- discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] Only Run as root
by Tanner L. Nelson 17 Feb '25

17 Feb '25

Good morning, We have a small cluster in an isolated environment. Our lead has spearheaded a reinstall to get away from Bright Cluster Manager and CentOS. This is the first time our project has done anything like this, and we have hit a snag that we believe is MPI-related. As root: * Job runs as expected (verified with 2 nodes) * mpirun with host file, same result As user account: * Single node works * 2 nodes, job hangs with no error output that we can find To us, and our colleagues at our main Computing Center, it seems related to permissions/authentication. But everything we know to check has been checked several times (i.e. user can ssh to nodes 001-002 where root was successful, user was not; permissions on mpi directory 755, even tried 777 just to rule it out). Details I believe may be asked: * RHEL9 * Mpich4.2.3 was compiled with gcc11.5.0 * Consistent results with scheduler and mpirun * Tested HPL and our primary application, both built against the same mpich version Hoping we can get some suggestions. Thanks in advance! Respectfully, Tanner Nelson

1 0

Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3
by Zhou, Hui 23 Dec '24

23 Dec '24

It is not lost. It is just not backported to 4.2.x. It is in the current 4.3.0rc1 release. Hui ________________________________ From: Edric Ellis <eellis(a)mathworks.com> Sent: Monday, December 23, 2024 9:25 AM To: Zhou, Hui <zhouh(a)anl.gov>; discuss(a)mpich.org <discuss(a)mpich.org> Subject: RE: Problems with "make check" in romio building MPICH-4.2.3 Exactly that. However, a change that should have fixed it (which set FILENAME to “”) was already submitted to “runtests. in” here: https: //github. com/pmodels/mpich/commit/90b0b0614039295d5bde1a7a8b2d430813448467 However, that change isn’t present ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Exactly that. However, a change that should have fixed it (which set FILENAME to “”) was already submitted to “runtests.in” here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061… <https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061…> However, that change isn’t present in the mpich-4.2.3 download. I don’t quite follow how it got lost. But you can see it isn’t present here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/v4.2.3/src… <https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/v4.2.3/src…> Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov> Sent: 23 December 2024 15:13 To: discuss(a)mpich.org Cc: Edric Ellis <eellis(a)mathworks.com> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 I see! It is the * in "rm -f ${FILENAME}*" causing the issue. We should either remove the "*" or rename "test_hintfile". Sorry for being dense. 🙂 Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Monday, December 23, 2024 1:46 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Yes, runtests deletes the file! As mentioned below, the runtests scripts states “FILENAME=test”, and then at various stages, runtests executes “rm -f ${FILENAME}*” Cheers, Edric. From: Zhou, Hui <zhouh@ anl. gov> Sent: 17 December 2024 16: 11 ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Yes, runtests deletes the file! As mentioned below, the runtests scripts states “FILENAME=test”, and then at various stages, runtests executes “rm -f ${FILENAME}*” Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>> Sent: 17 December 2024 16:11 To: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>>; discuss(a)mpich.org<mailto:[email protected]> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Thanks for the test log. The test-suite.log is complaining "test_hintfile" not exist. This is not a temporary file created during testing, but part of the static test files. Something is causing the tests not able to open "/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile". Any clues? Hui ________________________________ From: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Sent: Tuesday, December 17, 2024 9:49 AM To: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>>; discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Subject: RE: Problems with "make check" in romio building MPICH-4.2.3 I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4. 2. 3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary for ROMIO 4.2.3 ============================================================================ # TOTAL: 1 # PASS: 0 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 ============================================================================ See test/test-suite.log Please report to discuss(a)mpich.org<mailto:[email protected]> ============================================================================ make[2]: *** [Makefile:1167: test-suite.log] Error 1 and the test-suite.log has this: **** Testing syshints **** MPI_File_open(MPI_COMM_WORLD, argv[1], MPI_MODE_RDONLY, info_mine, &fh): File does not exist, error stack: ADIOI_NFS_OPEN(37): File /mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile does not exist Abort(1) on node 0 (rank 0 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0 FAIL runtests (exit status: 1) The file test_hintfile did exist before running the tests. Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>> Sent: 13 December 2024 16:04 To: discuss(a)mpich.org<mailto:[email protected]> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Could you show the relevant console log for the failure? Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Friday, December 13, 2024 2:59 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Hi mpich-discuss, "make check" is failing with mpich-4. 2. 3, in the src/mpi/romio/test subdirectory. I think it's because this file: https: //urldefense. us/v3/__https: //github. com/pmodels/mpich/blob/4. 2. x/src/mpi/romio/test/runtests. in__;!!G_uCfscf7eWS!c86jBfO2_nn7840CXToEK-Dl8-y3_afw3ARA6rXELHbuMyXHtcqIGwBNDcw6oUUN8brlyugyCb5NBiY$ ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hi mpich-discuss, "make check" is failing with mpich-4.2.3, in the src/mpi/romio/test subdirectory. I think it's because this file: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/… still says "FILENAME=test" on line 25. (The test fails because "FILENAME=test" eventually causes "test_hintfile" to be delete before it's needed by one of the tests). There's a change over here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061… to make it say "FILENAME=", which I think would fix things. I'm not sufficiently familiar with github to fathom how that change didn't make it in to 4.2.3 when this change to the Makefile https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a… (which effectively makes top-level "make check" enter the romio/test subdirectory) did make it in to 4.2.3... Cheers, Edric.

