iscsi: Perform connection failure entirely in kernel space
PATCH 1/1 iscsi: Perform connection failure entirely in kernel space
From: Bharath Ravi <[email protected]>
Connection failure processing depends on a daemon being present to (at least) stop the connection and start recovery. This is a problem on a multipath scenario, where if the daemon failed for whatever reason, the SCSI path is never marked as down, multipath won't perform the failover and IO to the device will be forever waiting for that connection to come back.
This patch implements an optional feature in the iscsi module, to perform the connection failure inside the kernel. This way, the failover can happen and pending IO can continue even if the daemon is dead. Once the daemon comes alive again, it can perform recovery procedures if applicable.
--- drivers/scsi/scsi_transport_iscsi.c | 46 +++++++++++++++++++++++++++++ include/scsi/scsi_transport_iscsi.h | 1 + 2 files changed, 47 insertions(+)
drivers/scsi/scsi_transport_iscsi.c +46 -0
@@ -36,6 +36,12 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(iscsi_dbg_session);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_tcp);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_sw_tcp); +static bool kern_conn_failure; +module_param (kern_conn_failure, bool,S_IRUGO |S_IWUSR ); +MODULE_PARM_DESC (kern_conn_failure, +"Allow the kernel to detect and disable broken connections " +"without requiring userspace intervention" ); +static int dbg_session;module_param_named (debug_session, dbg_session, int,S_IRUGO |S_IWUSR );
@@ -84,6 +90,12 @@ struct iscsi_internal {struct transport_container session_cont; }; +/* Worker to perform connection failure on unresponsive connections +* completely in kernel space. +*/ +static void stop_conn_work_fn (struct work_struct *work); +static DECLARE_WORK(stop_conn_work, stop_conn_work_fn); +static atomic_t iscsi_session_nr;/* sysfs session id for next new session */ static struct workqueue_struct *iscsi_eh_timer_workq;
@@ -1609,6 +1621,7 @@ static DEFINE_MUTEX(rx_queue_mutex);static LIST_HEAD (sesslist);static DEFINE_SPINLOCK (sesslock);static LIST_HEAD (connlist); +static LIST_HEAD (connlist_err);static DEFINE_SPINLOCK (connlock);static uint32_t iscsi_conn_get_sid (struct iscsi_cls_conn *conn)
@@ -2245,6 +2258,7 @@ iscsi_create_conn(struct iscsi_cls_session *session, int dd_size, uint32_t cid)mutex_init (&conn->ep_mutex);INIT_LIST_HEAD (&conn->conn_list); +INIT_LIST_HEAD (&conn->conn_list_err); conn->transport = transport; conn->cid = cid;
@@ -2291,6 +2305,7 @@ int iscsi_destroy_conn(struct iscsi_cls_conn *conn)spin_lock_irqsave (&connlock, flags);list_del (&conn->conn_list); +list_del (&conn->conn_list_err);spin_unlock_irqrestore (&connlock, flags);transport_unregister_device (&conn->dev);
@@ -2405,6 +2420,28 @@ int iscsi_offload_mesg(struct Scsi_Host *shost, }EXPORT_SYMBOL_GPL (iscsi_offload_mesg); +static void stop_conn_work_fn (struct work_struct *work) +{ +struct iscsi_cls_conn *conn, *tmp; +unsigned long flags; +LIST_HEAD (recovery_list); + +spin_lock_irqsave (&connlock, flags); +if (list_empty (&connlist_err)) { +spin_unlock_irqrestore (&connlock, flags); +return ; + } +list_splice_init (&connlist_err, &recovery_list); +spin_unlock_irqrestore (&connlock, flags); + +mutex_lock (&rx_queue_mutex); +list_for_each_entry_safe (conn, tmp, &recovery_list, conn_list_err) { + conn->transport->stop_conn (conn,STOP_CONN_RECOVER ); +list_del_init (&conn->conn_list_err); + } +mutex_unlock (&rx_queue_mutex); +} +void iscsi_conn_error_event (struct iscsi_cls_conn *conn,enum iscsi_err error) {struct nlmsghdr *nlh;
