMALL;
goto fail;
}
dev_warn(DEV, "==> truncating very big lower level device "
"to currently maximum possible %llu sectors <==\n",
(unsigned long long) max_possible_sectors);
- if ((int)nbc->dc.meta_dev_idx >= 0)
+ if (new_disk_conf->meta_dev_idx >= 0)
dev_warn(DEV, "==>> using internal or flexible "
"meta data may help <<==\n");
}
drbd_suspend_io(mdev);
/* also wait for the last barrier ack. */
- wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt) || is_susp(mdev->state));
+ /* FIXME see also https://daiquiri.linbit/cgi-bin/bugzilla/show_bug.cgi?id=171
+ * We need a way to either ignore barrier acks for barriers sent before a device
+ * was attached, or a way to wait for all pending barrier acks to come in.
+ * As barriers are counted per resource,
+ * we'd need to suspend io on all devices of a resource.
+ */
+ wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt) || drbd_suspended(mdev));
/* and for any other previously queued work */
drbd_flush_workqueue(mdev);
}
/* Since we are diskless, fix the activity log first... */
- if (drbd_check_al_size(mdev)) {
+ if (drbd_check_al_size(mdev, new_disk_conf)) {
retcode = ERR_NOMEM;
goto force_diskless_dec;
}
/* Prevent shrinking of consistent devices ! */
if (drbd_md_test_flag(nbc, MDF_CONSISTENT) &&
- drbd_new_dev_size(mdev, nbc, 0) < nbc->md.la_size_sect) {
+ drbd_new_dev_size(mdev, nbc, nbc->disk_conf->disk_size, 0) < nbc->md.la_size_sect) {
dev_warn(DEV, "refusing to truncate a consistent device\n");
- retcode = ERR_DISK_TO_SMALL;
- goto force_diskless_dec;
- }
-
- if (!drbd_al_read_log(mdev, nbc)) {
- retcode = ERR_IO_MD_DISK;
+ retcode = ERR_DISK_TOO_SMALL;
goto force_diskless_dec;
}
/* Reset the "barriers don't work" bits here, then force meta data to
* be written, to ensure we determine if barriers are supported. */
- if (nbc->dc.no_md_flush)
- set_bit(MD_NO_FUA, &mdev->flags);
- else
+ if (new_disk_conf->md_flushes)
clear_bit(MD_NO_FUA, &mdev->flags);
+ else
+ set_bit(MD_NO_FUA, &mdev->flags);
/* Point of no return reached.
* Devices and memory are no longer released by error cleanup below.
D_ASSERT(mdev->ldev == NULL);
mdev->ldev = nbc;
mdev->resync = resync_lru;
+ mdev->rs_plan_s = new_plan;
nbc = NULL;
resync_lru = NULL;
+ new_disk_conf = NULL;
+ new_plan = NULL;
- mdev->write_ordering = WO_bdev_flush;
- drbd_bump_write_ordering(mdev, WO_bdev_flush);
+ drbd_bump_write_ordering(mdev->tconn, WO_bdev_flush);
if (drbd_md_test_flag(mdev->ldev, MDF_CRASHED_PRIMARY))
set_bit(CRASHED_PRIMARY, &mdev->flags);
clear_bit(CRASHED_PRIMARY, &mdev->flags);
if (drbd_md_test_flag(mdev->ldev, MDF_PRIMARY_IND) &&
- !(mdev->state.role == R_PRIMARY && mdev->state.susp_nod)) {
+ !(mdev->state.role == R_PRIMARY && mdev->tconn->susp_nod))
set_bit(CRASHED_PRIMARY, &mdev->flags);
- cp_discovered = 1;
- }
mdev->send_cnt = 0;
mdev->recv_cnt = 0;
} else if (dd == grew)
set_bit(RESYNC_AFTER_NEG, &mdev->flags);
- if (drbd_md_test_flag(mdev->ldev, MDF_FULL_SYNC)) {
+ if (drbd_md_test_flag(mdev->ldev, MDF_FULL_SYNC) ||
+ (test_bit(CRASHED_PRIMARY, &mdev->flags) &&
+ drbd_md_test_flag(mdev->ldev, MDF_AL_DISABLED))) {
dev_info(DEV, "Assuming that all blocks are out of sync "
"(aka FullSync)\n");
if (drbd_bitmap_io(mdev, &drbd_bmio_set_n_write,
}
} else {
if (drbd_bitmap_io(mdev, &drbd_bm_read,
- "read from attaching", BM_LOCKED_MASK) < 0) {
- retcode = ERR_IO_MD_DISK;
- goto force_diskless_dec;
- }
- }
-
- if (cp_discovered) {
- drbd_al_apply_to_bm(mdev);
- if (drbd_bitmap_io(mdev, &drbd_bm_write,
- "crashed primary apply AL", BM_LOCKED_MASK)) {
+ "read from attaching", BM_LOCKED_MASK)) {
retcode = ERR_IO_MD_DISK;
goto force_diskless_dec;
}
drbd_suspend_al(mdev); /* IO is still suspended here... */
spin_lock_irq(&mdev->tconn->req_lock);
- os = mdev->state;
- ns.i = os.i;
+ os = drbd_read_state(mdev);
+ ns = os;
/* If MDF_CONSISTENT is not set go into inconsistent state,
otherwise investigate MDF_WasUpToDate...
If MDF_WAS_UP_TO_DATE is not set go into D_OUTDATED disk state,
if (drbd_md_test_flag(mdev->ldev, MDF_PEER_OUT_DATED))
ns.pdsk = D_OUTDATED;
- if ( ns.disk == D_CONSISTENT &&
- (ns.pdsk == D_OUTDATED || mdev->ldev->dc.fencing == FP_DONT_CARE))
+ rcu_read_lock();
+ if (ns.disk == D_CONSISTENT &&
+ (ns.pdsk == D_OUTDATED || rcu_dereference(mdev->ldev->disk_conf)->fencing == FP_DONT_CARE))
ns.disk = D_UP_TO_DATE;
/* All tests on MDF_PRIMARY_IND, MDF_CONNECTED_IND,
this point, because drbd_request_state() modifies these
flags. */
+ if (rcu_dereference(mdev->ldev->disk_conf)->al_updates)
+ mdev->ldev->md.flags &= ~MDF_AL_DISABLED;
+ else
+ mdev->ldev->md.flags |= MDF_AL_DISABLED;
+
+ rcu_read_unlock();
+
/* In case we are C_CONNECTED postpone any decision on the new disk
state after the negotiation phase. */
if (mdev->state.conn == C_CONNECTED) {
}
rv = _drbd_set_state(mdev, ns, CS_VERBOSE, NULL);
- ns = mdev->state;
spin_unlock_irq(&mdev->tconn->req_lock);
if (rv < SS_SUCCESS)
goto force_diskless_dec;
+ mod_timer(&mdev->request_timer, jiffies + HZ);
+
if (mdev->state.role == R_PRIMARY)
mdev->ldev->md.uuid[UI_CURRENT] |= (u64)1;
else
force_diskless_dec:
put_ldev(mdev);
force_diskless:
- drbd_force_state(mdev, NS(disk, D_FAILED));
+ drbd_force_state(mdev, NS(disk, D_DISKLESS));
drbd_md_sync(mdev);
- conn_reconfig_done(mdev->tconn);
fail:
+ conn_reconfig_done(mdev->tconn);
if (nbc) {
if (nbc->backing_bdev)
blkdev_put(nbc->backing_bdev,
FMODE_READ | FMODE_WRITE | FMODE_EXCL);
kfree(nbc);
}
+ kfree(new_disk_conf);
lc_destroy(resync_lru);
+ kfree(new_plan);
+ finish:
drbd_adm_finish(info, retcode);
return 0;
}
+static int adm_detach(struct drbd_conf *mdev, int force)
+{
+ enum drbd_state_rv retcode;
+ int ret;
+
+ if (force) {
+ set_bit(FORCE_DETACH, &mdev->flags);
+ drbd_force_state(mdev, NS(disk, D_FAILED));
+ retcode = SS_SUCCESS;
+ goto out;
+ }
+
+ drbd_suspend_io(mdev); /* so no-one is stuck in drbd_al_begin_io */
+ drbd_md_get_buffer(mdev); /* make sure there is no in-flight meta-data IO */
+ retcode = drbd_request_state(mdev, NS(disk, D_FAILED));
+ drbd_md_put_buffer(mdev);
+ /* D_FAILED will transition to DISKLESS. */
+ ret = wait_event_interruptible(mdev->misc_wait,
+ mdev->state.disk != D_FAILED);
+ drbd_resume_io(mdev);
+ if ((int)retcode == (int)SS_IS_DISKLESS)
+ retcode = SS_NOTHING_TO_DO;
+ if (ret)
+ retcode = ERR_INTR;
+out:
+ return retcode;
+}
+
/* Detaching the disk is a process in multiple stages. First we need to lock
* out application IO, in-flight IO, IO stuck in drbd_al_begin_io.