1 0

Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3
by Edric Ellis 23 Dec '24

23 Dec '24

Exactly that. However, a change that should have fixed it (which set FILENAME to “”) was already submitted to “runtests.in” here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061… However, that change isn’t present in the mpich-4.2.3 download. I don’t quite follow how it got lost. But you can see it isn’t present here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/v4.2.3/src… Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov> Sent: 23 December 2024 15:13 To: discuss(a)mpich.org Cc: Edric Ellis <eellis(a)mathworks.com> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 I see! It is the * in "rm -f ${FILENAME}*" causing the issue. We should either remove the "*" or rename "test_hintfile". Sorry for being dense. 🙂 Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Monday, December 23, 2024 1:46 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Yes, runtests deletes the file! As mentioned below, the runtests scripts states “FILENAME=test”, and then at various stages, runtests executes “rm -f ${FILENAME}*” Cheers, Edric. From: Zhou, Hui <zhouh@ anl. gov> Sent: 17 December 2024 16: 11 ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Yes, runtests deletes the file! As mentioned below, the runtests scripts states “FILENAME=test”, and then at various stages, runtests executes “rm -f ${FILENAME}*” Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>> Sent: 17 December 2024 16:11 To: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>>; discuss(a)mpich.org<mailto:[email protected]> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Thanks for the test log. The test-suite.log is complaining "test_hintfile" not exist. This is not a temporary file created during testing, but part of the static test files. Something is causing the tests not able to open "/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile". Any clues? Hui ________________________________ From: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Sent: Tuesday, December 17, 2024 9:49 AM To: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>>; discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Subject: RE: Problems with "make check" in romio building MPICH-4.2.3 I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4. 2. 3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary for ROMIO 4.2.3 ============================================================================ # TOTAL: 1 # PASS: 0 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 ============================================================================ See test/test-suite.log Please report to discuss(a)mpich.org<mailto:[email protected]> ============================================================================ make[2]: *** [Makefile:1167: test-suite.log] Error 1 and the test-suite.log has this: **** Testing syshints **** MPI_File_open(MPI_COMM_WORLD, argv[1], MPI_MODE_RDONLY, info_mine, &fh): File does not exist, error stack: ADIOI_NFS_OPEN(37): File /mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile does not exist Abort(1) on node 0 (rank 0 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0 FAIL runtests (exit status: 1) The file test_hintfile did exist before running the tests. Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>> Sent: 13 December 2024 16:04 To: discuss(a)mpich.org<mailto:[email protected]> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Could you show the relevant console log for the failure? Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Friday, December 13, 2024 2:59 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Hi mpich-discuss, "make check" is failing with mpich-4. 2. 3, in the src/mpi/romio/test subdirectory. I think it's because this file: https: //urldefense. us/v3/__https: //github. com/pmodels/mpich/blob/4. 2. x/src/mpi/romio/test/runtests. in__;!!G_uCfscf7eWS!c86jBfO2_nn7840CXToEK-Dl8-y3_afw3ARA6rXELHbuMyXHtcqIGwBNDcw6oUUN8brlyugyCb5NBiY$ ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hi mpich-discuss, "make check" is failing with mpich-4.2.3, in the src/mpi/romio/test subdirectory. I think it's because this file: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/…> still says "FILENAME=test" on line 25. (The test fails because "FILENAME=test" eventually causes "test_hintfile" to be delete before it's needed by one of the tests). There's a change over here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061…> to make it say "FILENAME=", which I think would fix things. I'm not sufficiently familiar with github to fathom how that change didn't make it in to 4.2.3 when this change to the Makefile https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a…> (which effectively makes top-level "make check" enter the romio/test subdirectory) did make it in to 4.2.3... Cheers, Edric.

