[Gluster-devel] Crashing glusterfs server / sync not working

[Guido Smit]

Hi all,

I have on my 2 Centos5 machines fuse2.7.2gls8 and glusterfs tla .
Everything works fine, as long as I don't sync the machines.
One of the things I see al the time is that most files are not synced on 
both servers. When I try to force a sync using find /mail -type f -exec 
head -c 1 {} \; >/dev/null

I get the following crash after a few minutes:

2008-01-31 10:08:00 E [server-protocol.c:178:generic_reply] server: 
transport_writev failed
2008-01-31 10:08:00 D [inode.c:308:__destroy_inode] mail/inode: destroy 
inode(0) [@0xb7e23a60]

---------
got signal (11), printing backtrace
---------
[0x537420]
//lib/libglusterfs.so.0[0xd4390c]
//lib/libglusterfs.so.0[0xd4390c]
//lib/libglusterfs.so.0[0xd4390c]
//lib/glusterfs/1.3.8/xlator/cluster/unify.so(unify_opendir_cbk+0xa3)[0x2e07f3]
//lib/glusterfs/1.3.8/xlator/cluster/afr.so(afr_opendir_cbk+0x138)[0x91c9f8]
//lib/glusterfs/1.3.8/xlator/protocol/client.so[0x1125b8]
//lib/glusterfs/1.3.8/xlator/protocol/client.so(notify+0xa97)[0x116717]
//lib/libglusterfs.so.0(transport_notify+0x37)[0xd47aa7]
//lib/libglusterfs.so.0(sys_epoll_iteration+0xd7)[0xd487e7]
//lib/libglusterfs.so.0(poll_iteration+0x7c)[0xd47bdc]
[glusterfsd][0x8049432]
//lib/libc.so.6(__libc_start_main+0xdc)[0xbe0dec]
[glusterfsd][0x8048cf1]
---------

My glusterfs-server.vol:

volume pop1-mail-ns
        type protocol/client
        option transport-type tcp/client
        option remote-host 62.59.252.41
        option remote-subvolume pop1-mail-ns
        option transport-timeout 10
end-volume

volume pop1-mail-ds
        type protocol/client
        option transport-type tcp/client
        option remote-host 62.59.252.41
        option remote-subvolume pop1-mail-ds
        option transport-timeout 10
end-volume

volume pop2-mail-ns
        type storage/posix
        option directory /home/export/namespace
end-volume

volume pop2-mail-ds
        type storage/posix
        option directory /home/export/mailspool
end-volume

volume ns-afr
        type cluster/afr
        subvolumes pop1-mail-ns pop2-mail-ns
        option scheduler random
end-volume

volume ds-afr
        type cluster/afr
        subvolumes pop1-mail-ds pop2-mail-ds
        option scheduler random
end-volume

volume mail-unify
        type cluster/unify
        subvolumes ds-afr
        option namespace ns-afr
        option scheduler alu
        option alu.limits.max-open-files 10000   # Don't create files on 
a volume with more than 10000 files open
        option alu.order 
disk-usage:read-usage:write-usage:open-files-usage:disk-speed-usage
        option alu.disk-usage.entry-threshold 2GB   # Kick in if the 
discrepancy in disk-usage between volumes is more than 2GB
        option alu.disk-usage.exit-threshold  60MB   # Don't stop 
writing to the least-used volume until the discrepancy is 1988MB
        option alu.open-files-usage.entry-threshold 1024   # Kick in if 
the discrepancy in open files is 1024
        option alu.open-files-usage.exit-threshold 32   # Don't stop 
until 992 files have been written the least-used volume
        option alu.stat-refresh.interval 10sec   # Refresh the 
statistics used for decision-making every 10 seconds
end-volume

volume mail-iothreads
        type performance/io-threads
        option thread-count 8
        option cache-size 64MB
        subvolumes mail-unify
end-volume

volume mail-wb
        type performance/write-behind
        subvolumes mail-iothreads
end-volume

volume mail
        type performance/read-ahead
        subvolumes mail-wb
end-volume

volume server
        type protocol/server
        option transport-type tcp/server
        subvolumes mail
        option auth.ip.pop2-mail-ds.allow 62.59.252.*,127.0.0.1
        option auth.ip.pop2-mail-ns.allow 62.59.252.*,127.0.0.1
        option auth.ip.mail.allow 62.59.252.*,127.0.0.1
end-volume


My glusterfs-client.vol:

volume mailspool
        type protocol/client
        option transport-type tcp/client
        option remote-host 127.0.0.1
        option remote-subvolume mail
end-volume

volume writeback
        type performance/write-behind
        option aggregate-size 131072
        subvolumes mailspool
end-volume

volume readahead
        type performance/read-ahead
        option page-size 65536
        option page-count 16
        subvolumes writeback
end-volume

-- 
Regards,

Guido Smit
DevInet