@@ -2412,6 +2449,15 @@ void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error)struct iscsi_uevent *ev;struct iscsi_internal *priv;int len =nlmsg_total_size (sizeof (*ev)); +unsigned long flags; + +if (kern_conn_failure) { +spin_lock_irqsave (&connlock, flags); +list_add (&conn->conn_list_err, &connlist_err); +spin_unlock_irqrestore (&connlock, flags); + +queue_work (system_unbound_wq, &stop_conn_work); + } priv =iscsi_if_transport_lookup (conn->transport);if (!priv)
include/scsi/scsi_transport_iscsi.h +1 -0
@@ -190,6 +190,7 @@ extern void iscsi_ping_comp_event(uint32_t host_no,struct iscsi_cls_conn {struct list_head conn_list;/* item in connlist */ +struct list_head conn_list_err;/* item in connlist_err */ void *dd_data;/* LLD private data */ struct iscsi_transport *transport;uint32_t cid;/* connection id */
Signature · 1 line
2.24.0
Lee Duncan
maintainerQuoted 9 lines · Gabriel Krisman Bertazi
From: Bharath Ravi <[email protected]>
Connection failure processing depends on a daemon being present to (at least) stop the connection and start recovery. This is a problem on a multipath scenario, where if the daemon failed for whatever reason, the SCSI path is never marked as down, multipath won't perform the failover and IO to the device will be forever waiting for that connection to come back.
Quoted 5 lines
This patch implements an optional feature in the iscsi module, to perform the connection failure inside the kernel. This way, the failover can happen and pending IO can continue even if the daemon is dead. Once the daemon comes alive again, it can perform recovery procedures if applicable.
I don't suppose you've tested how this plays with the daemon, when present?
Quoted 137 lines
Co-developed-by Dave Clausen <[email protected]>Signed-off-by Dave Clausen <[email protected]>Co-developed-by Nick Black <[email protected]>Signed-off-by Nick Black <[email protected]>Co-developed-by Vaibhav Nagarnaik <[email protected]>Signed-off-by Vaibhav Nagarnaik <[email protected]>Co-developed-by Anatol Pomazau <[email protected]>Signed-off-by Anatol Pomazau <[email protected]>Co-developed-by Tahsin Erdogan <[email protected]>Signed-off-by Tahsin Erdogan <[email protected]>Co-developed-by Frank Mayhar <[email protected]>Signed-off-by Frank Mayhar <[email protected]>Co-developed-by Junho Ryu <[email protected]>Signed-off-by Junho Ryu <[email protected]>Co-developed-by Khazhismel Kumykov <[email protected]>Signed-off-by Khazhismel Kumykov <[email protected]>Signed-off-by Bharath Ravi <[email protected]>Co-developed-by Gabriel Krisman Bertazi <[email protected]>Signed-off-by Gabriel Krisman Bertazi <[email protected]>--- drivers/scsi/scsi_transport_iscsi.c | 46 +++++++++++++++++++++++++++++ include/scsi/scsi_transport_iscsi.h | 1 + 2 files changed, 47 insertions(+)drivers/scsi/scsi_transport_iscsi.c +46
-0@@ -36,6 +36,12 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(iscsi_dbg_session);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_tcp);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_sw_tcp); +static bool kern_conn_failure; +module_param (kern_conn_failure, bool,S_IRUGO |S_IWUSR ); +MODULE_PARM_DESC (kern_conn_failure, +"Allow the kernel to detect and disable broken connections " +"without requiring userspace intervention" ); +static int dbg_session;module_param_named (debug_session, dbg_session, int,S_IRUGO |S_IWUSR );@@ -84,6 +90,12 @@ struct iscsi_internal {struct transport_container session_cont; }; +/* Worker to perform