1 0

Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3
by Zhou, Hui 23 Dec '24

23 Dec '24

I see! It is the * in "rm -f ${FILENAME}*" causing the issue. We should either remove the "*" or rename "test_hintfile". Sorry for being dense. 🙂 Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org> Sent: Monday, December 23, 2024 1:46 AM To: discuss(a)mpich.org <discuss(a)mpich.org> Cc: Edric Ellis <eellis(a)mathworks.com> Subject: Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Yes, runtests deletes the file! As mentioned below, the runtests scripts states “FILENAME=test”, and then at various stages, runtests executes “rm -f ${FILENAME}*” Cheers, Edric. From: Zhou, Hui <zhouh@ anl. gov> Sent: 17 December 2024 16: 11 ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Yes, runtests deletes the file! As mentioned below, the runtests scripts states “FILENAME=test”, and then at various stages, runtests executes “rm -f ${FILENAME}*” Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov> Sent: 17 December 2024 16:11 To: Edric Ellis <eellis(a)mathworks.com>; discuss(a)mpich.org Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Thanks for the test log. The test-suite.log is complaining "test_hintfile" not exist. This is not a temporary file created during testing, but part of the static test files. Something is causing the tests not able to open "/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile". Any clues? Hui ________________________________ From: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Sent: Tuesday, December 17, 2024 9:49 AM To: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>>; discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Subject: RE: Problems with "make check" in romio building MPICH-4.2.3 I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4. 2. 3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary for ROMIO 4.2.3 ============================================================================ # TOTAL: 1 # PASS: 0 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 ============================================================================ See test/test-suite.log Please report to discuss(a)mpich.org<mailto:[email protected]> ============================================================================ make[2]: *** [Makefile:1167: test-suite.log] Error 1 and the test-suite.log has this: **** Testing syshints **** MPI_File_open(MPI_COMM_WORLD, argv[1], MPI_MODE_RDONLY, info_mine, &fh): File does not exist, error stack: ADIOI_NFS_OPEN(37): File /mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile does not exist Abort(1) on node 0 (rank 0 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0 FAIL runtests (exit status: 1) The file test_hintfile did exist before running the tests. Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>> Sent: 13 December 2024 16:04 To: discuss(a)mpich.org<mailto:[email protected]> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Could you show the relevant console log for the failure? Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Friday, December 13, 2024 2:59 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Hi mpich-discuss, "make check" is failing with mpich-4. 2. 3, in the src/mpi/romio/test subdirectory. I think it's because this file: https: //urldefense. us/v3/__https: //github. com/pmodels/mpich/blob/4. 2. x/src/mpi/romio/test/runtests. in__;!!G_uCfscf7eWS!c86jBfO2_nn7840CXToEK-Dl8-y3_afw3ARA6rXELHbuMyXHtcqIGwBNDcw6oUUN8brlyugyCb5NBiY$ ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hi mpich-discuss, "make check" is failing with mpich-4.2.3, in the src/mpi/romio/test subdirectory. I think it's because this file: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/… still says "FILENAME=test" on line 25. (The test fails because "FILENAME=test" eventually causes "test_hintfile" to be delete before it's needed by one of the tests). There's a change over here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061… to make it say "FILENAME=", which I think would fix things. I'm not sufficiently familiar with github to fathom how that change didn't make it in to 4.2.3 when this change to the Makefile https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a… (which effectively makes top-level "make check" enter the romio/test subdirectory) did make it in to 4.2.3... Cheers, Edric.