* Then we transition to D_DISKLESS, and wait for put_ldev() to return all
* Only then we have finally detached. */
int drbd_adm_detach(struct sk_buff *skb, struct genl_info *info)
{
- struct drbd_conf *mdev;
enum drbd_ret_code retcode;
+ struct detach_parms parms = { };
+ int err;
retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
if (!adm_ctx.reply_skb)
if (retcode != NO_ERROR)
goto out;
- mdev = adm_ctx.mdev;
- drbd_suspend_io(mdev); /* so no-one is stuck in drbd_al_begin_io */
- retcode = drbd_request_state(mdev, NS(disk, D_DISKLESS));
- wait_event(mdev->misc_wait,
- mdev->state.disk != D_DISKLESS ||
- !atomic_read(&mdev->local_cnt));
- drbd_resume_io(mdev);
+ if (info->attrs[DRBD_NLA_DETACH_PARMS]) {
+ err = detach_parms_from_attrs(&parms, info);
+ if (err) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto out;
+ }
+ }
+
+ retcode = adm_detach(adm_ctx.mdev, parms.force_detach);
out:
drbd_adm_finish(info, retcode);
return 0;
}
-int drbd_adm_connect(struct sk_buff *skb, struct genl_info *info)
+static bool conn_resync_running(struct drbd_tconn *tconn)
{
- char hmac_name[CRYPTO_MAX_ALG_NAME];
struct drbd_conf *mdev;
- struct net_conf *new_conf = NULL;
- struct crypto_hash *tfm = NULL;
- struct crypto_hash *integrity_w_tfm = NULL;
- struct crypto_hash *integrity_r_tfm = NULL;
- void *int_dig_out = NULL;
- void *int_dig_in = NULL;
- void *int_dig_vv = NULL;
- struct drbd_tconn *oconn;
- struct drbd_tconn *tconn;
- struct sockaddr *new_my_addr, *new_peer_addr, *taken_addr;
- enum drbd_ret_code retcode;
+ bool rv = false;
+ int vnr;
+
+ rcu_read_lock();
+ idr_for_each_entry(&tconn->volumes, mdev, vnr) {
+ if (mdev->state.conn == C_SYNC_SOURCE ||
+ mdev->state.conn == C_SYNC_TARGET ||
+ mdev->state.conn == C_PAUSED_SYNC_S ||
+ mdev->state.conn == C_PAUSED_SYNC_T) {
+ rv = true;
+ break;
+ }
+ }
+ rcu_read_unlock();
+
+ return rv;
+}
+
+static bool conn_ov_running(struct drbd_tconn *tconn)
+{
+ struct drbd_conf *mdev;
+ bool rv = false;
+ int vnr;
+
+ rcu_read_lock();
+ idr_for_each_entry(&tconn->volumes, mdev, vnr) {
+ if (mdev->state.conn == C_VERIFY_S ||
+ mdev->state.conn == C_VERIFY_T) {
+ rv = true;
+ break;
+ }
+ }
+ rcu_read_unlock();
+
+ return rv;
+}
+
+static enum drbd_ret_code
+_check_net_options(struct drbd_tconn *tconn, struct net_conf *old_conf, struct net_conf *new_conf)
+{
+ struct drbd_conf *mdev;
+ int i;
+
+ if (old_conf && tconn->cstate == C_WF_REPORT_PARAMS && tconn->agreed_pro_version < 100) {
+ if (new_conf->wire_protocol != old_conf->wire_protocol)
+ return ERR_NEED_APV_100;
+
+ if (new_conf->two_primaries != old_conf->two_primaries)
+ return ERR_NEED_APV_100;
+
+ if (strcmp(new_conf->integrity_alg, old_conf->integrity_alg))
+ return ERR_NEED_APV_100;
+ }
+
+ if (!new_conf->two_primaries &&
+ conn_highest_role(tconn) == R_PRIMARY &&
+ conn_highest_peer(tconn) == R_PRIMARY)
+ return ERR_NEED_ALLOW_TWO_PRI;
+
+ if (new_conf->two_primaries &&
+ (new_conf->wire_protocol != DRBD_PROT_C))
+ return ERR_NOT_PROTO_C;
+
+ idr_for_each_entry(&tconn->volumes, mdev, i) {
+ if (get_ldev(mdev)) {
+ enum drbd_fencing_p fp = rcu_dereference(mdev->ldev->disk_conf)->fencing;
+ put_ldev(mdev);
+ if (new_conf->wire_protocol == DRBD_PROT_A && fp == FP_STONITH)
+ return ERR_STONITH_AND_PROT_A;
+ }
+ if (mdev->state.role == R_PRIMARY && new_conf->discard_my_data)
+ return ERR_DISCARD_IMPOSSIBLE;
+ }
+
+ if (new_conf->on_congestion != OC_BLOCK && new_conf->wire_protocol != DRBD_PROT_A)
+ return ERR_CONG_NOT_PROTO_A;
+
+ return NO_ERROR;
+}
+
+static enum drbd_ret_code
+check_net_options(struct drbd_tconn *tconn, struct net_conf *new_conf)
+{
+ static enum drbd_ret_code rv;
+ struct drbd_conf *mdev;
int i;
+
+ rcu_read_lock();
+ rv = _check_net_options(tconn, rcu_dereference(tconn->net_conf), new_conf);
+ rcu_read_unlock();
+
+ /* tconn->volumes protected by genl_lock() here */
+ idr_for_each_entry(&tconn->volumes, mdev, i) {
+ if (!mdev->bitmap) {
+ if(drbd_bm_init(mdev))
+ return ERR_NOMEM;
+ }
+ }
+
+ return rv;
+}
+
+struct crypto {
+ struct crypto_hash *verify_tfm;
+ struct crypto_hash *csums_tfm;
+ struct crypto_hash *cram_hmac_tfm;
+ struct crypto_hash *integrity_tfm;
+};
+
+static int
+alloc_hash(struct crypto_hash **tfm, char *tfm_name, int err_alg)
+{
+ if (!tfm_name[0])
+ return NO_ERROR;
+
+ *tfm = crypto_alloc_hash(tfm_name, 0, CRYPTO_ALG_ASYNC);
+ if (IS_ERR(*tfm)) {
+ *tfm = NULL;
+ return err_alg;
+ }
+
+ return NO_ERROR;
+}
+
+static enum drbd_ret_code
+alloc_crypto(struct crypto *crypto, struct net_conf *new_conf)
+{
+ char hmac_name[CRYPTO_MAX_ALG_NAME];
+ enum drbd_ret_code rv;
+
+ rv = alloc_hash(&crypto->csums_tfm, new_conf->csums_alg,
+ ERR_CSUMS_ALG);
+ if (rv != NO_ERROR)
+ return rv;
+ rv = alloc_hash(&crypto->verify_tfm, new_conf->verify_alg,
+ ERR_VERIFY_ALG);
+ if (rv != NO_ERROR)
+ return rv;
+ rv = alloc_hash(&crypto->integrity_tfm, new_conf->integrity_alg,
+ ERR_INTEGRITY_ALG);
+ if (rv != NO_ERROR)
+ return rv;
+ if (new_conf->cram_hmac_alg[0] != 0) {
+ snprintf(hmac_name, CRYPTO_MAX_ALG_NAME, "hmac(%s)",
+ new_conf->cram_hmac_alg);
+
+ rv = alloc_hash(&crypto->cram_hmac_tfm, hmac_name,
+ ERR_AUTH_ALG);
+ }
+
+ return rv;
+}
+
+static void free_crypto(struct crypto *crypto)
+{
+ crypto_free_hash(crypto->cram_hmac_tfm);
+ crypto_free_hash(crypto->integrity_tfm);
+ crypto_free_hash(crypto->csums_tfm);
+ crypto_free_hash(crypto->verify_tfm);
+}
+
+int drbd_adm_net_opts(struct sk_buff *skb, struct genl_info *info)
+{
+ enum drbd_ret_code retcode;
+ struct drbd_tconn *tconn;
+ struct net_conf *old_conf, *new_conf = NULL;
int err;
+ int ovr; /* online verify running */
+ int rsr; /* re-sync running */
+ struct crypto crypto = { };
- retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONNECTION);
if (!adm_ctx.reply_skb)
return retcode;
if (retcode != NO_ERROR)
goto out;
tconn = adm_ctx.tconn;
- conn_reconfig_start(tconn);
-
- if (tconn->cstate > C_STANDALONE) {
- retcode = ERR_NET_CONFIGURED;
- goto fail;
- }
- /* allocation not in the IO path, cqueue thread context */
new_conf = kzalloc(sizeof(struct net_conf), GFP_KERNEL);
if (!new_conf) {
retcode = ERR_NOMEM;
+ goto out;
+ }
+
+ conn_reconfig_start(tconn);
+
+ mutex_lock(&tconn->data.mutex);
+ mutex_lock(&tconn->conf_update);
+ old_conf = tconn->net_conf;
+
+ if (!old_conf) {
+ drbd_msg_put_info("net conf missing, try connect");
+ retcode = ERR_INVALID_REQUEST;
goto fail;
}
- new_conf->timeout = DRBD_TIMEOUT_DEF;
- new_conf->try_connect_int = DRBD_CONNECT_INT_DEF;
- new_conf->ping_int = DRBD_PING_INT_DEF;
- new_conf->max_epoch_size = DRBD_MAX_EPOCH_SIZE_DEF;
- new_conf->max_buffers = DRBD_MAX_BUFFERS_DEF;
- new_conf->unplug_watermark = DRBD_UNPLUG_WATERMARK_DEF;
- new_conf->sndbuf_size = DRBD_SNDBUF_SIZE_DEF;
- new_conf->rcvbuf_size = DRBD_RCVBUF_SIZE_DEF;
- new_conf->ko_count = DRBD_KO_COUNT_DEF;
- new_conf->after_sb_0p = DRBD_AFTER_SB_0P_DEF;
- new_conf->after_sb_1p = DRBD_AFTER_SB_1P_DEF;
- new_conf->after_sb_2p = DRBD_AFTER_SB_2P_DEF;
- new_conf->want_lose = 0;
- new_conf->two_primaries = 0;
- new_conf->wire_protocol = DRBD_PROT_C;
- new_conf->ping_timeo = DRBD_PING_TIMEO_DEF;
- new_conf->rr_conflict = DRBD_RR_CONFLICT_DEF;
- new_conf->on_congestion = DRBD_ON_CONGESTION_DEF;
- new_conf->cong_extents = DRBD_CONG_EXTENTS_DEF;
-
- err = net_conf_from_attrs(new_conf, info->attrs);
- if (err) {
+ *new_conf = *old_conf;
+ if (should_set_defaults(info))
+ set_net_conf_defaults(new_conf);
