lib/iscsi: Fix unregistered login timer when exiting

Request logout asynchronously when exiting the
running connection. The login timer should be
unregistered before exiting.

Change-Id: I3df28f7104

lib/iscsi: Fix unregistered login timer when exiting

Request logout asynchronously when exiting the
running connection. The login timer should be
unregistered before exiting.

Change-Id: I3df28f7104315da915f488378d4faeb713e7709d
Signed-off-by: sunzhenliang <hisunzhenliang@outlook.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25262
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
Community-CI: Community CI Samsung <spdk.community.ci.samsung@gmail.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot

show more ...

lib/iscsi: Fix: Unregister logout timer

Unregister the logout timer like the login timer.

Change-Id: I6331243f7ff68d86e2aab73c18dad1c95af76ad4
Signed-off-by: sunzhenliang <hisunzhenliang@outlook.co

lib/iscsi: Fix: Unregister logout timer

Unregister the logout timer like the login timer.

Change-Id: I6331243f7ff68d86e2aab73c18dad1c95af76ad4
Signed-off-by: sunzhenliang <hisunzhenliang@outlook.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24990
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

trace: do not define fn in SPDK_TRACE_REGISTER_FN

Newer lcov try to detect issues with function definitions, and throw
warnings if function definitions have overlapping lines. Since
SPDK_TRACE_REGIS

trace: do not define fn in SPDK_TRACE_REGISTER_FN

Newer lcov try to detect issues with function definitions, and throw
warnings if function definitions have overlapping lines. Since
SPDK_TRACE_REGISTER_FN defines both the constructor function and
the trace function, it treats this as an overlap.

Note this is the only one of our REGISTER functions in SPDK that
effectively defines two functions in the same macro.

So change the macro so that the user actually defines the
trace function instead.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: Ib9d0d1af7674c5ff9b65c12c1e4a5b05e6e110f5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24836
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot

show more ...

lib/util: Make md5 API public

Make md5 realted functions in iscsi lib public.
Interface remained unchaged except of spdk_ prefix
added to func names. Implementation slightly updated
with unlikely()

lib/util: Make md5 API public

Make md5 realted functions in iscsi lib public.
Interface remained unchaged except of spdk_ prefix
added to func names. Implementation slightly updated
with unlikely() hints.

Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: I8bff008aad0c44b639b9249e1b48f3a6f9421f74
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23133
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Ben Walker <ben@nvidia.com>

show more ...

iscsi: Schedule active target to idlest poll group

We used round robin to schedule target connections. However, this
had problems so that targets are not balanced equally among poll groups
when targ

iscsi: Schedule active target to idlest poll group

We used round robin to schedule target connections. However, this
had problems so that targets are not balanced equally among poll groups
when target connection up/down was repeated.

Fix the problem by scheduling target connections to the idlest poll
group.

target->num_active_targets is incremented only if target has the first
active connection. If connection is deleted before becoming active,
target->num_active_conns is not incremented. Add a variable scheduled to
iscsi_conn structure to cover such case. Get the global mutex to decrement
pg->num_active_targets and keep the same ordering, i.e. get first and free
last global mutex, as iscsi_conn_schedule().

Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Suggested-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I0671583424d9f08e93839718c543ee3325f179b4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24611
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

env: add SPDK_ENV_NUMA_ID_ANY and replace socket_id with numa_id

We will try to avoid further proliferation of "SOCKET_ID" to refer
to a NUMA socket ID moving forward, and just use "NUMA_ID" to avoi

env: add SPDK_ENV_NUMA_ID_ANY and replace socket_id with numa_id

We will try to avoid further proliferation of "SOCKET_ID" to refer
to a NUMA socket ID moving forward, and just use "NUMA_ID" to avoid
confusion with TCP sockets.

Change all of the existing in-tree SPDK_ENV_SOCKET_ID_ANY uses to
SPDK_ENV_NUMA_ID_ANY, but keep the old #define around, at least for
now. Also change all 'socket_id' parameters to 'numa_id'.

We still have spdk_env_get_socket_id(), we will need to keep this
but next patch will add spdk_env_get_numa_id().

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: Idc31c29e32b708c24d88f9c6fecaf9a99e34ba1e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24607
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Ben Walker <ben@nvidia.com>
Community-CI: Mellanox Build Bot

show more ...

iscs: chap mutual authentication should apply when configured.

mutual chap authentication should be used when the -m option is sent
via iscsi_auth_group_add_secret rpc.
The commit comes to fix the p

iscs: chap mutual authentication should apply when configured.

mutual chap authentication should be used when the -m option is sent
via iscsi_auth_group_add_secret rpc.
The commit comes to fix the problem where the mutual authentication
depends solely on the existance of the muser + msecret configured in the auth
group.