connection failure on unresponsive connections +* completely in kernel space. +*/ +static void stop_conn_work_fn (struct work_struct *work); +static DECLARE_WORK(stop_conn_work, stop_conn_work_fn); +static atomic_t iscsi_session_nr;/* sysfs session id for next new session */ static struct workqueue_struct *iscsi_eh_timer_workq;@@ -1609,6 +1621,7 @@ static DEFINE_MUTEX(rx_queue_mutex);static LIST_HEAD (sesslist);static DEFINE_SPINLOCK (sesslock);static LIST_HEAD (connlist); +static LIST_HEAD (connlist_err);static DEFINE_SPINLOCK (connlock);static uint32_t iscsi_conn_get_sid (struct iscsi_cls_conn *conn)@@ -2245,6 +2258,7 @@ iscsi_create_conn(struct iscsi_cls_session *session, int dd_size, uint32_t cid)mutex_init (&conn->ep_mutex);INIT_LIST_HEAD (&conn->conn_list); +INIT_LIST_HEAD (&conn->conn_list_err); conn->transport = transport; conn->cid = cid;@@ -2291,6 +2305,7 @@ int iscsi_destroy_conn(struct iscsi_cls_conn *conn)spin_lock_irqsave (&connlock, flags);list_del (&conn->conn_list); +list_del (&conn->conn_list_err);spin_unlock_irqrestore (&connlock, flags);transport_unregister_device (&conn->dev);@@ -2405,6 +2420,28 @@ int iscsi_offload_mesg(struct Scsi_Host *shost, }EXPORT_SYMBOL_GPL (iscsi_offload_mesg); +static void stop_conn_work_fn (struct work_struct *work) +{ +struct iscsi_cls_conn *conn, *tmp; +unsigned long flags; +LIST_HEAD (recovery_list); + +spin_lock_irqsave (&connlock, flags); +if (list_empty (&connlist_err)) { +spin_unlock_irqrestore (&connlock, flags); +return ; + } +list_splice_init (&connlist_err, &recovery_list); +spin_unlock_irqrestore (&connlock, flags); + +mutex_lock (&rx_queue_mutex); +list_for_each_entry_safe (conn, tmp, &recovery_list, conn_list_err) { + conn->transport->stop_conn (conn,STOP_CONN_RECOVER ); +list_del_init (&conn->conn_list_err); + } +mutex_unlock (&rx_queue_mutex); +} +void iscsi_conn_error_event (struct iscsi_cls_conn *conn,enum iscsi_err error) {struct nlmsghdr *nlh;@@ -2412,6 +2449,15 @@ void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error)struct iscsi_uevent *ev;struct iscsi_internal *priv;int len =nlmsg_total_size (sizeof (*ev)); +unsigned long flags; + +if (kern_conn_failure) { +spin_lock_irqsave (&connlock, flags); +list_add (&conn->conn_list_err, &connlist_err); +spin_unlock_irqrestore (&connlock, flags); + +queue_work (system_unbound_wq, &stop_conn_work); + } priv =iscsi_if_transport_lookup (conn->transport);if (!priv)include/scsi/scsi_transport_iscsi.h +1
-0@@ -190,6 +190,7 @@ extern void iscsi_ping_comp_event(uint32_t host_no,struct iscsi_cls_conn {struct list_head conn_list;/* item in connlist */ +struct list_head conn_list_err;/* item in connlist_err */ void *dd_data;/* LLD private data */ struct iscsi_transport *transport;uint32_t cid;/* connection id */
Quoted 10 lines · Lee Duncan
Quoted 9 lines · Gabriel Krisman Bertazi
From: Bharath Ravi <[email protected]>
Connection failure processing depends on a daemon being present to (at least) stop the connection and start recovery. This is a problem on a multipath scenario, where if the daemon failed for whatever reason, the SCSI path is never marked as down, multipath won't perform the failover and IO to the device will be forever waiting for that connection to come back.
Quoted 8 lines
Quoted 5 lines
This patch implements an optional feature in the iscsi module, to perform the connection failure inside the kernel. This way, the failover can happen and pending IO can continue even if the daemon is dead. Once the daemon comes alive again, it can perform recovery procedures if applicable.