+
+ err = net_conf_from_attrs_for_change(new_conf, info);
+ if (err && err != -ENOMSG) {
retcode = ERR_MANDATORY_TAG;
drbd_msg_put_info(from_attrs_err_to_txt(err));
goto fail;
}
- if (new_conf->two_primaries
- && (new_conf->wire_protocol != DRBD_PROT_C)) {
- retcode = ERR_NOT_PROTO_C;
+ retcode = check_net_options(tconn, new_conf);
+ if (retcode != NO_ERROR)
+ goto fail;
+
+ /* re-sync running */
+ rsr = conn_resync_running(tconn);
+ if (rsr && strcmp(new_conf->csums_alg, old_conf->csums_alg)) {
+ retcode = ERR_CSUMS_RESYNC_RUNNING;
goto fail;
}
- idr_for_each_entry(&tconn->volumes, mdev, i) {
- if (get_ldev(mdev)) {
- enum drbd_fencing_p fp = mdev->ldev->dc.fencing;
- put_ldev(mdev);
- if (new_conf->wire_protocol == DRBD_PROT_A && fp == FP_STONITH) {
- retcode = ERR_STONITH_AND_PROT_A;
- goto fail;
- }
- }
- if (mdev->state.role == R_PRIMARY && new_conf->want_lose) {
- retcode = ERR_DISCARD;
- goto fail;
- }
- if (!mdev->bitmap) {
- if(drbd_bm_init(mdev)) {
- retcode = ERR_NOMEM;
- goto fail;
- }
- }
+ /* online verify running */
+ ovr = conn_ov_running(tconn);
+ if (ovr && strcmp(new_conf->verify_alg, old_conf->verify_alg)) {
+ retcode = ERR_VERIFY_RUNNING;
+ goto fail;
}
- if (new_conf->on_congestion != OC_BLOCK && new_conf->wire_protocol != DRBD_PROT_A) {
- retcode = ERR_CONG_NOT_PROTO_A;
+ retcode = alloc_crypto(&crypto, new_conf);
+ if (retcode != NO_ERROR)
goto fail;
+
+ rcu_assign_pointer(tconn->net_conf, new_conf);
+
+ if (!rsr) {
+ crypto_free_hash(tconn->csums_tfm);
+ tconn->csums_tfm = crypto.csums_tfm;
+ crypto.csums_tfm = NULL;
+ }
+ if (!ovr) {
+ crypto_free_hash(tconn->verify_tfm);
+ tconn->verify_tfm = crypto.verify_tfm;
+ crypto.verify_tfm = NULL;
}
- retcode = NO_ERROR;
+ crypto_free_hash(tconn->integrity_tfm);
+ tconn->integrity_tfm = crypto.integrity_tfm;
+ if (tconn->cstate >= C_WF_REPORT_PARAMS && tconn->agreed_pro_version >= 100)
+ /* Do this without trying to take tconn->data.mutex again. */
+ __drbd_send_protocol(tconn, P_PROTOCOL_UPDATE);
- new_my_addr = (struct sockaddr *)&new_conf->my_addr;
- new_peer_addr = (struct sockaddr *)&new_conf->peer_addr;
- list_for_each_entry(oconn, &drbd_tconns, all_tconn) {
- if (oconn == tconn)
- continue;
- if (get_net_conf(oconn)) {
- taken_addr = (struct sockaddr *)&oconn->net_conf->my_addr;
- if (new_conf->my_addr_len == oconn->net_conf->my_addr_len &&
- !memcmp(new_my_addr, taken_addr, new_conf->my_addr_len))
- retcode = ERR_LOCAL_ADDR;
-
- taken_addr = (struct sockaddr *)&oconn->net_conf->peer_addr;
- if (new_conf->peer_addr_len == oconn->net_conf->peer_addr_len &&
- !memcmp(new_peer_addr, taken_addr, new_conf->peer_addr_len))
- retcode = ERR_PEER_ADDR;
-
- put_net_conf(oconn);
- if (retcode != NO_ERROR)
- goto fail;
- }
+ crypto_free_hash(tconn->cram_hmac_tfm);
+ tconn->cram_hmac_tfm = crypto.cram_hmac_tfm;
+
+ mutex_unlock(&tconn->conf_update);
+ mutex_unlock(&tconn->data.mutex);
+ synchronize_rcu();
+ kfree(old_conf);
+
+ if (tconn->cstate >= C_WF_REPORT_PARAMS)
+ drbd_send_sync_param(minor_to_mdev(conn_lowest_minor(tconn)));
+
+ goto done;
+
+ fail:
+ mutex_unlock(&tconn->conf_update);
+ mutex_unlock(&tconn->data.mutex);
+ free_crypto(&crypto);
+ kfree(new_conf);
+ done:
+ conn_reconfig_done(tconn);
+ out:
+ drbd_adm_finish(info, retcode);
+ return 0;
+}
+
+int drbd_adm_connect(struct sk_buff *skb, struct genl_info *info)
+{
+ struct drbd_conf *mdev;
+ struct net_conf *old_conf, *new_conf = NULL;
+ struct crypto crypto = { };
+ struct drbd_tconn *tconn;
+ enum drbd_ret_code retcode;
+ int i;
+ int err;
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_RESOURCE);
+
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+ if (!(adm_ctx.my_addr && adm_ctx.peer_addr)) {
+ drbd_msg_put_info("connection endpoint(s) missing");
+ retcode = ERR_INVALID_REQUEST;
+ goto out;
}
- if (new_conf->cram_hmac_alg[0] != 0) {
- snprintf(hmac_name, CRYPTO_MAX_ALG_NAME, "hmac(%s)",
- new_conf->cram_hmac_alg);
- tfm = crypto_alloc_hash(hmac_name, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(tfm)) {
- tfm = NULL;
- retcode = ERR_AUTH_ALG;
- goto fail;
+ /* No need for _rcu here. All reconfiguration is
+ * strictly serialized on genl_lock(). We are protected against
+ * concurrent reconfiguration/addition/deletion */
+ list_for_each_entry(tconn, &drbd_tconns, all_tconn) {
+ if (nla_len(adm_ctx.my_addr) == tconn->my_addr_len &&
+ !memcmp(nla_data(adm_ctx.my_addr), &tconn->my_addr, tconn->my_addr_len)) {
+ retcode = ERR_LOCAL_ADDR;
+ goto out;
}
- if (!drbd_crypto_is_hash(crypto_hash_tfm(tfm))) {
- retcode = ERR_AUTH_ALG_ND;
- goto fail;
+ if (nla_len(adm_ctx.peer_addr) == tconn->peer_addr_len &&
+ !memcmp(nla_data(adm_ctx.peer_addr), &tconn->peer_addr, tconn->peer_addr_len)) {
+ retcode = ERR_PEER_ADDR;
+ goto out;
}
}
- if (new_conf->integrity_alg[0]) {
- integrity_w_tfm = crypto_alloc_hash(new_conf->integrity_alg, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(integrity_w_tfm)) {
- integrity_w_tfm = NULL;
- retcode=ERR_INTEGRITY_ALG;
- goto fail;
- }
+ tconn = adm_ctx.tconn;
+ conn_reconfig_start(tconn);
- if (!drbd_crypto_is_hash(crypto_hash_tfm(integrity_w_tfm))) {
- retcode=ERR_INTEGRITY_ALG_ND;
- goto fail;
- }
+ if (tconn->cstate > C_STANDALONE) {
+ retcode = ERR_NET_CONFIGURED;
+ goto fail;
+ }
- integrity_r_tfm = crypto_alloc_hash(new_conf->integrity_alg, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(integrity_r_tfm)) {
- integrity_r_tfm = NULL;
- retcode=ERR_INTEGRITY_ALG;
- goto fail;
- }
+ /* allocation not in the IO path, drbdsetup / netlink process context */
+ new_conf = kzalloc(sizeof(*new_conf), GFP_KERNEL);
+ if (!new_conf) {
+ retcode = ERR_NOMEM;
+ goto fail;
}
- ((char *)new_conf->shared_secret)[SHARED_SECRET_MAX-1] = 0;
+ set_net_conf_defaults(new_conf);
- /* allocation not in the IO path, cqueue thread context */
- if (integrity_w_tfm) {
- i = crypto_hash_digestsize(integrity_w_tfm);
- int_dig_out = kmalloc(i, GFP_KERNEL);
- if (!int_dig_out) {
- retcode = ERR_NOMEM;
- goto fail;
- }
- int_dig_in = kmalloc(i, GFP_KERNEL);
- if (!int_dig_in) {
- retcode = ERR_NOMEM;
- goto fail;
- }
- int_dig_vv = kmalloc(i, GFP_KERNEL);
- if (!int_dig_vv) {
- retcode = ERR_NOMEM;
- goto fail;
- }
+ err = net_conf_from_attrs(new_conf, info);
+ if (err && err != -ENOMSG) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto fail;
}
+ retcode = check_net_options(tconn, new_conf);
+ if (retcode != NO_ERROR)
+ goto fail;
+
+ retcode = alloc_crypto(&crypto, new_conf);
+ if (retcode != NO_ERROR)
+ goto fail;
+
+ ((char *)new_conf->shared_secret)[SHARED_SECRET_MAX-1] = 0;
+
conn_flush_workqueue(tconn);
- spin_lock_irq(&tconn->req_lock);
- if (tconn->net_conf != NULL) {
+
+ mutex_lock(&tconn->conf_update);
+ old_conf = tconn->net_conf;
+ if (old_conf) {
retcode = ERR_NET_CONFIGURED;
- spin_unlock_irq(&tconn->req_lock);
+ mutex_unlock(&tconn->conf_update);
goto fail;
}
- tconn->net_conf = new_conf;
+ rcu_assign_pointer(tconn->net_conf, new_conf);
- crypto_free_hash(tconn->cram_hmac_tfm);
- tconn->cram_hmac_tfm = tfm;
-
- crypto_free_hash(tconn->integrity_w_tfm);
- tconn->integrity_w_tfm = integrity_w_tfm;
+ conn_free_crypto(tconn);
+ tconn->cram_hmac_tfm = crypto.cram_hmac_tfm;
+ tconn->integrity_tfm = crypto.integrity_tfm;
+ tconn->csums_tfm = crypto.csums_tfm;
+ tconn->verify_tfm = crypto.verify_tfm;
- crypto_free_hash(tconn->integrity_r_tfm);
- tconn->integrity_r_tfm = integrity_r_tfm;
