Move the scsi_free_scsi_task() call from receive_copy_results() to the
callers of this function to avoid that accessing the unmarshalled data
triggers a use-after-free.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
This patch avoids that Valgrind reports that an uninitialized buffer
is passed to writev():
Syscall param writev(vector[...]) points to uninitialised byte(s)
at 0x5557D07: writev (in /lib64/libc-2.19.so)
by 0x5269BFD: iscsi_iovector_readv_writev (socket.c:509)
by 0x526A7A9: iscsi_write_to_socket (socket.c:727)
by 0x526AE47: iscsi_service (socket.c:888)
by 0x526B632: event_loop (sync.c:67)
by 0x526D925: iscsi_scsi_command_sync (sync.c:1223)
by 0x405970: send_scsi_command (iscsi-support.c:276)
by 0x40B190: extendedcopy (iscsi-support.c:2673)
by 0x40E9FE: test_extendedcopy_param (test_extendedcopy_param.c:63)
by 0x503E260: ??? (in /usr/lib64/libcunit.so.1.0.1)
by 0x503E56E: ??? (in /usr/lib64/libcunit.so.1.0.1)
by 0x503E8A5: CU_run_all_tests (in /usr/lib64/libcunit.so.1.0.1)
Address 0xffefff864 is on thread 1's stack
in frame #8, created by test_extendedcopy_param (test_extendedcopy_param.c:31)
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
The test tool is simplistic and uses the synchronous API. In this API
we will only poll() the socket and process events during the lifetime of any
iscsi_*_sync() calls.
A problem with this approach arises with multipath tests. When iscsi-test-cu
is used with multiple devices we will first open a session for each
multipath leg, but then as most tests are only using the first device we end
up with creating secondary sessions we don't do any iscsi_*_sync() calls on
for very long times. If a target is set up to send NOPs on these
secondary paths it will not get any responses and eventually tear down the
connection.
To avoid this, try to make sure we service any secondary paths we have
created by calling iscsi_service(POLLIN|POLLOUT) every few seconds
to make sure we do service and respond to any NOPs.
Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Normal applications want the current behaviour where we have the library
consume any/all of the UnitAttentions that the target may have queued on the
initial connection, but when we reconnect the session after a failure the
library will pass all the UAs back to the application to process.
Some applications, such as the test suite or really trivial applications
might not want to have to deal with handling of UAs and just "make it work".
Those applications can now request that upon any reconnection of the session
that libiscsi will automatically consume any and all UAs and hide them from
the application.
Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
If a previous run of the test tool had been interrupted one or
more persistent reservation keys can still be registered. This
can cause tests in the new run to fail. Avoid this by clearing
these keys before running any tests. Do this only if --dataloss
has been specified.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Avoid that running the regression tests against a LUN with size zero triggers
the following assertion failure:
lt-iscsi-test-cu: iscsi-support.c:1666: read6: Assertion `task != ((void *)0)' failed.
Aborted
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
This patch does not change any functionality but removes the number
of casts in the test tool source code.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Check that the Persistent Reserve Out PREEMT request removes the
registration referenced by the service action key.
Signed-off-by: David Disseldorp <ddiss@suse.de>
test_iscsi_tur_until_good() dispatches TUR requests until the target has
cleared all UAs for the given sd, or the maximum number of retries is
reached.
This helper function is useful for any test that needs to deal with UAs
(e.g. PRs), so should be moved out into common code.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Issue a Persistent Reservation Out PREEMPT request, with keys and type
provided.
PREEMPT allows for the removal of registrations, and replacement of
reservations.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Check that the Persistent Reserve In REPORT CAPABILITIES response
carries valid LENGTH, ALLOW COMMANDS, and PERSISTENT RESERVATION TYPE
MASK fields.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Issue a Persistent Reservation In REPORT CAPABILITIES request and parse
the response. Callers can obtain the unmarshalled response data via the
_rcaps parameter.
Signed-off-by: David Disseldorp <ddiss@suse.de>
This test currently reconnects to the base sd for multi-initiator
testing. Use the secondary multipath device instead, if available.
Signed-off-by: David Disseldorp <ddiss@suse.de>
These tests currently reconnect to the base sd for multi-initiator
testing. Use the secondary multipath device instead, if available.
Signed-off-by: David Disseldorp <ddiss@suse.de>
These tests currently reconnect to the base sd for multi-initiator
testing. Use the secondary multipath device instead, if available.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Many tests currently reconnect to the base sd for multi-initiator
testing. Use the secondary multipath device instead, if available.
Signed-off-by: David Disseldorp <ddiss@suse.de>
A number of existing tests create a secondary iSCSI session by logging
into the target specified in the base sd during the test.
With multipath sessions now connected during iscsi-test-cu startup
(when launched with multiple URIs), secondary sessions can be
provided via multipath when available, rather than reconnecting.
This change adds a helper that conditionally returns the secondary
multipath device if available, or falls back to creating a new session
via the base sd target details.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Check that a Persistent Reserve Out CLEAR request:
- clears an existing reservation
- clears existing registration keys
- bumps the PRgeneration number
Signed-off-by: David Disseldorp <ddiss@suse.de>
Confirm that data is consistent across paths when using the atomic
COMPARE AND WRITE operation.
This should be extended in future to issue the requests simultaneously
across both paths, in an attempt to trigger a read/write race.
Signed-off-by: David Disseldorp <ddiss@suse.de>
This test issues a logical unit reset iSCSI TMF request, and then
confirms that all paths report a subsequent unit attention condition.
Signed-off-by: David Disseldorp <ddiss@suse.de>
MPATH_SKIP_UNLESS_ISCSI(_sds, _num_sds) iterates over all _sds, and
skips the test unless all paths are iSCSI based.
Signed-off-by: David Disseldorp <ddiss@suse.de>
From Alexander Motin <mav@freebsd.org>
Treat BlockLimits VPD atomic granulatity/alignment reported as 0
(no restrictions) as == 1 (also means no restriction).
Both 0 and 1 mean that there is no restriction on alignment or
granularity but with the difference being that the second value can
be used to increment to the next valid LBA.
Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Avoid that this test crashes as follows if the RECEIVE COPY RESULTS
command is not supported:
[FAILED] RECEIVECOPYRESULT command failed with sense. (null)
Segmentation fault
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: Sushma Gurram <Sushma.Gurram@sandisk.com>