I don't suppose you've tested how this plays with the daemon, when present?
Hello,
Yes, I did. It seemed to work properly over TCP.
The stop calls are serialized by the rx_mutex, whoever gets it first, gets the job done. The second should return immediately since the socket is no longer there.
What am I missing?
Signature · 1 line
Gabriel Krisman Bertazi
Quoted 126 lines · Gabriel Krisman Bertazi
From: Bharath Ravi <[email protected]>
Connection failure processing depends on a daemon being present to (at least) stop the connection and start recovery. This is a problem on a multipath scenario, where if the daemon failed for whatever reason, the SCSI path is never marked as down, multipath won't perform the failover and IO to the device will be forever waiting for that connection to come back.
This patch implements an optional feature in the iscsi module, to perform the connection failure inside the kernel. This way, the failover can happen and pending IO can continue even if the daemon is dead. Once the daemon comes alive again, it can perform recovery procedures if applicable.
Co-developed-by Dave Clausen <[email protected]>Signed-off-by Dave Clausen <[email protected]>Co-developed-by Nick Black <[email protected]>Signed-off-by Nick Black <[email protected]>Co-developed-by Vaibhav Nagarnaik <[email protected]>Signed-off-by Vaibhav Nagarnaik <[email protected]>Co-developed-by Anatol Pomazau <[email protected]>Signed-off-by Anatol Pomazau <[email protected]>Co-developed-by Tahsin Erdogan <[email protected]>Signed-off-by Tahsin Erdogan <[email protected]>Co-developed-by Frank Mayhar <[email protected]>Signed-off-by Frank Mayhar <[email protected]>Co-developed-by Junho Ryu <[email protected]>Signed-off-by Junho Ryu <[email protected]>Co-developed-by Khazhismel Kumykov <[email protected]>Signed-off-by Khazhismel Kumykov <[email protected]>Signed-off-by Bharath Ravi <[email protected]>Co-developed-by Gabriel Krisman Bertazi <[email protected]>Signed-off-by Gabriel Krisman Bertazi <[email protected]>--- drivers/scsi/scsi_transport_iscsi.c | 46 +++++++++++++++++++++++++++++ include/scsi/scsi_transport_iscsi.h | 1 + 2 files changed, 47 insertions(+)drivers/scsi/scsi_transport_iscsi.c +37
-0@@ -36,6 +36,12 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(iscsi_dbg_session);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_tcp);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_sw_tcp); +static bool kern_conn_failure; +module_param (kern_conn_failure, bool,S_IRUGO |S_IWUSR ); +MODULE_PARM_DESC (kern_conn_failure, +"Allow the kernel to detect and disable broken connections " +"without requiring userspace intervention" ); +static int dbg_session;module_param_named (debug_session, dbg_session, int,S_IRUGO |S_IWUSR );@@ -84,6 +90,12 @@ struct iscsi_internal {struct transport_container session_cont; }; +/* Worker to perform connection failure on unresponsive connections +* completely in kernel space. +*/ +static void stop_conn_work_fn (struct work_struct *work); +static DECLARE_WORK(stop_conn_work, stop_conn_work_fn); +static atomic_t iscsi_session_nr;/* sysfs session id for next new session */ static struct workqueue_struct *iscsi_eh_timer_workq;@@ -1609,6 +1621,7 @@ static DEFINE_MUTEX(rx_queue_mutex);static LIST_HEAD (sesslist);static DEFINE_SPINLOCK (sesslock);static LIST_HEAD (connlist); +static LIST_HEAD (connlist_err);static DEFINE_SPINLOCK (connlock);static