+ tconn->my_addr_len = nla_len(adm_ctx.my_addr);
+ memcpy(&tconn->my_addr, nla_data(adm_ctx.my_addr), tconn->my_addr_len);
+ tconn->peer_addr_len = nla_len(adm_ctx.peer_addr);
+ memcpy(&tconn->peer_addr, nla_data(adm_ctx.peer_addr), tconn->peer_addr_len);
- kfree(tconn->int_dig_out);
- kfree(tconn->int_dig_in);
- kfree(tconn->int_dig_vv);
- tconn->int_dig_out=int_dig_out;
- tconn->int_dig_in=int_dig_in;
- tconn->int_dig_vv=int_dig_vv;
- retcode = _conn_request_state(tconn, NS(conn, C_UNCONNECTED), CS_VERBOSE);
- spin_unlock_irq(&tconn->req_lock);
+ mutex_unlock(&tconn->conf_update);
+ rcu_read_lock();
idr_for_each_entry(&tconn->volumes, mdev, i) {
mdev->send_cnt = 0;
mdev->recv_cnt = 0;
- kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE);
}
+ rcu_read_unlock();
+
+ retcode = conn_request_state(tconn, NS(conn, C_UNCONNECTED), CS_VERBOSE);
+
conn_reconfig_done(tconn);
drbd_adm_finish(info, retcode);
return 0;
fail:
- kfree(int_dig_out);
- kfree(int_dig_in);
- kfree(int_dig_vv);
- crypto_free_hash(tfm);
- crypto_free_hash(integrity_w_tfm);
- crypto_free_hash(integrity_r_tfm);
+ free_crypto(&crypto);
kfree(new_conf);
conn_reconfig_done(tconn);
return 0;
}
+static enum drbd_state_rv conn_try_disconnect(struct drbd_tconn *tconn, bool force)
+{
+ enum drbd_state_rv rv;
+
+ rv = conn_request_state(tconn, NS(conn, C_DISCONNECTING),
+ force ? CS_HARD : 0);
+
+ switch (rv) {
+ case SS_NOTHING_TO_DO:
+ break;
+ case SS_ALREADY_STANDALONE:
+ return SS_SUCCESS;
+ case SS_PRIMARY_NOP:
+ /* Our state checking code wants to see the peer outdated. */
+ rv = conn_request_state(tconn, NS2(conn, C_DISCONNECTING,
+ pdsk, D_OUTDATED), CS_VERBOSE);
+ break;
+ case SS_CW_FAILED_BY_PEER:
+ /* The peer probably wants to see us outdated. */
+ rv = conn_request_state(tconn, NS2(conn, C_DISCONNECTING,
+ disk, D_OUTDATED), 0);
+ if (rv == SS_IS_DISKLESS || rv == SS_LOWER_THAN_OUTDATED) {
+ rv = conn_request_state(tconn, NS(conn, C_DISCONNECTING),
+ CS_HARD);
+ }
+ break;
+ default:;
+ /* no special handling necessary */
+ }
+
+ if (rv >= SS_SUCCESS) {
+ enum drbd_state_rv rv2;
+ /* No one else can reconfigure the network while I am here.
+ * The state handling only uses drbd_thread_stop_nowait(),
+ * we want to really wait here until the receiver is no more.
+ */
+ drbd_thread_stop(&adm_ctx.tconn->receiver);
+
+ /* Race breaker. This additional state change request may be
+ * necessary, if this was a forced disconnect during a receiver
+ * restart. We may have "killed" the receiver thread just
+ * after drbdd_init() returned. Typically, we should be
+ * C_STANDALONE already, now, and this becomes a no-op.
+ */
+ rv2 = conn_request_state(tconn, NS(conn, C_STANDALONE),
+ CS_VERBOSE | CS_HARD);
+ if (rv2 < SS_SUCCESS)
+ conn_err(tconn,
+ "unexpected rv2=%d in conn_try_disconnect()\n",
+ rv2);
+ }
+ return rv;
+}
+
int drbd_adm_disconnect(struct sk_buff *skb, struct genl_info *info)
{
struct disconnect_parms parms;
struct drbd_tconn *tconn;
+ enum drbd_state_rv rv;
enum drbd_ret_code retcode;
int err;
- retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONNECTION);
if (!adm_ctx.reply_skb)
return retcode;
if (retcode != NO_ERROR)
tconn = adm_ctx.tconn;
memset(&parms, 0, sizeof(parms));
if (info->attrs[DRBD_NLA_DISCONNECT_PARMS]) {
- err = disconnect_parms_from_attrs(&parms, info->attrs);
+ err = disconnect_parms_from_attrs(&parms, info);
if (err) {
retcode = ERR_MANDATORY_TAG;
drbd_msg_put_info(from_attrs_err_to_txt(err));
}
}
- if (parms.force_disconnect) {
- spin_lock_irq(&tconn->req_lock);
- if (tconn->cstate >= C_WF_CONNECTION)
- _conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD);
- spin_unlock_irq(&tconn->req_lock);
- goto done;
- }
-
- retcode = conn_request_state(tconn, NS(conn, C_DISCONNECTING), 0);
-
- if (retcode == SS_NOTHING_TO_DO)
- goto done;
- else if (retcode == SS_ALREADY_STANDALONE)
- goto done;
- else if (retcode == SS_PRIMARY_NOP) {
- /* Our state checking code wants to see the peer outdated. */
- retcode = conn_request_state(tconn, NS2(conn, C_DISCONNECTING,
- pdsk, D_OUTDATED), CS_VERBOSE);
- } else if (retcode == SS_CW_FAILED_BY_PEER) {
- /* The peer probably wants to see us outdated. */
- retcode = conn_request_state(tconn, NS2(conn, C_DISCONNECTING,
- disk, D_OUTDATED), 0);
- if (retcode == SS_IS_DISKLESS || retcode == SS_LOWER_THAN_OUTDATED) {
- conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD);
- retcode = SS_SUCCESS;
- }
- }
-
- if (retcode < SS_SUCCESS)
- goto fail;
-
- if (wait_event_interruptible(tconn->ping_wait,
- tconn->cstate != C_DISCONNECTING)) {
- /* Do not test for mdev->state.conn == C_STANDALONE, since
- someone else might connect us in the mean time! */
- retcode = ERR_INTR;
- goto fail;
- }
-
- done:
- retcode = NO_ERROR;
+ rv = conn_try_disconnect(tconn, parms.force_disconnect);
+ if (rv < SS_SUCCESS)
+ retcode = rv; /* FIXME: Type mismatch. */
+ else
+ retcode = NO_ERROR;
fail:
drbd_adm_finish(info, retcode);
return 0;
if (mdev->state.role != mdev->state.peer)
iass = (mdev->state.role == R_PRIMARY);
else
- iass = test_bit(DISCARD_CONCURRENT, &mdev->tconn->flags);
+ iass = test_bit(RESOLVE_CONFLICTS, &mdev->tconn->flags);
if (iass)
drbd_start_resync(mdev, C_SYNC_SOURCE);
int drbd_adm_resize(struct sk_buff *skb, struct genl_info *info)
{
+ struct disk_conf *old_disk_conf, *new_disk_conf = NULL;
struct resize_parms rs;
struct drbd_conf *mdev;
enum drbd_ret_code retcode;
enum determine_dev_size dd;
enum dds_flags ddsf;
+ sector_t u_size;
int err;
retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
goto fail;
memset(&rs, 0, sizeof(struct resize_parms));
- if (info->attrs[DRBD_NLA_RESIZE_PARMS]) {
- err = resize_parms_from_attrs(&rs, info->attrs);
- if (err) {
- retcode = ERR_MANDATORY_TAG;
- drbd_msg_put_info(from_attrs_err_to_txt(err));
- goto fail;
- }
- }
-
- mdev = adm_ctx.mdev;
- if (mdev->state.conn > C_CONNECTED) {
- retcode = ERR_RESIZE_RESYNC;
- goto fail;
- }
-
- if (mdev->state.role == R_SECONDARY &&
- mdev->state.peer == R_SECONDARY) {
- retcode = ERR_NO_PRIMARY;
- goto fail;
- }
-
- if (!get_ldev(mdev)) {
- retcode = ERR_NO_DISK;
- goto fail;
- }
-
- if (rs.no_resync && mdev->tconn->agreed_pro_version < 93) {
- retcode = ERR_NEED_APV_93;
- goto fail;
- }
-
- if (mdev->ldev->known_size != drbd_get_capacity(mdev->ldev->backing_bdev))
- mdev->ldev->known_size = drbd_get_capacity(mdev->ldev->backing_bdev);
-
- mdev->ldev->dc.disk_size = (sector_t)rs.resize_size;
- ddsf = (rs.resize_force ? DDSF_FORCED : 0) | (rs.no_resync ? DDSF_NO_RESYNC : 0);
- dd = drbd_determine_dev_size(mdev, ddsf);
- drbd_md_sync(mdev);
- put_ldev(mdev);
- if (dd == dev_size_error) {
- retcode = ERR_NOMEM_BITMAP;
- goto fail;
- }
-
- if (mdev->state.conn == C_CONNECTED) {
- if (dd == grew)
- set_bit(RESIZE_PENDING, &mdev->flags);
-
- drbd_send_uuids(mdev);
- drbd_send_sizes(mdev, 1, ddsf);
- }
-
- fail:
- drbd_adm_finish(info, retcode);
- return 0;
-}
-
-int drbd_adm_syncer(struct sk_buff *skb, struct genl_info *info)
-{
- struct drbd_conf *mdev;
- enum drbd_ret_code retcode;
- int err;
- int ovr; /* online verify running */
- int rsr; /* re-sync running */
- struct crypto_hash *verify_tfm = NULL;
- struct crypto_hash *csums_tfm = NULL;
- struct syncer_conf sc;
- cpumask_var_t new_cpu_mask;
- int *rs_plan_s = NULL;
- int fifo_size;
-
- retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
- if (!adm_ctx.reply_skb)
- return retcode;
- if (retcode != NO_ERROR)
