From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3a) Gecko/20021212 Description of problem: I upgraded from 8.0 to 8.1b3. Running Zope with Python had the strange effect that after the upgrade Zope seems to hang after some HTTP request (nothing happens, no new connections were accepted). I straced Zope and it always hangs here: such file or directory) [pid 9547] brk(0) = 0x8c89000 [pid 9547] brk(0x8c95000) = 0x8c95000 [pid 9547] gettimeofday({1046349158, 697639}, NULL) = 0 [pid 9547] gettimeofday({1046349158, 698009}, NULL) = 0 [pid 9547] _llseek(5, 8192, [8192], SEEK_SET) = 0 [pid 9547] read(5, "\10\0\0\0\0\0\0\0Tq\16(h\7U\nDTMLMethodq\17ttQU"..., 4096) = 4096 [pid 9547] _llseek(5, 12288, [12288], SEEK_SET) = 0 [pid 9547] read(5, "/((U$Products.Sessions.SessionDa"..., 4096) = 4096 [pid 9547] _llseek(5, 69632, [69632], SEEK_SET) = 0 [pid 9547] read(5, "\0\0_getattr_s\10\0\0\0_getattrs\t\0\0\0con"..., 4096) = 4096 [pid 9547] gettimeofday({1046349158, 705371}, NULL) = 0 [pid 9547] write(10, "x", 1) = 1 [pid 9540] <... select resumed> ) = 1 (in [9], left {29, 990000}) [pid 9547] gettimeofday( <unfinished ...> [pid 9540] futex(0x8c87ea0, FUTEX_WAIT, 3, NULL <unfinished ...> [pid 9547] <... gettimeofday resumed> {1046349158, 706733}, NULL) = 0 [pid 9547] futex(0x8c87ea0, FUTEX_WAKE, 1, NULL) = 1 [pid 9547] write(10, "x", 1) = 1 [pid 9547] futex(0x8c881b0, FUTEX_WAIT, 0, NULL <unfinished ...> [pid 9544] futex(0x8c87c80, FUTEX_WAIT, 0, NULL <unfinished ...> [pid 9545] <... futex resumed> ) = 0 [pid 9545] futex(0x8c87b90, FUTEX_WAIT, 0, NULL <unfinished ...> No idea what's going one here. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Install Python 2.2.2 2. Install Zope 2.6.1 3. Additional info:
The same problems (hanging process in futex call) have been found in RH9. 1. Suddenly, any running of rpm hangs (fixed by `rm -f /var/lib/rpm/--db*`) 2. And while Oracle 9.2 installation, java machine hangs with this strace: ... gettimeofday({1063351776, 3307}, NULL) = 0 futex(0x82cd410, FUTEX_WAKE, 2147483647, {0, 0}) = 1 futex(0x82cce1c, FUTEX_WAKE, 1, {137155584, 1702259060}) = 1 futex(0x809dff4, FUTEX_WAKE, 1, NULL) = 1 write(9, "\n*** Install Page***", 20) = 20 clone(child_stack=0x5d139950, flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID|CLONE_DETACHED, [28496], {entry_number:6, base_addr:0x5d139d30, limit:1048575, seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0, useable:1}) = 28496 futex(0x8323468, FUTEX_WAKE, 2147483647, {0, 0}) = 1 futex(0x84ab9dc, FUTEX_WAKE, 1, {137507928, 0}) = 1 stat64("/tmp/OraInstall2003-09-12_11-28-10AM/oracle/sysman/oii/oiic/OiicInstallActionsPhase$OiInstRun.class", 0xbfffa938) = -1 ENOENT (No such file or directory) clone(child_stack=0x5e5a2950, flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID|CLONE_DETACHED, [28497], {entry_number:6, base_addr:0x5e5a2d30, limit:1048575, seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0, useable:1}) = 28497 futex(0x82cf348, FUTEX_WAKE, 2147483647, {0, 0}) = 1 futex(0x84a7364, FUTEX_WAKE, 1, {137163576, 1700147052}) = 1 futex(0x804e18c, FUTEX_WAKE, 1, {137163576, 1700147052}) = 1 futex(0x81a70d0, FUTEX_WAIT, 40, NULL
There have been many different issues wrt to using libpthread since RHL9. In libpthread or the programs using it. AFAIK the current code is fine. I'm closing the bug since the information is outdated in any case. Should there be a problem in current code, file new bugs.