uint32_t iscsi_conn_get_sid (struct iscsi_cls_conn *conn)@@ -2245,6 +2258,7 @@ iscsi_create_conn(struct iscsi_cls_session *session, int dd_size, uint32_t cid)mutex_init (&conn->ep_mutex);INIT_LIST_HEAD (&conn->conn_list); +INIT_LIST_HEAD (&conn->conn_list_err); conn->transport = transport; conn->cid = cid;@@ -2291,6 +2305,7 @@ int iscsi_destroy_conn(struct iscsi_cls_conn *conn)spin_lock_irqsave (&connlock, flags);list_del (&conn->conn_list); +list_del (&conn->conn_list_err);spin_unlock_irqrestore (&connlock, flags);transport_unregister_device (&conn->dev);@@ -2405,6 +2420,28 @@ int iscsi_offload_mesg(struct Scsi_Host *shost, }EXPORT_SYMBOL_GPL (iscsi_offload_mesg); +static void stop_conn_work_fn (struct work_struct *work) +{ +struct iscsi_cls_conn *conn, *tmp; +unsigned long flags; +LIST_HEAD (recovery_list); + +spin_lock_irqsave (&connlock, flags); +if (list_empty (&connlist_err)) { +spin_unlock_irqrestore (&connlock, flags); +return ; + } +list_splice_init (&connlist_err, &recovery_list); +spin_unlock_irqrestore (&connlock, flags); + +mutex_lock (&rx_queue_mutex); +list_for_each_entry_safe (conn, tmp, &recovery_list, conn_list_err) { + conn->transport->stop_conn (conn,STOP_CONN_RECOVER ); +list_del_init (&conn->conn_list_err); + } +mutex_unlock (&rx_queue_mutex); +} +void iscsi_conn_error_event (struct iscsi_cls_conn *conn,enum iscsi_err error) {struct nlmsghdr *nlh;@@ -2412,6 +2449,15 @@ void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error)struct iscsi_uevent *ev;struct iscsi_internal *priv;int len =nlmsg_total_size (sizeof (*ev));
drivers/scsi/scsi_transport_iscsi.c void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error) · line 2461
Quoted 10 lines
drivers/scsi/scsi_transport_iscsi.c +9
-0+unsigned long flags; + +if (kern_conn_failure) { +spin_lock_irqsave (&connlock, flags); +list_add (&conn->conn_list_err, &connlist_err); +spin_unlock_irqrestore (&connlock, flags); + +queue_work (system_unbound_wq, &stop_conn_work); + }
Do you need the modparam? I think you could handle this issue and the similar one during shutdown at the same time, and you would always want to do the kernel based error handler when userspace is not answering for both cases.
You could do the following:
- Modify __iscsi_block_session so it does the stop_conn callout instead of reverse, and change the iscsi_stop_conn/ISCSI_UEVENT_STOP_CONN:
related code accordingly.
- In iscsi_conn_error_event you would then do:
iscsi_multicast_skb();iscsi_block_session();
- You can then drop the system_state check in iscsi_eh_cmd_timed_out because those running commands are always handled by the stop_conn call in __iscsi_block_session now.
Quoted 143 lines · Mike Christie
Quoted 126 lines · Gabriel Krisman Bertazi
From: Bharath Ravi <[email protected]>
Connection failure processing depends on a daemon being present to (at least) stop the connection and start recovery. This is a problem on a multipath scenario, where if the daemon failed for whatever reason, the SCSI path is never marked as down, multipath won't perform the failover and IO to the device will be forever waiting for that connection to come back.
This patch implements an optional feature in the iscsi module, to perform the connection failure inside the kernel. This way, the failover can happen and pending IO can continue even if the daemon is dead. Once the daemon comes alive again, it can perform recovery procedures if applicable.