- goto fail;
- mdev = adm_ctx.mdev;
-
- if (!zalloc_cpumask_var(&new_cpu_mask, GFP_KERNEL)) {
- retcode = ERR_NOMEM;
- drbd_msg_put_info("unable to allocate cpumask");
- goto fail;
- }
-
- if (((struct drbd_genlmsghdr*)info->userhdr)->flags
- & DRBD_GENL_F_SET_DEFAULTS) {
- memset(&sc, 0, sizeof(struct syncer_conf));
- sc.rate = DRBD_RATE_DEF;
- sc.after = DRBD_AFTER_DEF;
- sc.al_extents = DRBD_AL_EXTENTS_DEF;
- sc.on_no_data = DRBD_ON_NO_DATA_DEF;
- sc.c_plan_ahead = DRBD_C_PLAN_AHEAD_DEF;
- sc.c_delay_target = DRBD_C_DELAY_TARGET_DEF;
- sc.c_fill_target = DRBD_C_FILL_TARGET_DEF;
- sc.c_max_rate = DRBD_C_MAX_RATE_DEF;
- sc.c_min_rate = DRBD_C_MIN_RATE_DEF;
- } else
- memcpy(&sc, &mdev->sync_conf, sizeof(struct syncer_conf));
-
- err = syncer_conf_from_attrs(&sc, info->attrs);
- if (err) {
- retcode = ERR_MANDATORY_TAG;
- drbd_msg_put_info(from_attrs_err_to_txt(err));
- goto fail;
- }
-
- /* re-sync running */
- rsr = ( mdev->state.conn == C_SYNC_SOURCE ||
- mdev->state.conn == C_SYNC_TARGET ||
- mdev->state.conn == C_PAUSED_SYNC_S ||
- mdev->state.conn == C_PAUSED_SYNC_T );
-
- if (rsr && strcmp(sc.csums_alg, mdev->sync_conf.csums_alg)) {
- retcode = ERR_CSUMS_RESYNC_RUNNING;
- goto fail;
- }
-
- if (!rsr && sc.csums_alg[0]) {
- csums_tfm = crypto_alloc_hash(sc.csums_alg, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(csums_tfm)) {
- csums_tfm = NULL;
- retcode = ERR_CSUMS_ALG;
- goto fail;
- }
-
- if (!drbd_crypto_is_hash(crypto_hash_tfm(csums_tfm))) {
- retcode = ERR_CSUMS_ALG_ND;
- goto fail;
- }
- }
-
- /* online verify running */
- ovr = (mdev->state.conn == C_VERIFY_S || mdev->state.conn == C_VERIFY_T);
-
- if (ovr) {
- if (strcmp(sc.verify_alg, mdev->sync_conf.verify_alg)) {
- retcode = ERR_VERIFY_RUNNING;
- goto fail;
- }
- }
-
- if (!ovr && sc.verify_alg[0]) {
- verify_tfm = crypto_alloc_hash(sc.verify_alg, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(verify_tfm)) {
- verify_tfm = NULL;
- retcode = ERR_VERIFY_ALG;
- goto fail;
- }
-
- if (!drbd_crypto_is_hash(crypto_hash_tfm(verify_tfm))) {
- retcode = ERR_VERIFY_ALG_ND;
- goto fail;
- }
- }
-
- /* silently ignore cpu mask on UP kernel */
- if (nr_cpu_ids > 1 && sc.cpu_mask[0] != 0) {
- err = __bitmap_parse(sc.cpu_mask, 32, 0,
- cpumask_bits(new_cpu_mask), nr_cpu_ids);
+ if (info->attrs[DRBD_NLA_RESIZE_PARMS]) {
+ err = resize_parms_from_attrs(&rs, info);
if (err) {
- dev_warn(DEV, "__bitmap_parse() failed with %d\n", err);
- retcode = ERR_CPU_MASK_PARSE;
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
goto fail;
}
}
- if (!expect(sc.rate >= 1))
- sc.rate = 1;
+ mdev = adm_ctx.mdev;
+ if (mdev->state.conn > C_CONNECTED) {
+ retcode = ERR_RESIZE_RESYNC;
+ goto fail;
+ }
- /* clip to allowed range */
- if (!expect(sc.al_extents >= DRBD_AL_EXTENTS_MIN))
- sc.al_extents = DRBD_AL_EXTENTS_MIN;
- if (!expect(sc.al_extents <= DRBD_AL_EXTENTS_MAX))
- sc.al_extents = DRBD_AL_EXTENTS_MAX;
+ if (mdev->state.role == R_SECONDARY &&
+ mdev->state.peer == R_SECONDARY) {
+ retcode = ERR_NO_PRIMARY;
+ goto fail;
+ }
- /* most sanity checks done, try to assign the new sync-after
- * dependency. need to hold the global lock in there,
- * to avoid a race in the dependency loop check. */
- retcode = drbd_alter_sa(mdev, sc.after);
- if (retcode != NO_ERROR)
+ if (!get_ldev(mdev)) {
+ retcode = ERR_NO_DISK;
goto fail;
+ }
- fifo_size = (sc.c_plan_ahead * 10 * SLEEP_TIME) / HZ;
- if (fifo_size != mdev->rs_plan_s.size && fifo_size > 0) {
- rs_plan_s = kzalloc(sizeof(int) * fifo_size, GFP_KERNEL);
- if (!rs_plan_s) {
- dev_err(DEV, "kmalloc of fifo_buffer failed");
+ if (rs.no_resync && mdev->tconn->agreed_pro_version < 93) {
+ retcode = ERR_NEED_APV_93;
+ goto fail_ldev;
+ }
+
+ rcu_read_lock();
+ u_size = rcu_dereference(mdev->ldev->disk_conf)->disk_size;
+ rcu_read_unlock();
+ if (u_size != (sector_t)rs.resize_size) {
+ new_disk_conf = kmalloc(sizeof(struct disk_conf), GFP_KERNEL);
+ if (!new_disk_conf) {
retcode = ERR_NOMEM;
- goto fail;
+ goto fail_ldev;
}
}
- /* ok, assign the rest of it as well.
- * lock against receive_SyncParam() */
- spin_lock(&mdev->peer_seq_lock);
- mdev->sync_conf = sc;
+ if (mdev->ldev->known_size != drbd_get_capacity(mdev->ldev->backing_bdev))
+ mdev->ldev->known_size = drbd_get_capacity(mdev->ldev->backing_bdev);
- if (!rsr) {
- crypto_free_hash(mdev->csums_tfm);
- mdev->csums_tfm = csums_tfm;
- csums_tfm = NULL;
+ if (new_disk_conf) {
+ mutex_lock(&mdev->tconn->conf_update);
+ old_disk_conf = mdev->ldev->disk_conf;
+ *new_disk_conf = *old_disk_conf;
+ new_disk_conf->disk_size = (sector_t)rs.resize_size;
+ rcu_assign_pointer(mdev->ldev->disk_conf, new_disk_conf);
+ mutex_unlock(&mdev->tconn->conf_update);
+ synchronize_rcu();
+ kfree(old_disk_conf);
}
- if (!ovr) {
- crypto_free_hash(mdev->verify_tfm);
- mdev->verify_tfm = verify_tfm;
- verify_tfm = NULL;
+ ddsf = (rs.resize_force ? DDSF_FORCED : 0) | (rs.no_resync ? DDSF_NO_RESYNC : 0);
+ dd = drbd_determine_dev_size(mdev, ddsf);
+ drbd_md_sync(mdev);
+ put_ldev(mdev);
+ if (dd == dev_size_error) {
+ retcode = ERR_NOMEM_BITMAP;
+ goto fail;
}
- if (fifo_size != mdev->rs_plan_s.size) {
- kfree(mdev->rs_plan_s.values);
- mdev->rs_plan_s.values = rs_plan_s;
- mdev->rs_plan_s.size = fifo_size;
- mdev->rs_planed = 0;
- rs_plan_s = NULL;
+ if (mdev->state.conn == C_CONNECTED) {
+ if (dd == grew)
+ set_bit(RESIZE_PENDING, &mdev->flags);
+
+ drbd_send_uuids(mdev);
+ drbd_send_sizes(mdev, 1, ddsf);
}
- spin_unlock(&mdev->peer_seq_lock);
+ fail:
+ drbd_adm_finish(info, retcode);
+ return 0;
- if (get_ldev(mdev)) {
- wait_event(mdev->al_wait, lc_try_lock(mdev->act_log));
- drbd_al_shrink(mdev);
- err = drbd_check_al_size(mdev);
- lc_unlock(mdev->act_log);
- wake_up(&mdev->al_wait);
+ fail_ldev:
+ put_ldev(mdev);
+ goto fail;
+}
- put_ldev(mdev);
- drbd_md_sync(mdev);
+int drbd_adm_resource_opts(struct sk_buff *skb, struct genl_info *info)
+{
+ enum drbd_ret_code retcode;
+ struct drbd_tconn *tconn;
+ struct res_opts res_opts;
+ int err;
- if (err) {
- retcode = ERR_NOMEM;
- goto fail;
- }
- }
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_RESOURCE);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto fail;
+ tconn = adm_ctx.tconn;
- if (mdev->state.conn >= C_CONNECTED)
- drbd_send_sync_param(mdev, &sc);
+ res_opts = tconn->res_opts;
+ if (should_set_defaults(info))
+ set_res_opts_defaults(&res_opts);
+
+ err = res_opts_from_attrs(&res_opts, info);
+ if (err && err != -ENOMSG) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto fail;
+ }
- if (!cpumask_equal(mdev->tconn->cpu_mask, new_cpu_mask)) {
- cpumask_copy(mdev->tconn->cpu_mask, new_cpu_mask);
- drbd_calc_cpu_mask(mdev->tconn);
- mdev->tconn->receiver.reset_cpu_mask = 1;
- mdev->tconn->asender.reset_cpu_mask = 1;
- mdev->tconn->worker.reset_cpu_mask = 1;
+ err = set_resource_options(tconn, &res_opts);
+ if (err) {
+ retcode = ERR_INVALID_REQUEST;
+ if (err == -ENOMEM)
+ retcode = ERR_NOMEM;
}
- kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE);
fail:
- kfree(rs_plan_s);
- free_cpumask_var(new_cpu_mask);
- crypto_free_hash(csums_tfm);
- crypto_free_hash(verify_tfm);
-
drbd_adm_finish(info, retcode);
return 0;
}
mdev = adm_ctx.mdev;
/* If there is still bitmap IO pending, probably because of a previous
- * resync just being finished, wait for it before requesting a new resync. */
+ * resync just being finished, wait for it before requesting a new resync.