1 0

Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3
by Edric Ellis 23 Dec '24

23 Dec '24

Yes, runtests deletes the file! As mentioned below, the runtests scripts states “FILENAME=test”, and then at various stages, runtests executes “rm -f ${FILENAME}*” Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov> Sent: 17 December 2024 16:11 To: Edric Ellis <eellis(a)mathworks.com>; discuss(a)mpich.org Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Thanks for the test log. The test-suite.log is complaining "test_hintfile" not exist. This is not a temporary file created during testing, but part of the static test files. Something is causing the tests not able to open "/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile". Any clues? Hui ________________________________ From: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Sent: Tuesday, December 17, 2024 9:49 AM To: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>>; discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Subject: RE: Problems with "make check" in romio building MPICH-4.2.3 I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4. 2. 3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary for ROMIO 4.2.3 ============================================================================ # TOTAL: 1 # PASS: 0 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 ============================================================================ See test/test-suite.log Please report to discuss(a)mpich.org<mailto:[email protected]> ============================================================================ make[2]: *** [Makefile:1167: test-suite.log] Error 1 and the test-suite.log has this: **** Testing syshints **** MPI_File_open(MPI_COMM_WORLD, argv[1], MPI_MODE_RDONLY, info_mine, &fh): File does not exist, error stack: ADIOI_NFS_OPEN(37): File /mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile does not exist Abort(1) on node 0 (rank 0 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0 FAIL runtests (exit status: 1) The file test_hintfile did exist before running the tests. Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>> Sent: 13 December 2024 16:04 To: discuss(a)mpich.org<mailto:[email protected]> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Could you show the relevant console log for the failure? Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Friday, December 13, 2024 2:59 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Hi mpich-discuss, "make check" is failing with mpich-4. 2. 3, in the src/mpi/romio/test subdirectory. I think it's because this file: https: //urldefense. us/v3/__https: //github. com/pmodels/mpich/blob/4. 2. x/src/mpi/romio/test/runtests. in__;!!G_uCfscf7eWS!c86jBfO2_nn7840CXToEK-Dl8-y3_afw3ARA6rXELHbuMyXHtcqIGwBNDcw6oUUN8brlyugyCb5NBiY$ ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hi mpich-discuss, "make check" is failing with mpich-4.2.3, in the src/mpi/romio/test subdirectory. I think it's because this file: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/…> still says "FILENAME=test" on line 25. (The test fails because "FILENAME=test" eventually causes "test_hintfile" to be delete before it's needed by one of the tests). There's a change over here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061…> to make it say "FILENAME=", which I think would fix things. I'm not sufficiently familiar with github to fathom how that change didn't make it in to 4.2.3 when this change to the Makefile https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a…> (which effectively makes top-level "make check" enter the romio/test subdirectory) did make it in to 4.2.3... Cheers, Edric.