Co-developed-by Dave Clausen <[email protected]>Signed-off-by Dave Clausen <[email protected]>Co-developed-by Nick Black <[email protected]>Signed-off-by Nick Black <[email protected]>Co-developed-by Vaibhav Nagarnaik <[email protected]>Signed-off-by Vaibhav Nagarnaik <[email protected]>Co-developed-by Anatol Pomazau <[email protected]>Signed-off-by Anatol Pomazau <[email protected]>Co-developed-by Tahsin Erdogan <[email protected]>Signed-off-by Tahsin Erdogan <[email protected]>Co-developed-by Frank Mayhar <[email protected]>Signed-off-by Frank Mayhar <[email protected]>Co-developed-by Junho Ryu <[email protected]>Signed-off-by Junho Ryu <[email protected]>Co-developed-by Khazhismel Kumykov <[email protected]>Signed-off-by Khazhismel Kumykov <[email protected]>Signed-off-by Bharath Ravi <[email protected]>Co-developed-by Gabriel Krisman Bertazi <[email protected]>Signed-off-by Gabriel Krisman Bertazi <[email protected]>--- drivers/scsi/scsi_transport_iscsi.c | 46 +++++++++++++++++++++++++++++ include/scsi/scsi_transport_iscsi.h | 1 + 2 files changed, 47 insertions(+)drivers/scsi/scsi_transport_iscsi.c +37
-0@@ -36,6 +36,12 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(iscsi_dbg_session);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_tcp);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_sw_tcp); +static bool kern_conn_failure; +module_param (kern_conn_failure, bool,S_IRUGO |S_IWUSR ); +MODULE_PARM_DESC (kern_conn_failure, +"Allow the kernel to detect and disable broken connections " +"without requiring userspace intervention" ); +static int dbg_session;module_param_named (debug_session, dbg_session, int,S_IRUGO |S_IWUSR );@@ -84,6 +90,12 @@ struct iscsi_internal {struct transport_container session_cont; }; +/* Worker to perform connection failure on unresponsive connections +* completely in kernel space. +*/ +static void stop_conn_work_fn (struct work_struct *work); +static DECLARE_WORK(stop_conn_work, stop_conn_work_fn); +static atomic_t iscsi_session_nr;/* sysfs session id for next new session */ static struct workqueue_struct *iscsi_eh_timer_workq;@@ -1609,6 +1621,7 @@ static DEFINE_MUTEX(rx_queue_mutex);static LIST_HEAD (sesslist);static DEFINE_SPINLOCK (sesslock);static LIST_HEAD (connlist); +static LIST_HEAD (connlist_err);static DEFINE_SPINLOCK (connlock);static uint32_t iscsi_conn_get_sid (struct iscsi_cls_conn *conn)@@ -2245,6 +2258,7 @@ iscsi_create_conn(struct iscsi_cls_session *session, int dd_size, uint32_t cid)mutex_init (&conn->ep_mutex);INIT_LIST_HEAD (&conn->conn_list); +INIT_LIST_HEAD (&conn->conn_list_err); conn->transport = transport; conn->cid = cid;@@ -2291,6 +2305,7 @@ int iscsi_destroy_conn(struct iscsi_cls_conn *conn)spin_lock_irqsave (&connlock, flags);list_del (&conn->conn_list); +list_del (&conn->conn_list_err);spin_unlock_irqrestore (&connlock, flags);transport_unregister_device (&conn->dev);@@ -2405,6 +2420,28 @@ int iscsi_offload_mesg(struct Scsi_Host *shost, }EXPORT_SYMBOL_GPL (iscsi_offload_mesg); +static void stop_conn_work_fn (struct work_struct *work) +{ +struct iscsi_cls_conn *conn, *tmp; +unsigned long flags; +LIST_HEAD (recovery_list); + +spin_lock_irqsave (&connlock, flags); +if (list_empty (&connlist_err)) { +spin_unlock_irqrestore (&connlock, flags); +return ; + } +list_splice_init (&connlist_err, &recovery_list); +spin_unlock_irqrestore (&connlock, flags); + +mutex_lock (&rx_queue_mutex); +list_for_each_entry_safe (conn, tmp, &recovery_list, conn_list_err) { + conn->transport->stop_conn (conn,STOP_CONN_RECOVER ); +list_del_init (&conn->conn_list_err); + } +mutex_unlock (&rx_queue_mutex); +} +void iscsi_conn_error_event (struct iscsi_cls_conn *conn,enum iscsi_err error) {struct nlmsghdr *nlh;@@ -2412,6 +2449,15 @@ void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error)struct iscsi_uevent *ev;struct iscsi_internal *priv;int len =nlmsg_total_size (sizeof (*ev));drivers/scsi/scsi_transport_iscsi.c void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error) · line 2461
Quoted 10 lines
drivers/scsi/scsi_transport_iscsi.c +9
-0+unsigned long flags; + +if (kern_conn_failure) { +spin_lock_irqsave (&connlock, flags); +list_add (&conn->conn_list_err, &connlist_err); +spin_unlock_irqrestore (&connlock, flags); + +queue_work (system_unbound_wq, &stop_conn_work); + }Do you need the modparam? I think you could handle this issue and the similar one during shutdown at the same time, and you would always want to do the kernel based error handler when userspace is not answering for both cases.