+ * Also wait for it's after_state_ch(). */
+ drbd_suspend_io(mdev);
wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags));
+ drbd_flush_workqueue(mdev);
retcode = _drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_T), CS_ORDERED);
retcode = drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_T));
}
+ drbd_resume_io(mdev);
+
+out:
+ drbd_adm_finish(info, retcode);
+ return 0;
+}
+
+static int drbd_adm_simple_request_state(struct sk_buff *skb, struct genl_info *info,
+ union drbd_state mask, union drbd_state val)
+{
+ enum drbd_ret_code retcode;
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+ retcode = drbd_request_state(adm_ctx.mdev, mask, val);
out:
drbd_adm_finish(info, retcode);
return 0;
return rv;
}
-static int drbd_adm_simple_request_state(struct sk_buff *skb, struct genl_info *info,
- union drbd_state mask, union drbd_state val)
+int drbd_adm_invalidate_peer(struct sk_buff *skb, struct genl_info *info)
{
- enum drbd_ret_code retcode;
+ int retcode; /* drbd_ret_code, drbd_state_rv */
+ struct drbd_conf *mdev;
retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
if (!adm_ctx.reply_skb)
if (retcode != NO_ERROR)
goto out;
- retcode = drbd_request_state(adm_ctx.mdev, mask, val);
+ mdev = adm_ctx.mdev;
+
+ /* If there is still bitmap IO pending, probably because of a previous
+ * resync just being finished, wait for it before requesting a new resync.
+ * Also wait for it's after_state_ch(). */
+ drbd_suspend_io(mdev);
+ wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags));
+ drbd_flush_workqueue(mdev);
+
+ retcode = _drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_S), CS_ORDERED);
+ if (retcode < SS_SUCCESS) {
+ if (retcode == SS_NEED_CONNECTION && mdev->state.role == R_PRIMARY) {
+ /* The peer will get a resync upon connect anyways.
+ * Just make that into a full resync. */
+ retcode = drbd_request_state(mdev, NS(pdsk, D_INCONSISTENT));
+ if (retcode >= SS_SUCCESS) {
+ if (drbd_bitmap_io(mdev, &drbd_bmio_set_susp_al,
+ "set_n_write from invalidate_peer",
+ BM_LOCKED_SET_ALLOWED))
+ retcode = ERR_IO_MD_DISK;
+ }
+ } else
+ retcode = drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_S));
+ }
+ drbd_resume_io(mdev);
+
out:
drbd_adm_finish(info, retcode);
return 0;
}
-int drbd_adm_invalidate_peer(struct sk_buff *skb, struct genl_info *info)
-{
- return drbd_adm_simple_request_state(skb, info, NS(conn, C_STARTING_SYNC_S));
-}
-
int drbd_adm_pause_sync(struct sk_buff *skb, struct genl_info *info)
{
enum drbd_ret_code retcode;
int drbd_adm_resume_sync(struct sk_buff *skb, struct genl_info *info)
{
- union drbd_state s;
+ union drbd_dev_state s;
enum drbd_ret_code retcode;
retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
return drbd_adm_simple_request_state(skb, info, NS(disk, D_OUTDATED));
}
+int nla_put_drbd_cfg_context(struct sk_buff *skb, struct drbd_tconn *tconn, unsigned vnr)
+{
+ struct nlattr *nla;
+ nla = nla_nest_start(skb, DRBD_NLA_CFG_CONTEXT);
+ if (!nla)
+ goto nla_put_failure;
+ if (vnr != VOLUME_UNSPECIFIED &&
+ nla_put_u32(skb, T_ctx_volume, vnr))
+ goto nla_put_failure;
+ if (nla_put_string(skb, T_ctx_resource_name, tconn->name))
+ goto nla_put_failure;
+ if (tconn->my_addr_len &&
+ nla_put(skb, T_ctx_my_addr, tconn->my_addr_len, &tconn->my_addr))
+ goto nla_put_failure;
+ if (tconn->peer_addr_len &&
+ nla_put(skb, T_ctx_peer_addr, tconn->peer_addr_len, &tconn->peer_addr))
+ goto nla_put_failure;
+ nla_nest_end(skb, nla);
+ return 0;
+
+nla_put_failure:
+ if (nla)
+ nla_nest_cancel(skb, nla);
+ return -EMSGSIZE;
+}
+
int nla_put_status_info(struct sk_buff *skb, struct drbd_conf *mdev,
const struct sib_info *sib)
{
struct state_info *si = NULL; /* for sizeof(si->member); */
+ struct net_conf *nc;
struct nlattr *nla;
int got_ldev;
- int got_net;
int err = 0;
int exclude_sensitive;
exclude_sensitive = sib || !capable(CAP_SYS_ADMIN);
got_ldev = get_ldev(mdev);
- got_net = get_net_conf(mdev->tconn);
/* We need to add connection name and volume number information still.
* Minor number is in drbd_genlmsghdr. */
- nla = nla_nest_start(skb, DRBD_NLA_CFG_CONTEXT);
- if (!nla)
+ if (nla_put_drbd_cfg_context(skb, mdev->tconn, mdev->vnr))
+ goto nla_put_failure;
+
+ if (res_opts_to_skb(skb, &mdev->tconn->res_opts, exclude_sensitive))
goto nla_put_failure;
- NLA_PUT_U32(skb, T_ctx_volume, mdev->vnr);
- NLA_PUT_STRING(skb, T_ctx_conn_name, mdev->tconn->name);
- nla_nest_end(skb, nla);
+ rcu_read_lock();
if (got_ldev)
- if (disk_conf_to_skb(skb, &mdev->ldev->dc, exclude_sensitive))
- goto nla_put_failure;
- if (got_net)
- if (net_conf_to_skb(skb, mdev->tconn->net_conf, exclude_sensitive))
+ if (disk_conf_to_skb(skb, rcu_dereference(mdev->ldev->disk_conf), exclude_sensitive))
goto nla_put_failure;
- if (syncer_conf_to_skb(skb, &mdev->sync_conf, exclude_sensitive))
- goto nla_put_failure;
+ nc = rcu_dereference(mdev->tconn->net_conf);
+ if (nc)
+ err = net_conf_to_skb(skb, nc, exclude_sensitive);
+ rcu_read_unlock();
+ if (err)
+ goto nla_put_failure;
nla = nla_nest_start(skb, DRBD_NLA_STATE_INFO);
if (!nla)
goto nla_put_failure;
- NLA_PUT_U32(skb, T_sib_reason, sib ? sib->sib_reason : SIB_GET_STATUS_REPLY);
- NLA_PUT_U32(skb, T_current_state, mdev->state.i);
- NLA_PUT_U64(skb, T_ed_uuid, mdev->ed_uuid);
- NLA_PUT_U64(skb, T_capacity, drbd_get_capacity(mdev->this_bdev));
+ if (nla_put_u32(skb, T_sib_reason, sib ? sib->sib_reason : SIB_GET_STATUS_REPLY) ||
+ nla_put_u32(skb, T_current_state, mdev->state.i) ||
+ nla_put_u64(skb, T_ed_uuid, mdev->ed_uuid) ||
+ nla_put_u64(skb, T_capacity, drbd_get_capacity(mdev->this_bdev)) ||
+ nla_put_u64(skb, T_send_cnt, mdev->send_cnt) ||
+ nla_put_u64(skb, T_recv_cnt, mdev->recv_cnt) ||
+ nla_put_u64(skb, T_read_cnt, mdev->read_cnt) ||
+ nla_put_u64(skb, T_writ_cnt, mdev->writ_cnt) ||
+ nla_put_u64(skb, T_al_writ_cnt, mdev->al_writ_cnt) ||
+ nla_put_u64(skb, T_bm_writ_cnt, mdev->bm_writ_cnt) ||
+ nla_put_u32(skb, T_ap_bio_cnt, atomic_read(&mdev->ap_bio_cnt)) ||
+ nla_put_u32(skb, T_ap_pending_cnt, atomic_read(&mdev->ap_pending_cnt)) ||
+ nla_put_u32(skb, T_rs_pending_cnt, atomic_read(&mdev->rs_pending_cnt)))
+ goto nla_put_failure;
if (got_ldev) {
- NLA_PUT_U32(skb, T_disk_flags, mdev->ldev->md.flags);
- NLA_PUT(skb, T_uuids, sizeof(si->uuids), mdev->ldev->md.uuid);
- NLA_PUT_U64(skb, T_bits_total, drbd_bm_bits(mdev));
- NLA_PUT_U64(skb, T_bits_oos, drbd_bm_total_weight(mdev));
+ int err;
+
+ spin_lock_irq(&mdev->ldev->md.uuid_lock);
+ err = nla_put(skb, T_uuids, sizeof(si->uuids), mdev->ldev->md.uuid);
+ spin_unlock_irq(&mdev->ldev->md.uuid_lock);
+
+ if (err)
+ goto nla_put_failure;
+
+ if (nla_put_u32(skb, T_disk_flags, mdev->ldev->md.flags) ||
+ nla_put_u64(skb, T_bits_total, drbd_bm_bits(mdev)) ||
+ nla_put_u64(skb, T_bits_oos, drbd_bm_total_weight(mdev)))
+ goto nla_put_failure;
if (C_SYNC_SOURCE <= mdev->state.conn &&
C_PAUSED_SYNC_T >= mdev->state.conn) {
- NLA_PUT_U64(skb, T_bits_rs_total, mdev->rs_total);
- NLA_PUT_U64(skb, T_bits_rs_failed, mdev->rs_failed);
+ if (nla_put_u64(skb, T_bits_rs_total, mdev->rs_total) ||
+ nla_put_u64(skb, T_bits_rs_failed, mdev->rs_failed))
+ goto nla_put_failure;
}
}
case SIB_GET_STATUS_REPLY:
break;
case SIB_STATE_CHANGE:
- NLA_PUT_U32(skb, T_prev_state, sib->os.i);
- NLA_PUT_U32(skb, T_new_state, sib->ns.i);
+ if (nla_put_u32(skb, T_prev_state, sib->os.i) ||
+ nla_put_u32(skb, T_new_state, sib->ns.i))
+ goto nla_put_failure;
break;
case SIB_HELPER_POST:
- NLA_PUT_U32(skb,
- T_helper_exit_code, sib->helper_exit_code);
+ if (nla_put_u32(skb, T_helper_exit_code,
+ sib->helper_exit_code))
+ goto nla_put_failure;
/* fall through */
case SIB_HELPER_PRE:
- NLA_PUT_STRING(skb, T_helper, sib->helper_name);
+ if (nla_put_string(skb, T_helper, sib->helper_name))
+ goto nla_put_failure;
break;
}
}
err = -EMSGSIZE;
if (got_ldev)
put_ldev(mdev);
- if (got_net)
- put_net_conf(mdev->tconn);
return err;
}
return 0;
}
-int drbd_adm_get_status_all(struct sk_buff *skb, struct netlink_callback *cb)
+int get_one_status(struct sk_buff *skb, struct netlink_callback *cb)
{
struct drbd_conf *mdev;
struct drbd_genlmsghdr *dh;
- int minor = cb->args[0];
-
- /* Open coded deferred single idr_for_each_entry iteration.