1 0

Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3
by Zhou, Hui 17 Dec '24

17 Dec '24

Hi Edric, Thanks for the test log. The test-suite.log is complaining "test_hintfile" not exist. This is not a temporary file created during testing, but part of the static test files. Something is causing the tests not able to open "/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile". Any clues? Hui ________________________________ From: Edric Ellis <eellis(a)mathworks.com> Sent: Tuesday, December 17, 2024 9:49 AM To: Zhou, Hui <zhouh(a)anl.gov>; discuss(a)mpich.org <discuss(a)mpich.org> Subject: RE: Problems with "make check" in romio building MPICH-4.2.3 I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4. 2. 3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary for ROMIO 4.2.3 ============================================================================ # TOTAL: 1 # PASS: 0 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 ============================================================================ See test/test-suite.log Please report to discuss(a)mpich.org<mailto:[email protected]> ============================================================================ make[2]: *** [Makefile:1167: test-suite.log] Error 1 and the test-suite.log has this: **** Testing syshints **** MPI_File_open(MPI_COMM_WORLD, argv[1], MPI_MODE_RDONLY, info_mine, &fh): File does not exist, error stack: ADIOI_NFS_OPEN(37): File /mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile does not exist Abort(1) on node 0 (rank 0 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0 FAIL runtests (exit status: 1) The file test_hintfile did exist before running the tests. Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov> Sent: 13 December 2024 16:04 To: discuss(a)mpich.org Cc: Edric Ellis <eellis(a)mathworks.com> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Could you show the relevant console log for the failure? Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Friday, December 13, 2024 2:59 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Hi mpich-discuss, "make check" is failing with mpich-4. 2. 3, in the src/mpi/romio/test subdirectory. I think it's because this file: https: //urldefense. us/v3/__https: //github. com/pmodels/mpich/blob/4. 2. x/src/mpi/romio/test/runtests. in__;!!G_uCfscf7eWS!c86jBfO2_nn7840CXToEK-Dl8-y3_afw3ARA6rXELHbuMyXHtcqIGwBNDcw6oUUN8brlyugyCb5NBiY$ ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hi mpich-discuss, "make check" is failing with mpich-4.2.3, in the src/mpi/romio/test subdirectory. I think it's because this file: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/… still says "FILENAME=test" on line 25. (The test fails because "FILENAME=test" eventually causes "test_hintfile" to be delete before it's needed by one of the tests). There's a change over here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061… to make it say "FILENAME=", which I think would fix things. I'm not sufficiently familiar with github to fathom how that change didn't make it in to 4.2.3 when this change to the Makefile https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a… (which effectively makes top-level "make check" enter the romio/test subdirectory) did make it in to 4.2.3... Cheers, Edric.

1 0

Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3
by Edric Ellis 17 Dec '24

17 Dec '24

I see this: make[2]: Entering directory '/mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test' FAIL: runtests ============================================================================ Testsuite summary for ROMIO 4.2.3 ============================================================================ # TOTAL: 1 # PASS: 0 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 ============================================================================ See test/test-suite.log Please report to discuss(a)mpich.org<mailto:[email protected]> ============================================================================ make[2]: *** [Makefile:1167: test-suite.log] Error 1 and the test-suite.log has this: **** Testing syshints **** MPI_File_open(MPI_COMM_WORLD, argv[1], MPI_MODE_RDONLY, info_mine, &fh): File does not exist, error stack: ADIOI_NFS_OPEN(37): File /mathworks/devel/sandbox/eellis/3p-mpich-4.2.3/3p/derived/glnxa64/mpich/src/mpi/romio/test/test_hintfile does not exist Abort(1) on node 0 (rank 0 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0 FAIL runtests (exit status: 1) The file test_hintfile did exist before running the tests. Cheers, Edric. From: Zhou, Hui <zhouh(a)anl.gov> Sent: 13 December 2024 16:04 To: discuss(a)mpich.org Cc: Edric Ellis <eellis(a)mathworks.com> Subject: Re: Problems with "make check" in romio building MPICH-4.2.3 Hi Edric, Could you show the relevant console log for the failure? Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Friday, December 13, 2024 2:59 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Edric Ellis <eellis(a)mathworks.com<mailto:[email protected]>> Subject: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Hi mpich-discuss, "make check" is failing with mpich-4. 2. 3, in the src/mpi/romio/test subdirectory. I think it's because this file: https: //urldefense. us/v3/__https: //github. com/pmodels/mpich/blob/4. 2. x/src/mpi/romio/test/runtests. in__;!!G_uCfscf7eWS!c86jBfO2_nn7840CXToEK-Dl8-y3_afw3ARA6rXELHbuMyXHtcqIGwBNDcw6oUUN8brlyugyCb5NBiY$ ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hi mpich-discuss, "make check" is failing with mpich-4.2.3, in the src/mpi/romio/test subdirectory. I think it's because this file: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/…> still says "FILENAME=test" on line 25. (The test fails because "FILENAME=test" eventually causes "test_hintfile" to be delete before it's needed by one of the tests). There's a change over here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061…> to make it say "FILENAME=", which I think would fix things. I'm not sufficiently familiar with github to fathom how that change didn't make it in to 4.2.3 when this change to the Makefile https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a…<https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a…> (which effectively makes top-level "make check" enter the romio/test subdirectory) did make it in to 4.2.3... Cheers, Edric.

