Cannot run CST on Blue Waters

Create issue
Issue #2006 closed
Erik Schnetter created an issue

After upgrading to the Piraha parser, building on Blue Waters fails with this error:

child error: tags at /mnt/a/u/sciteam/eschnett/Cvanilla/lib/sbin/ScheduleParser.pl line 281
        main::parse_schedule_statement('Group=HASH(0x3ce23e0)', 'HASH(0xd1efd0)', 'SCALAR(0xd1c478)', 'SCALAR(0xd1c2f8)', 'SCALAR(0xd1c778)', 'ML_BSSN_CL') called at /mnt/a/u/sciteam/eschnett/Cvanilla/lib/sbin/ScheduleParser.pl line 334
        main::parse_schedule_statement('Group=HASH(0x7921800)', 'HASH(0xd1efd0)', 'SCALAR(0xd1c478)', 'SCALAR(0xd1c2f8)', 'SCALAR(0xd1c778)', 'ML_BSSN_CL') called at /mnt/a/u/sciteam/eschnett/Cvanilla/lib/sbin/ScheduleParser.pl line 304
        main::parse_schedule_statement('Group=HASH(0x3a41e58)', 'HASH(0xd1efd0)', 'SCALAR(0xd1c478)', 'SCALAR(0xd1c2f8)', 'SCALAR(0xd1c778)', 'ML_BSSN_CL') called at /mnt/a/u/sciteam/eschnett/Cvanilla/lib/sbin/ScheduleParser.pl line 382
        main::parse_schedule_ccl('ML_BSSN_CL', 'Group=HASH(0x3a41e58)', 'STORAGE: ML_log_confac[timelevels]', 'STORAGE: ML_metric[timelevels]', 'STORAGE: ML_Gamma[timelevels]', 'STORAGE: ML_trace_curv[timelevels]', 'STORAGE: ML_curv[timelevels]', 'STORAGE: ML_Theta[timelevels]', 'STORAGE: ML_lapse[timelevels]', ...) called at /mnt/a/u/sciteam/eschnett/Cvanilla/lib/sbin/ScheduleParser.pl line 111
        main::create_schedule_database('TestStrings', '/mnt/a/u/sciteam/eschnett/Cvanilla/arrangements/CactusTest/Te...', 'CartesianCoordinates', '/mnt/a/u/sciteam/eschnett/Cvanilla/arrangements/CTGamma/Carte...', 'WaveToyC', '/mnt/a/u/sciteam/eschnett/Cvanilla/arrangements/CactusWave/Wa...', 'ML_BSSN_DG3', '/mnt/a/u/sciteam/eschnett/Cvanilla/arrangements/McLachlan_dgt...', 'ML_Test_FD4', ...) called at /mnt/a/u/sciteam/eschnett/Cvanilla/lib/sbin/CST line 144

Things work fine on other systems. A first guess would be some kind of Perl incompatibility?

Keyword:

Comments (4)

  1. Steven R. Brandt
    • removed comment

    No, it's not a perl version problem. The issue is that the parser isn't handling the TAGS attribute in ML_BSSN_CL. I've replicated the issue.

  2. Log in to comment