Change-Id: I8d703ea2b643508c97e7e64bd2194e47cc6ab8cd
Signed-off-by: Oded Mashbach <oded.mashbach@silk.us>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23617
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

iscsi: Authenticating discovery based on givven credentials.

This commits verifies that iSCSI CHAP authentication functions correctly
during the discovery phase.

Previously, configuring CHAP using

iscsi: Authenticating discovery based on givven credentials.

This commits verifies that iSCSI CHAP authentication functions correctly
during the discovery phase.

Previously, configuring CHAP using iscsi_set_discovery_auth did not
enforce authentication for host discovery sessions. This occurred
because the connection object's CHAP property remained unset.

The commit ensures that updating the connection's CHAP properties, when
discovery session negotiates its chap params - addresses this issue and enforces CHAP
authentication during discovery.

Change-Id: I1f9e59a1b7a4a507b3c9ddcd2a6e4b070176c8c3
Signed-off-by: Oded Mashbach <oded.mashbach@silk.us>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23616
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

lib/iscsi: add json config saving for iscsi_enable_histogram

When the "save_config" RPC command is called in SPDK,
the current environment is saved. However, the environment
created by "iscsi_enable

lib/iscsi: add json config saving for iscsi_enable_histogram

When the "save_config" RPC command is called in SPDK,
the current environment is saved. However, the environment
created by "iscsi_enable_histogram" is not included
in the saved configuration.

In a production environment, it is crucial to monitor
various IO indicators, and IO latency is one of the
important metrics. Fortunately, SPDK provides an RPC
method called "iscsi_get_histogram" to obtain the IO
latency histogram of an iscsi target device. However, before
using this RPC method, it is necessary to first call
the "iscsi_enable_histogram" command.

In a production environment, SPDK is typically configured
to automatically restart using the saved configuration
file (by using the "save_config" command in rpc.py).
However, since the "iscsi_enable_histogram" configuration
is missing from the saved configuration, the IO latency
monitoring will fail once SPDK restarts.

Change-Id: I3fe79debc337ae42f99cd23856a2cee405dbb3cd
Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23486
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>

show more ...

lib/iscsi: set target_short_name when target_name missing ":"

When ":" is missing in the target_name of iscsi,
the target_short_name of the target is empty.
In above case, it is difficult to test. S

lib/iscsi: set target_short_name when target_name missing ":"

When ":" is missing in the target_name of iscsi,
the target_short_name of the target is empty.
In above case, it is difficult to test. So it is
set to target_name by default.

Change-Id: Ifd81a7436a9472bfadd6217564bb7b7a0ab22dff
Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23443
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Community-CI: Mellanox Build Bot

show more ...

iscsi: add trace owner support

Note: spdk_trace_record() in the fuzzing app really shouldn't be
there, so this patch removes it.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I4b0da

iscsi: add trace owner support

Note: spdk_trace_record() in the fuzzing app really shouldn't be
there, so this patch removes it.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I4b0dad56775beee3563d15786c4d0ccd41142997
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22892
Community-CI: Mellanox Build Bot
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

lib/iscsi: add rpc method iscsi_get_histogram

A new rpc method(iscsi_get_histogram) is added to get
histogram of iscsi target.

Change-Id: I4bbda04763e4db734b798657c8acb9e702c69f83
Signed-off-by: Ch

lib/iscsi: add rpc method iscsi_get_histogram

A new rpc method(iscsi_get_histogram) is added to get
histogram of iscsi target.

Change-Id: I4bbda04763e4db734b798657c8acb9e702c69f83
Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21001
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

lib/iscsi: add rpc method iscsi_enable_histogram

SPDK provides a histogram function about io delay in the bdev layer,
but when using the iscsi protocol, especially when the io size is large
and io s

lib/iscsi: add rpc method iscsi_enable_histogram

SPDK provides a histogram function about io delay in the bdev layer,
but when using the iscsi protocol, especially when the io size is large
and io splitting is required, we need to be more concerned about
the end-to-end io latency of the iscsi layer.

An example of WRITE16 command with bulk data:
I T
WRITE16 ->
<- R2T
DATAOUT ->
DATAOUT ->
...
DATAOUT(FINAL) ->

Therefore, a new rpc method is added to control the opening or
closing of the histogram function.