+ struct drbd_tconn *pos = (struct drbd_tconn*)cb->args[0];
+ struct drbd_tconn *tconn = NULL;
+ struct drbd_tconn *tmp;
+ unsigned volume = cb->args[1];
+
+ /* Open coded, deferred, iteration:
+ * list_for_each_entry_safe(tconn, tmp, &drbd_tconns, all_tconn) {
+ * idr_for_each_entry(&tconn->volumes, mdev, i) {
+ * ...
+ * }
+ * }
+ * where tconn is cb->args[0];
+ * and i is cb->args[1];
+ *
+ * cb->args[2] indicates if we shall loop over all resources,
+ * or just dump all volumes of a single resource.
+ *
* This may miss entries inserted after this dump started,
* or entries deleted before they are reached.
- * But we need to make sure the mdev won't disappear while
- * we are looking at it. */
+ *
+ * We need to make sure the mdev won't disappear while
+ * we are looking at it, and revalidate our iterators
+ * on each iteration.
+ */
+ /* synchronize with conn_create()/conn_destroy() */
rcu_read_lock();
- mdev = idr_get_next(&minors, &minor);
- if (mdev) {
- dh = genlmsg_put(skb, NETLINK_CB(cb->skb).pid,
+ /* revalidate iterator position */
+ list_for_each_entry_rcu(tmp, &drbd_tconns, all_tconn) {
+ if (pos == NULL) {
+ /* first iteration */
+ pos = tmp;
+ tconn = pos;
+ break;
+ }
+ if (tmp == pos) {
+ tconn = pos;
+ break;
+ }
+ }
+ if (tconn) {
+next_tconn:
+ mdev = idr_get_next(&tconn->volumes, &volume);
+ if (!mdev) {
+ /* No more volumes to dump on this tconn.
+ * Advance tconn iterator. */
+ pos = list_entry_rcu(tconn->all_tconn.next,
+ struct drbd_tconn, all_tconn);
+ /* Did we dump any volume on this tconn yet? */
+ if (volume != 0) {
+ /* If we reached the end of the list,
+ * or only a single resource dump was requested,
+ * we are done. */
+ if (&pos->all_tconn == &drbd_tconns || cb->args[2])
+ goto out;
+ volume = 0;
+ tconn = pos;
+ goto next_tconn;
+ }
+ }
+
+ dh = genlmsg_put(skb, NETLINK_CB(cb->skb).portid,
cb->nlh->nlmsg_seq, &drbd_genl_family,
NLM_F_MULTI, DRBD_ADM_GET_STATUS);
if (!dh)
- goto errout;
+ goto out;
+
+ if (!mdev) {
+ /* This is a tconn without a single volume.
+ * Suprisingly enough, it may have a network
+ * configuration. */
+ struct net_conf *nc;
+ dh->minor = -1U;
+ dh->ret_code = NO_ERROR;
+ if (nla_put_drbd_cfg_context(skb, tconn, VOLUME_UNSPECIFIED))
+ goto cancel;
+ nc = rcu_dereference(tconn->net_conf);
+ if (nc && net_conf_to_skb(skb, nc, 1) != 0)
+ goto cancel;
+ goto done;
+ }
- D_ASSERT(mdev->minor == minor);
+ D_ASSERT(mdev->vnr == volume);
+ D_ASSERT(mdev->tconn == tconn);
- dh->minor = minor;
+ dh->minor = mdev_to_minor(mdev);
dh->ret_code = NO_ERROR;
if (nla_put_status_info(skb, mdev, NULL)) {
+cancel:
genlmsg_cancel(skb, dh);
- goto errout;
+ goto out;
}
+done:
genlmsg_end(skb, dh);
}
-errout:
+out:
rcu_read_unlock();
- /* where to start idr_get_next with the next iteration */
- cb->args[0] = minor+1;
+ /* where to start the next iteration */
+ cb->args[0] = (long)pos;
+ cb->args[1] = (pos == tconn) ? volume + 1 : 0;
- /* No more minors found: empty skb. Which will terminate the dump. */
+ /* No more tconns/volumes/minors found results in an empty skb.
+ * Which will terminate the dump. */
return skb->len;
}
+/*
+ * Request status of all resources, or of all volumes within a single resource.
+ *
+ * This is a dump, as the answer may not fit in a single reply skb otherwise.
+ * Which means we cannot use the family->attrbuf or other such members, because
+ * dump is NOT protected by the genl_lock(). During dump, we only have access
+ * to the incoming skb, and need to opencode "parsing" of the nlattr payload.
+ *
+ * Once things are setup properly, we call into get_one_status().
+ */
+int drbd_adm_get_status_all(struct sk_buff *skb, struct netlink_callback *cb)
+{
+ const unsigned hdrlen = GENL_HDRLEN + GENL_MAGIC_FAMILY_HDRSZ;
+ struct nlattr *nla;
+ const char *resource_name;
+ struct drbd_tconn *tconn;
+ int maxtype;
+
+ /* Is this a followup call? */
+ if (cb->args[0]) {
+ /* ... of a single resource dump,
+ * and the resource iterator has been advanced already? */
+ if (cb->args[2] && cb->args[2] != cb->args[0])
+ return 0; /* DONE. */
+ goto dump;
+ }
+
+ /* First call (from netlink_dump_start). We need to figure out
+ * which resource(s) the user wants us to dump. */
+ nla = nla_find(nlmsg_attrdata(cb->nlh, hdrlen),
+ nlmsg_attrlen(cb->nlh, hdrlen),
+ DRBD_NLA_CFG_CONTEXT);
+
+ /* No explicit context given. Dump all. */
+ if (!nla)
+ goto dump;
+ maxtype = ARRAY_SIZE(drbd_cfg_context_nl_policy) - 1;
+ nla = drbd_nla_find_nested(maxtype, nla, __nla_type(T_ctx_resource_name));
+ if (IS_ERR(nla))
+ return PTR_ERR(nla);
+ /* context given, but no name present? */
+ if (!nla)
+ return -EINVAL;
+ resource_name = nla_data(nla);
+ tconn = conn_get_by_name(resource_name);
+
+ if (!tconn)
+ return -ENODEV;
+
+ kref_put(&tconn->kref, &conn_destroy); /* get_one_status() (re)validates tconn by itself */
+
+ /* prime iterators, and set "filter" mode mark:
+ * only dump this tconn. */
+ cb->args[0] = (long)tconn;
+ /* cb->args[1] = 0; passed in this way. */
+ cb->args[2] = (long)tconn;
+
+dump:
+ return get_one_status(skb, cb);
+}
+
int drbd_adm_get_timeout_type(struct sk_buff *skb, struct genl_info *info)
{
enum drbd_ret_code retcode;
{
struct drbd_conf *mdev;
enum drbd_ret_code retcode;
+ struct start_ov_parms parms;
retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
if (!adm_ctx.reply_skb)
goto out;
mdev = adm_ctx.mdev;
+
+ /* resume from last known position, if possible */
+ parms.ov_start_sector = mdev->ov_start_sector;
+ parms.ov_stop_sector = ULLONG_MAX;
if (info->attrs[DRBD_NLA_START_OV_PARMS]) {
- /* resume from last known position, if possible */
- struct start_ov_parms parms =
- { .ov_start_sector = mdev->ov_start_sector };
- int err = start_ov_parms_from_attrs(&parms, info->attrs);
+ int err = start_ov_parms_from_attrs(&parms, info);
if (err) {
retcode = ERR_MANDATORY_TAG;
drbd_msg_put_info(from_attrs_err_to_txt(err));
goto out;
}
- /* w_make_ov_request expects position to be aligned */
- mdev->ov_start_sector = parms.ov_start_sector & ~BM_SECT_PER_BIT;
}
+ /* w_make_ov_request expects position to be aligned */
+ mdev->ov_start_sector = parms.ov_start_sector & ~(BM_SECT_PER_BIT-1);
+ mdev->ov_stop_sector = parms.ov_stop_sector;
+
/* If there is still bitmap IO pending, e.g. previous resync or verify
* just being finished, wait for it before requesting a new resync. */
+ drbd_suspend_io(mdev);
wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags));
retcode = drbd_request_state(mdev,NS(conn,C_VERIFY_S));
+ drbd_resume_io(mdev);
out:
drbd_adm_finish(info, retcode);
return 0;
mdev = adm_ctx.mdev;
memset(&args, 0, sizeof(args));
if (info->attrs[DRBD_NLA_NEW_C_UUID_PARMS]) {
- err = new_c_uuid_parms_from_attrs(&args, info->attrs);
+ err = new_c_uuid_parms_from_attrs(&args, info);
if (err) {
retcode = ERR_MANDATORY_TAG;
drbd_msg_put_info(from_attrs_err_to_txt(err));
}