Quoted 5 lines
You could do the following:
- Modify
__iscsi_block_sessionso it does thestop_conncallout instead of reverse, and change the iscsi_stop_conn/ISCSI_UEVENT_STOP_CONN:
related code accordingly.
Oh yeah, on second thought, I think I like how your new function above calls into the stop_conn callout and everything works like it did before. Ignore the __iscsi_block_session changes. But, I would drop the modparam, always queue the work, and then fix up the system_state check.
Quoted 10 lines
- In
iscsi_conn_error_eventyou would then do:
iscsi_multicast_skb();iscsi_block_session();- You can then drop the
system_statecheck iniscsi_eh_cmd_timed_outbecause those running commands are always handled by thestop_conncall in__iscsi_block_sessionnow.
Quoted 108 lines · Gabriel Krisman Bertazi
From: Bharath Ravi <[email protected]>
Connection failure processing depends on a daemon being present to (at least) stop the connection and start recovery. This is a problem on a multipath scenario, where if the daemon failed for whatever reason, the SCSI path is never marked as down, multipath won't perform the failover and IO to the device will be forever waiting for that connection to come back.
This patch implements an optional feature in the iscsi module, to perform the connection failure inside the kernel. This way, the failover can happen and pending IO can continue even if the daemon is dead. Once the daemon comes alive again, it can perform recovery procedures if applicable.
Co-developed-by Dave Clausen <[email protected]>Signed-off-by Dave Clausen <[email protected]>Co-developed-by Nick Black <[email protected]>Signed-off-by Nick Black <[email protected]>Co-developed-by Vaibhav Nagarnaik <[email protected]>Signed-off-by Vaibhav Nagarnaik <[email protected]>Co-developed-by Anatol Pomazau <[email protected]>Signed-off-by Anatol Pomazau <[email protected]>Co-developed-by Tahsin Erdogan <[email protected]>Signed-off-by Tahsin Erdogan <[email protected]>Co-developed-by Frank Mayhar <[email protected]>Signed-off-by Frank Mayhar <[email protected]>Co-developed-by Junho Ryu <[email protected]>Signed-off-by Junho Ryu <[email protected]>Co-developed-by Khazhismel Kumykov <[email protected]>Signed-off-by Khazhismel Kumykov <[email protected]>Signed-off-by Bharath Ravi <[email protected]>Co-developed-by Gabriel Krisman Bertazi <[email protected]>Signed-off-by Gabriel Krisman Bertazi <[email protected]>--- drivers/scsi/scsi_transport_iscsi.c | 46 +++++++++++++++++++++++++++++ include/scsi/scsi_transport_iscsi.h | 1 + 2 files changed, 47 insertions(+)drivers/scsi/scsi_transport_iscsi.c +25
-0@@ -36,6 +36,12 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(iscsi_dbg_session);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_tcp);EXPORT_TRACEPOINT_SYMBOL_GPL (iscsi_dbg_sw_tcp); +static bool kern_conn_failure; +module_param (kern_conn_failure, bool,S_IRUGO |S_IWUSR ); +MODULE_PARM_DESC (kern_conn_failure, +"Allow the kernel to detect and disable broken connections " +"without requiring userspace intervention" ); +static int dbg_session;module_param_named (debug_session, dbg_session, int,S_IRUGO |S_IWUSR );@@ -84,6 +90,12 @@ struct iscsi_internal {struct transport_container