Change-Id: Iec96a54a8da2a58c0542e7b38e2161ade505f09f
Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21002
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

trace: rename owner to owner_type

We will be adding the real concept of an "owner" in upcoming patches. So change
the existing "owner" to "owner_type" which more accurate reflects its meaning.
For e

trace: rename owner to owner_type

We will be adding the real concept of an "owner" in upcoming patches. So change
the existing "owner" to "owner_type" which more accurate reflects its meaning.
For example, an "owner_type" can be OWNER_BDEV, but an actual "owner" would
be some specific bdev identified by its name.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: Ieb25eb5a574a88b3812291f2345796e5bb66a54f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22645
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

iscsi: associate trace ISCSI_PDU object with SOCK_REQ

This enables mapping sock request events with their respective iscsi_pdu
object.

This removes the need for the iSCSI FLUSH_WRITEBUF events - th

iscsi: associate trace ISCSI_PDU object with SOCK_REQ

This enables mapping sock request events with their respective iscsi_pdu
object.

This removes the need for the iSCSI FLUSH_WRITEBUF events - the new sock
request events serve the exact same purpose - so remove them.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I73ad7ef903edb65e0bedaecabcde85c34fd2ca9a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22107
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <ben@nvidia.com>
Reviewed-by: Seung yeon Shin <syeon.shin@samsung.com>

show more ...

lib/iscsi: Add RPC method iscsi_get_stats

In a production environment, the iscsi_get_connections RPC method is
commonly used to monitor the number of SPDK connections. It is often
necessary to issue

lib/iscsi: Add RPC method iscsi_get_stats

In a production environment, the iscsi_get_connections RPC method is
commonly used to monitor the number of SPDK connections. It is often
necessary to issue an alarm when there are too many iSCSI connections.
By default, the maximum number of connections in SPDK is defined by
MAX_ISCSI_CONNECTIONS as 1024.

However, the return value of the iscsi_get_connections method contains
a significant amount of information. In certain cases, only the number
of connections is required, and the overhead of parsing JSON becomes
a considerable burden, especially when there are numerous connections.

To address this issue, the rpc method iscsi_get_stats has been added.
When this method is used, only the count of connections in each state
will be provided, eliminating the need for detailed information parsing.

Change-Id: I45f4a4ebb2d4a5a6fe9eee28a3cbeb4e05009b56
Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21629
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

so_ver: increase all major versions after SPDK 24.01

To allow SO_MINOR updates on LTS for the whole year it is supported,
the major version for all components needs to be increased.
This is to preve

so_ver: increase all major versions after SPDK 24.01

To allow SO_MINOR updates on LTS for the whole year it is supported,
the major version for all components needs to be increased.
This is to prevent scenario where two versions exists with matching
versions, but conflicting ABI.
Ex. Next SPDK release adds an API call increasing the minor version,
then LTS needs just a subset of those additions.

Increasing major so version after LTS, allows the future releases
to update versions as needed. Yet allowing LTS to increase minor
version separately.

This patch:
- increases SO_VER by 1 for all components
- resets SO_MINOR to 0 for all components

Due to patch below being introduced after v24.01 code freeze,
bdev lib version should not be increased:
61623c5ca9 bdev: add spdk_bdev_io padding in place of io_submit_ch

Short reference to how the versions were changed:
MAX=$(git grep "SO_VER := " | cut -d" " -f 3 | sort -ubnr | head -1)
for((i=$MAX;i>0;i-=1)); do find . -name "Makefile" -exec \
sed -i -e "s/SO_VER := $i\$/SO_VER := $(($i+1))/g" {} +; done
find . -name "Makefile" -exec \
sed -i -e "s/SO_MINOR := .*/SO_MINOR := 0/g" {} +

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I0166d6be6c17bbbf985d48b5dfcb36f1d4af1b48
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21582
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

include: Remove duplicate includes in source files

Change-Id: I7dd6ae6fa11603a956c3d178b9b23d2c755913d1
Signed-off-by: Pierre Lestringant <plestringant@kalray.eu>
Reviewed-on: https://review.spdk.io

include: Remove duplicate includes in source files

Change-Id: I7dd6ae6fa11603a956c3d178b9b23d2c755913d1
Signed-off-by: Pierre Lestringant <plestringant@kalray.eu>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20106
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

iscsi: cleanup backend on accessed exclusively

Multiple initiators share a target:

+--+ +--+ +--+ +--+
|Im| ... |Ix| |Iy| |Iz|
+--+ +--+ +--+ +--+
| | |

iscsi: cleanup backend on accessed exclusively

Multiple initiators share a target:

+--+ +--+ +--+ +--+
|Im| ... |Ix| |Iy| |Iz|
+--+ +--+ +--+ +--+
| | | |
\ | / /
+---------------------+
| Target |
+---------------------+
|
+---------------------+
| Bdev |
+---------------------+

Once initiatorM(Im) gets disconnected, SPDK tries to cancel the
inflight IOs of Im by resetting backend bdev. Unfortunately, the
inflight IOs of Ix, Iy and Iz are canceled together,
'SENSE KEY:COMMAND ABORTED' error floods at that moment.