1 0

Re: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3
by Zhou, Hui 13 Dec '24

13 Dec '24

Hi Edric, Could you show the relevant console log for the failure? Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org> Sent: Friday, December 13, 2024 2:59 AM To: discuss(a)mpich.org <discuss(a)mpich.org> Cc: Edric Ellis <eellis(a)mathworks.com> Subject: [mpich-discuss] Problems with "make check" in romio building MPICH-4.2.3 Hi mpich-discuss, "make check" is failing with mpich-4. 2. 3, in the src/mpi/romio/test subdirectory. I think it's because this file: https: //urldefense. us/v3/__https: //github. com/pmodels/mpich/blob/4. 2. x/src/mpi/romio/test/runtests. in__;!!G_uCfscf7eWS!c86jBfO2_nn7840CXToEK-Dl8-y3_afw3ARA6rXELHbuMyXHtcqIGwBNDcw6oUUN8brlyugyCb5NBiY$ ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hi mpich-discuss, "make check" is failing with mpich-4.2.3, in the src/mpi/romio/test subdirectory. I think it's because this file: https://urldefense.us/v3/__https://github.com/pmodels/mpich/blob/4.2.x/src/… still says "FILENAME=test" on line 25. (The test fails because "FILENAME=test" eventually causes "test_hintfile" to be delete before it's needed by one of the tests). There's a change over here: https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/90b0b061… to make it say "FILENAME=", which I think would fix things. I'm not sufficiently familiar with github to fathom how that change didn't make it in to 4.2.3 when this change to the Makefile https://urldefense.us/v3/__https://github.com/pmodels/mpich/commit/6e1e892a… (which effectively makes top-level "make check" enter the romio/test subdirectory) did make it in to 4.2.3... Cheers, Edric.

1 0

Problems with "make check" in romio building MPICH-4.2.3
by Edric Ellis 13 Dec '24

13 Dec '24

1 0

Join us for the MPICH BoF on Wednesday evening at SC24
by Raffenetti, Ken 18 Nov '24

18 Nov '24

All, The MPICH team will host a Birds-of-a-Feather session Wednesday evening as part of the SC24 proceedings. We hope you will join us to discuss the latest MPICH developments and hear from the community. We will again feature an MPICH user presentation alongside the traditional partner updates this year. See below for more info. Date and Time: Wednesday, November 20th, 5:15pm - 6:45pm EST Location: B308 Agenda: * Welcome and ANL Update * Welcome – Ken Raffenetti * ANL Update – Yanfei Guo & Hui Zhou * Partner Presentations * Maria Garzaran, Intel * Hari Subramoni, MVAPICH * Simon Pickartz, ParTex AG * Naveen Ravi, HPE * Jim Dinan, NVIDIA * Charles Shereda, Cornelis Networks * User Presentations * Jeff Hammond, NVIDIA * Q&A Cheers, The MPICH team

1 0