http://defect.opensolaris.org/bz/show_bug.cgi?id=10090
--- Comment #2 from John Leser <John.Leser at sun.com> 2009-08-06 13:03:08 --- (In reply to comment #1) > (In reply to comment #0) > > Created an attachment (id=2261) --> (http://defect.opensolaris.org/bz/attachment.cgi?id=2261) [details] [details] > > Core file from nwamd > > > > This is with build nwam1-build:2009-07-13, running event tests that create > > several ENMs, then cause them to be enabled/disabled based on ip-address > > activation conditions. > > > > bash-3.2# mdb /var/cores/core.nwamd.208518 > > Loading modules: [ libumem.so.1 libuutil.so.1 libnvpair.so.1 > > libsysevent.so.1 > > ld.so.1 ] > > > $c > > libnvpair.so.1`nvlist_lookup_nvpair_ei_sep+0x71(deadbeef, 806b2e8, 0, > > 8047d38, > > 0 > > , 0) > > libnvpair.so.1`nvlist_lookup_nvpair+0x1a(deadbeef, 806b2e8, 8047d38, > > fead3ee5) > > libnwam.so.1`nwam_get_prop_value+0x4b(deadbeef, 806b2e8, 8047d90, feac8034, > > fee12a00) > > libnwam.so.1`nwam_enm_get_prop_value+0x2a(80e3648, 806b2e8, 8047d90, > > 80c47a4) > > nwamd_enm_check+0x2a(80c4708) > > nwamd_walk_objects+0x72(3, 805c5a4, 0, 805da38, 8082800, 80f679c) > > nwamd_enm_check_conditions+0x14(80f6718, 80fa708, 8047e48, 805dba4, 1, > > 809db14) > > nwamd_triggered_check_all_conditions+0xb(1, 809db14, 8047e68, 80604bc, 29, > > fef8c030) > > nwamd_event_handler+0x84(29, fef8c030, 1, 0, 8047e50, feffb804) > > main+0x1c8(1, 8047e94, 8047e9c, 80597af) > > _start+0x7d(1, 8047f24, 0, 8047f34, 8047f4c, 8047f70) > > > ::status > > debugging core file of nwamd (32-bit) from unknown > > file: /lib/inet/nwamd > > initial argv: /lib/inet/nwamd > > threading model: native threads > > status: process terminated by SIGSEGV (Segmentation Fault), addr=deadbef3 > > hmm, the ENM nvlist appears to have gone away during the object walk to check > conditions. Do your tests remove ENMs they create also? I'm wondering if this > may be caused by a locking issue that is exposed by a destroy happening around > the time of a triggered condition check. If this is reproducible, an nwamd log > would help too. Each test case starts by destroying all existing user state (nwamcfg destroy -a), so yes, the enms are being destroyed by the test suite. That said, I recently redid the startup sequence because it stopped being possible to destroy the User NCP while it was active and nwam was disabled. So, as it stands, I'm not having any luck reproducing this issue at this time. It may have been corrected or the test suite changes might have prevented it from being triggered. -- Configure bugmail: http://defect.opensolaris.org/bz/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the QA contact for the bug.