To avoid this issue, drain the inflight IOs of the exited connection
intead.

Change-Id: Ic7668d0be13052b6f4e3b0fd52c21db38f22ecf0
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20469
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>

show more ...

scsi: Add SPDK_SCSI_TASK_FUNC_TARGET_RESET

SCSI/iSCSI triggers bdev_scsi_reset() in two reasons:
- mgmt command from initiator side
- target side cleanup

For example, once we hit the following logs

scsi: Add SPDK_SCSI_TASK_FUNC_TARGET_RESET

SCSI/iSCSI triggers bdev_scsi_reset() in two reasons:
- mgmt command from initiator side
- target side cleanup

For example, once we hit the following logs:
nvme_ctrlr.c:1639:nvme_ctrlr_disconnect: *NOTICE*: [0000:b2:00.0] resetting controller
bdev_nvme.c:2027:_bdev_nvme_reset_ctrlr_complete: *NOTICE*: Resetting controller successful.

We need to know why the bdev/nvme resets, rather than a single
SPDK_SCSI_TASK_FUNC_LUN_RESET reason, separate it into two to
distinguish the root cause of the lower layer bdev reset.

Change-Id: I087add31c31e46780c6e7c60cc8af14290e84636
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20361
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

iscsi/portal_grp: improve port string validation

Detect bad port strings before trying
spdk_sock_listen() to improve the error messages
for the user.

This also will reject '0' and '-0' as port numb

iscsi/portal_grp: improve port string validation

Detect bad port strings before trying
spdk_sock_listen() to improve the error messages
for the user.

This also will reject '0' and '-0' as port numbers,
similar to recent fixes made for nvmf/tcp.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I8ea8a7fecd8a25020ab8505dec22f53b700622c9
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20528
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Community-CI: Mellanox Build Bot

show more ...

iscsi: fix possible login state error

The following steps are expected on iSCSI login:
Initiator Target
login req ->
set CONN STATE INVALID
<-

iscsi: fix possible login state error

The following steps are expected on iSCSI login:
Initiator Target
login req ->
set CONN STATE INVALID
<- login resp
...
PDU write cb
set CONN STATE RUNNING
CHAP req ->
Test CONN STATE(expected RUNNING)

Because of socket pending list for asynchronous requests mechanism[1],
unexpected INVALID state maybe hit on CHAP request:
Initiator Target
login req ->
set CONN STATE INVALID
<- login resp
CHAP req ->
Test CONN STATE(unexpected INVALID)
...
PDU write cb (delayed)
set CONN STATE RUNNING

This leads iscsi_op_login_phase_none() re-entry, a CHAP request
gets error with log "InitiatorName is empty".

In fact, once the target handles LOGIN request and prepare to send
response, it has already becomes RUNNING state. Set RUNNING state
immediately instead of callback funcion, any unlikely error occurs
before callback, EXITING will overwrite the state.

[1]: SPDK commit: ab22d249e2162b5f79ad0e7007de480a7f8a2c18

Change-Id: Ic41575cf600478496046888b6d5806e6beb09aea
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19944
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

iscsi: Move connection to EXITING if opening LUNs failed

Opening LUNs is a fatal error. However, the error had been ignored.
Fix the bug in this patch. Change the return type of
iscsi_conn_open_luns

iscsi: Move connection to EXITING if opening LUNs failed

Opening LUNs is a fatal error. However, the error had been ignored.
Fix the bug in this patch. Change the return type of
iscsi_conn_open_luns() from void to int, and change
iscsi_conn_open_luns() to return 0 if success or -1 otherwise.

iscsi_conn_full_feature_migrate() sets the state of the connection to
EXITING if opening LUNs failed. Then, the first poll in the assigned
poll group destructs the connection.

Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I8d64786287f5e1ad9120651439dd21282278cb2d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19382
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Vasuki Manikarnike <vasuki.manikarnike@hpe.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>

show more ...