static enum drbd_ret_code
-drbd_check_conn_name(const char *name)
+drbd_check_resource_name(const char *name)
{
if (!name || !name[0]) {
- drbd_msg_put_info("connection name missing");
+ drbd_msg_put_info("resource name missing");
return ERR_MANDATORY_TAG;
}
/* if we want to use these in sysfs/configfs/debugfs some day,
* we must not allow slashes */
if (strchr(name, '/')) {
- drbd_msg_put_info("invalid connection name");
+ drbd_msg_put_info("invalid resource name");
return ERR_INVALID_REQUEST;
}
return NO_ERROR;
}
-int drbd_adm_create_connection(struct sk_buff *skb, struct genl_info *info)
+int drbd_adm_new_resource(struct sk_buff *skb, struct genl_info *info)
{
enum drbd_ret_code retcode;
+ struct res_opts res_opts;
+ int err;
retcode = drbd_adm_prepare(skb, info, 0);
if (!adm_ctx.reply_skb)
if (retcode != NO_ERROR)
goto out;
- retcode = drbd_check_conn_name(adm_ctx.conn_name);
+ set_res_opts_defaults(&res_opts);
+ err = res_opts_from_attrs(&res_opts, info);
+ if (err && err != -ENOMSG) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto out;
+ }
+
+ retcode = drbd_check_resource_name(adm_ctx.resource_name);
if (retcode != NO_ERROR)
goto out;
if (adm_ctx.tconn) {
- retcode = ERR_INVALID_REQUEST;
- drbd_msg_put_info("connection exists");
+ if (info->nlhdr->nlmsg_flags & NLM_F_EXCL) {
+ retcode = ERR_INVALID_REQUEST;
+ drbd_msg_put_info("resource exists");
+ }
+ /* else: still NO_ERROR */
goto out;
}
- if (!drbd_new_tconn(adm_ctx.conn_name))
+ if (!conn_create(adm_ctx.resource_name, &res_opts))
retcode = ERR_NOMEM;
out:
drbd_adm_finish(info, retcode);
struct drbd_genlmsghdr *dh = info->userhdr;
enum drbd_ret_code retcode;
- retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_RESOURCE);
if (!adm_ctx.reply_skb)
return retcode;
if (retcode != NO_ERROR)
goto out;
- /* FIXME drop minor_count parameter, limit to MINORMASK */
- if (dh->minor >= minor_count) {
+ if (dh->minor > MINORMASK) {
drbd_msg_put_info("requested minor out of range");
retcode = ERR_INVALID_REQUEST;
goto out;
}
- /* FIXME we need a define here */
- if (adm_ctx.volume >= 256) {
+ if (adm_ctx.volume > DRBD_VOLUME_MAX) {
drbd_msg_put_info("requested volume id out of range");
retcode = ERR_INVALID_REQUEST;
goto out;
}
+ /* drbd_adm_prepare made sure already
+ * that mdev->tconn and mdev->vnr match the request. */
+ if (adm_ctx.mdev) {
+ if (info->nlhdr->nlmsg_flags & NLM_F_EXCL)
+ retcode = ERR_MINOR_EXISTS;
+ /* else: still NO_ERROR */
+ goto out;
+ }
+
retcode = conn_new_minor(adm_ctx.tconn, dh->minor, adm_ctx.volume);
out:
drbd_adm_finish(info, retcode);
return 0;
}
+static enum drbd_ret_code adm_delete_minor(struct drbd_conf *mdev)
+{
+ if (mdev->state.disk == D_DISKLESS &&
+ /* no need to be mdev->state.conn == C_STANDALONE &&
+ * we may want to delete a minor from a live replication group.
+ */
+ mdev->state.role == R_SECONDARY) {
+ _drbd_request_state(mdev, NS(conn, C_WF_REPORT_PARAMS),
+ CS_VERBOSE + CS_WAIT_COMPLETE);
+ idr_remove(&mdev->tconn->volumes, mdev->vnr);
+ idr_remove(&minors, mdev_to_minor(mdev));
+ del_gendisk(mdev->vdisk);
+ synchronize_rcu();
+ kref_put(&mdev->kref, &drbd_minor_destroy);
+ return NO_ERROR;
+ } else
+ return ERR_MINOR_CONFIGURED;
+}
+
int drbd_adm_delete_minor(struct sk_buff *skb, struct genl_info *info)
{
- struct drbd_conf *mdev;
enum drbd_ret_code retcode;
retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
if (retcode != NO_ERROR)
goto out;
- mdev = adm_ctx.mdev;
- if (mdev->state.disk == D_DISKLESS &&
- mdev->state.conn == C_STANDALONE &&
- mdev->state.role == R_SECONDARY) {
- drbd_delete_device(mdev_to_minor(mdev));
+ retcode = adm_delete_minor(adm_ctx.mdev);
+out:
+ drbd_adm_finish(info, retcode);
+ return 0;
+}
+
+int drbd_adm_down(struct sk_buff *skb, struct genl_info *info)
+{
+ int retcode; /* enum drbd_ret_code rsp. enum drbd_state_rv */
+ struct drbd_conf *mdev;
+ unsigned i;
+
+ retcode = drbd_adm_prepare(skb, info, 0);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ if (!adm_ctx.tconn) {
+ retcode = ERR_RES_NOT_KNOWN;
+ goto out;
+ }
+
+ /* demote */
+ idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) {
+ retcode = drbd_set_role(mdev, R_SECONDARY, 0);
+ if (retcode < SS_SUCCESS) {
+ drbd_msg_put_info("failed to demote");
+ goto out;
+ }
+ }
+
+ retcode = conn_try_disconnect(adm_ctx.tconn, 0);
+ if (retcode < SS_SUCCESS) {
+ drbd_msg_put_info("failed to disconnect");
+ goto out;
+ }
+
+ /* detach */
+ idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) {
+ retcode = adm_detach(mdev, 0);
+ if (retcode < SS_SUCCESS || retcode > NO_ERROR) {
+ drbd_msg_put_info("failed to detach");
+ goto out;
+ }
+ }
+
+ /* If we reach this, all volumes (of this tconn) are Secondary,
+ * Disconnected, Diskless, aka Unconfigured. Make sure all threads have
+ * actually stopped, state handling only does drbd_thread_stop_nowait(). */
+ drbd_thread_stop(&adm_ctx.tconn->worker);
+
+ /* Now, nothing can fail anymore */
+
+ /* delete volumes */
+ idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) {
+ retcode = adm_delete_minor(mdev);
+ if (retcode != NO_ERROR) {
+ /* "can not happen" */
+ drbd_msg_put_info("failed to delete volume");
+ goto out;
+ }
+ }
+
+ /* delete connection */
+ if (conn_lowest_minor(adm_ctx.tconn) < 0) {
+ list_del_rcu(&adm_ctx.tconn->all_tconn);
+ synchronize_rcu();
+ kref_put(&adm_ctx.tconn->kref, &conn_destroy);
+
retcode = NO_ERROR;
- } else
- retcode = ERR_MINOR_CONFIGURED;
+ } else {
+ /* "can not happen" */
+ retcode = ERR_RES_IN_USE;
+ drbd_msg_put_info("failed to delete connection");
+ }
+ goto out;
out:
drbd_adm_finish(info, retcode);
return 0;
}
-int drbd_adm_delete_connection(struct sk_buff *skb, struct genl_info *info)
+int drbd_adm_del_resource(struct sk_buff *skb, struct genl_info *info)
{
enum drbd_ret_code retcode;
- retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_RESOURCE);
if (!adm_ctx.reply_skb)
return retcode;
if (retcode != NO_ERROR)
goto out;
if (conn_lowest_minor(adm_ctx.tconn) < 0) {
- drbd_free_tconn(adm_ctx.tconn);
+ list_del_rcu(&adm_ctx.tconn->all_tconn);
+ synchronize_rcu();
+ kref_put(&adm_ctx.tconn->kref, &conn_destroy);
+
retcode = NO_ERROR;
} else {
- retcode = ERR_CONN_IN_USE;
+ retcode = ERR_RES_IN_USE;
}
+ if (retcode == NO_ERROR)
+ drbd_thread_stop(&adm_ctx.tconn->worker);
out:
drbd_adm_finish(info, retcode);
return 0;
unsigned seq;
int err = -ENOMEM;
+ if (sib->sib_reason == SIB_SYNC_PROGRESS) {
+ if (time_after(jiffies, mdev->rs_last_bcast + HZ))
+ mdev->rs_last_bcast = jiffies;
+ else
+ return;
+ }
+
seq = atomic_inc_return(&drbd_genl_seq);
msg = genlmsg_new(NLMSG_GOODSIZE, GFP_NOIO);
if (!msg)
if (!d_out) /* cannot happen, but anyways. */
goto nla_put_failure;
d_out->minor = mdev_to_minor(mdev);
- d_out->ret_code = 0;
+ d_out->ret_code = NO_ERROR;
if (nla_put_status_info(msg, mdev, sib))
goto nla_put_failure;
projects / firefly-linux-kernel-4.4.55.git / blobdiff