This bug is being fixed in the repository. You can disable flush-behind on the server spec file to not trigger the bug, till the next release is made available.
Avati On Thu, Apr 2, 2009 at 2:41 PM, Greg <[email protected]> wrote: > Shehjar Tikoo a écrit : >> >> Hi Greg >> >> Can you try running a test with the same config but without the >> io-threads translator? It has undergone some changes lately that might >> be the cause here. However, cursory look at the code suggests something >> else. Still, no harm in eliminating one potential reason. > > Hi Shehjar, > > I've tried, same crash on both servers : > > pending frames: > > patchset: 4e5c297d7c3480d0d3ab1c0c2a184c6a4fb801ef > signal received: 11 > configuration details:argp 1 > backtrace 1 > bdb->cursor->get 1 > db.h 1 > dlfcn 1 > fdatasync 1 > libpthread 1 > llistxattr 1 > setfsid 1 > spinlock 1 > epoll.h 1 > xattr.h 1 > st_atim.tv_nsec 1 > package-string: glusterfs 2.0.0rc7 > /lib/libc.so.6[0x7f5dd9794f60] > /usr/lib/glusterfs/2.0.0rc7/xlator/features/posix-locks.so(pl_inode_get+0x7a)[0x7f5dd934ac8a] > /usr/lib/glusterfs/2.0.0rc7/xlator/features/posix-locks.so(pl_flush+0x29)[0x7f5dd934c339] > /usr/lib/libglusterfs.so.0(default_flush+0xaa)[0x7f5dd9eebeba] > /usr/lib/glusterfs/2.0.0rc7/xlator/performance/write-behind.so(wb_flush+0x268)[0x7f5dd8f3b088] > /usr/lib/glusterfs/2.0.0rc7/xlator/performance/read-ahead.so(ra_flush+0xe0)[0x7f5dd8d30b30] > /usr/lib/glusterfs/2.0.0rc7/xlator/protocol/server.so(server_release+0xf9)[0x7f5dd8b1ae79] > /usr/lib/glusterfs/2.0.0rc7/xlator/protocol/server.so(protocol_server_pollin+0xa6)[0x7f5dd8b157d6] > /usr/lib/glusterfs/2.0.0rc7/xlator/protocol/server.so(notify+0x38)[0x7f5dd8b15818] > /usr/lib/glusterfs/2.0.0rc7/transport/socket.so(socket_event_handler+0xe0)[0x7f5dd8908b80] > /usr/lib/libglusterfs.so.0[0x7f5dd9efe1ef] > /usr/sbin/glusterfsd(main+0xa81)[0x403a21] > /lib/libc.so.6(__libc_start_main+0xe6)[0x7f5dd97811a6] > /usr/sbin/glusterfsd[0x402519] > --------- > ================================================================================ > Version : glusterfs 2.0.0rc7 built on Apr 1 2009 15:21:00 > TLA Revision : 4e5c297d7c3480d0d3ab1c0c2a184c6a4fb801ef > Starting Time: 2009-04-02 11:10:03 > Command line : /usr/sbin/glusterfsd -p /var/run/glusterfsd.pid -f > /etc/glusterfs/glusterfsd.vol > PID : 16858 > System name : Linux > Nodename : filer-04 > Kernel Release : 2.6.26-1-amd64 > Hardware Identifier: x86_64 > > Given volfile: > +------------------------------------------------------------------------------+ > 1: # file: /etc/glusterfs/glusterfsd.vol > 2: > 3: # > 4: # Volumes > 5: # > 6: volume media-small > 7: type storage/posix > 8: option directory /var/local/glusterfs/media_small > 9: end-volume > 10: > 11: volume media-medium > 12: type storage/posix > 13: option directory /var/local/glusterfs/media_medium > 14: end-volume > 15: > 16: # Lock posix > 17: volume media-small-locks > 18: type features/posix-locks > 19: option mandatory-locks on > 20: subvolumes media-small > 21: # subvolumes trash # enable this if you need trash can support (NOTE: > not present in 1.3.0-pre5+ releases) > 22: end-volume > 23: > 24: volume media-medium-locks > 25: type features/posix-locks > 26: option mandatory-locks on > 27: subvolumes media-medium > 28: # subvolumes trash # enable this if you need trash can support (NOTE: > not present in 1.3.0-pre5+ releases) > 29: end-volume > 30: > 31: > 32: # > 33: # Performance > 34: # > 35: #volume media-small-iot > 36: # type performance/io-threads > 37: # subvolumes media-small-locks > 38: # option thread-count 4 # default value is 1 > 39: #end-volume > 40: > 41: volume media-small-ioc > 42: type performance/io-cache > 43: option cache-size 128MB # default is 32MB > 44: option page-size 128KB # default is 128KB > 45: # subvolumes media-small-iot > 46: subvolumes media-small-locks > 47: end-volume > 48: > 49: volume media-small-wb > 50: type performance/write-behind > 51: option flush-behind on # default is off > 52: subvolumes media-small-ioc > 53: end-volume > 54: > 55: volume media-small-ra > 56: type performance/read-ahead > 57: subvolumes media-small-wb > 58: option page-size 256KB # default is 256KB > 59: option page-count 4 # default is 2 - cache per file = > (page-count x page-size) > 60: option force-atime-update no # defalut is 'no' > 61: end-volume > 62: > 63: > 64: #volume media-medium-iot > 65: # type performance/io-threads > 66: # subvolumes media-medium-locks > 67: # option thread-count 4 # default value is 1 > 68: #end-volume > 69: > 70: volume media-medium-ioc > 71: type performance/io-cache > 72: option cache-size 128MB # default is 32MB > 73: option page-size 128KB # default is 128KB > 74: # subvolumes media-medium-iot > 75: subvolumes media-medium-locks > 76: end-volume > 77: > 78: volume media-medium-wb > 79: type performance/write-behind > 80: option flush-behind on # default is off > 81: subvolumes media-medium-ioc > 82: end-volume > 83: > 84: volume media-medium-ra > 85: type performance/read-ahead > 86: subvolumes media-medium-wb > 87: option page-size 256KB # default is 256KB > 88: option page-count 4 # default is 2 - cache per file = > (page-count x page-size) > 89: option force-atime-update no # defalut is 'no' > 90: end-volume > 91: > 92: > 93: > 94: > 95: # > 96: # Serveur > 97: # > 98: volume server > 99: type protocol/server > 100: option transport-type tcp/server > 101: option auth.addr.media-small-ra.allow 10.0.*.* > 102: option auth.addr.media-medium-ra.allow 10.0.*.* > 103: # Autoconfiguration, e.g. : > 104: # glusterfs -l /tmp/glusterfs.log --server=filer-04 ./Cache > 105: option client-volume-filename /etc/glusterfs/glusterfs.vol > 106: subvolumes media-small-ra media-medium-ra # volumes exportés > 107: end-volume > 108: > > +------------------------------------------------------------------------------+ > 2009-04-02 11:10:03 N [glusterfsd.c:1134:main] glusterfs: Successfully > started > > -- > Greg > > > _______________________________________________ > Gluster-users mailing list > [email protected] > http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users > _______________________________________________ Gluster-users mailing list [email protected] http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users
