Tests: ex_librpmem_basic/TEST1, ex_librpmem_hello/TEST1, ex_librpmem_fibonacci/TEST[0-2] fail.

[Greg091]

ISSUE:

Environment Information

• PMDK package version(s): 1.6-141-g416e88f9f
• OS(es) version(s): Debian
• ndctl version(s): 64.1.1.g43bf086
• kernel version(s): 4.19.0-2-amd64
• bash version(s): 5.0.2(1)-release (x86_64-pc-linux-gnu)
• compiler, libraries, packaging and other related tools version(s):

Please provide a reproduction of the bug:

./RUNTESTS test_name -s test_number, example:

./RUNTESTS ex_librpmem_basic -s TEST1

How often bug is revealed: (always, often, rare): always

Actual behavior:

All the above tests fail in the same way. Below log of execution of ex_librpmem_basic TEST1.

./RUNTESTS ex_librpmem_basic -s TEST1
ex_librpmem_basic/TEST1: SETUP (short/non-pmem/debug/sockets/GPSPM)
../unittest/unittest.sh: line 2070: no match: pool_hdr.checksum_gen\(\)
RUNTESTS: stopping: ex_librpmem_basic/TEST1 failed, TEST=short FS=non-pmem BUILD=debug RPMEM_PROVIDER=sockets RPMEM_PM=GPSPM
ex_librpmem_basic/TEST1: SETUP (short/non-pmem/debug/sockets/APM)
../unittest/unittest.sh: line 2070: no match: pool_hdr.checksum_gen\(\)
RUNTESTS: stopping: ex_librpmem_basic/TEST1 failed, TEST=short FS=non-pmem BUILD=debug RPMEM_PROVIDER=sockets RPMEM_PM=APM
./RUNTESTS: line 142: /tmp/data-location.RlLrpOye: No such file or directory

Expected behavior:

Tests should pass.

Details

Additional information about Priority and Help Requested:

Are you willing to submit a pull request with a proposed change? (Yes, No)

Requested priority: (Showstopper, High, Medium, Low)

[Greg091]

Verified on: 1.6.1-253-ga6ad15e55. Still fails.

[janekmi]

The issue is caused by special characters passed multiple times as a function argument, through SSH and to the shell on the remote node. I see two possible solutions:

• get rid of the special characters from pmemspoil arguments e.g. "(", ")"
• do not require to pass these characters through the arguments e.g. by allowing providing commands from a file

[janekmi]

Please note the issue appears also in other tests:

$ ./RUNTESTS pmempool_sync_remote -s TEST20
$ ./RUNTESTS pmempool_sync_remote -s TEST21

[Greg091]

Verified on: 1.6.1-338-g392789288