session_cont; }; +/* Worker to perform connection failure on unresponsive connections +* completely in kernel space. +*/ +static void stop_conn_work_fn (struct work_struct *work); +static DECLARE_WORK(stop_conn_work, stop_conn_work_fn); +static atomic_t iscsi_session_nr;/* sysfs session id for next new session */ static struct workqueue_struct *iscsi_eh_timer_workq;@@ -1609,6 +1621,7 @@ static DEFINE_MUTEX(rx_queue_mutex);static LIST_HEAD (sesslist);static DEFINE_SPINLOCK (sesslock);static LIST_HEAD (connlist); +static LIST_HEAD (connlist_err);static DEFINE_SPINLOCK (connlock);static uint32_t iscsi_conn_get_sid (struct iscsi_cls_conn *conn)@@ -2245,6 +2258,7 @@ iscsi_create_conn(struct iscsi_cls_session *session, int dd_size, uint32_t cid)mutex_init (&conn->ep_mutex);INIT_LIST_HEAD (&conn->conn_list); +INIT_LIST_HEAD (&conn->conn_list_err); conn->transport = transport; conn->cid = cid;@@ -2291,6 +2305,7 @@ int iscsi_destroy_conn(struct iscsi_cls_conn *conn)spin_lock_irqsave (&connlock, flags);list_del (&conn->conn_list); +list_del (&conn->conn_list_err);spin_unlock_irqrestore (&connlock, flags);transport_unregister_device (&conn->dev);@@ -2405,6 +2420,28 @@ int iscsi_offload_mesg(struct Scsi_Host *shost, }EXPORT_SYMBOL_GPL (iscsi_offload_mesg); +static void stop_conn_work_fn (struct work_struct *work) +{ +struct iscsi_cls_conn *conn, *tmp; +unsigned long flags; +LIST_HEAD (recovery_list); + +spin_lock_irqsave (&connlock, flags); +if (list_empty (&connlist_err)) { +spin_unlock_irqrestore (&connlock, flags); +return ;
drivers/scsi/scsi_transport_iscsi.c int iscsi_offload_mesg(struct Scsi_Host *shost, · line 2442
Quoted 10 lines
drivers/scsi/scsi_transport_iscsi.c +10
-0+ } +list_splice_init (&connlist_err, &recovery_list); +spin_unlock_irqrestore (&connlock, flags); + +mutex_lock (&rx_queue_mutex); +list_for_each_entry_safe (conn, tmp, &recovery_list, conn_list_err) { + conn->transport->stop_conn (conn,STOP_CONN_RECOVER ); +list_del_init (&conn->conn_list_err); + } +mutex_unlock (&rx_queue_mutex);
Holding rx_queue_mutex for the entire conn_list_err may be problematic for long conn_list_err, could we drop on need_resched perhaps, or otherwise limit how long we hold this?
Quoted 36 lines
drivers/scsi/scsi_transport_iscsi.c +11
-0+} +void iscsi_conn_error_event (struct iscsi_cls_conn *conn,enum iscsi_err error) {struct nlmsghdr *nlh;@@ -2412,6 +2449,15 @@ void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error)struct iscsi_uevent *ev;struct iscsi_internal *priv;int len =nlmsg_total_size (sizeof (*ev)); +unsigned long flags; + +if (kern_conn_failure) { +spin_lock_irqsave (&connlock, flags); +list_add (&conn->conn_list_err, &connlist_err); +spin_unlock_irqrestore (&connlock, flags); + +queue_work (system_unbound_wq, &stop_conn_work); + } priv =iscsi_if_transport_lookup (conn->transport);if (!priv)include/scsi/scsi_transport_iscsi.h +1
-1@@ -190,6 +190,7 @@ extern void iscsi_ping_comp_event(uint32_t host_no,struct iscsi_cls_conn {struct list_head conn_list;/* item in connlist */ +struct list_head conn_list_err;/* item in connlist_err */ void *dd_data;/* LLD private data */ struct iscsi_transport *transport;uint32_t cid;/* connection id */ --2.24.0