Hello Christoph<div><br></div><div>For make some tuning on drbd you can look this link</div><div><br></div><div><meta http-equiv="content-type" content="text/html; charset=utf-8"><a href="http://www.drbd.org/users-guide/s-latency-tuning.html">http://www.drbd.org/users-guide/s-latency-tuning.html</a></div>
<div><br><div class="gmail_quote">2012/5/20 Christoph Bartoschek <span dir="ltr"><<a href="mailto:ponto@pontohonk.de" target="_blank">ponto@pontohonk.de</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi,<br>
<br>
we have a two node setup with drbd below LVM and an Ext4 filesystem that is<br>
shared vi NFS. The system shows low performance and lots of timeouts<br>
resulting in unnecessary failovers from pacemaker.<br>
<br>
The connection between both nodes is capable of 1 GByte/s as shown by iperf.<br>
The network between the clients and the nodes is capable of 110 MByte/s. The<br>
RAID can be filled with 450 MByte/s.<br>
<br>
Thus I would expect to have a write performance of about 100 MByte/s. But dd<br>
gives me only 20 MByte/s.<br>
<br>
dd if=/dev/zero of=bigfile.10G bs=8192 count=1310720<br>
1310720+0 records in<br>
1310720+0 records out<br>
10737418240 bytes (11 GB) copied, 498.26 s, 21.5 MB/s<br>
<br>
While the slow dd runs there are timeouts on the server resulting in a<br>
restart of some resources. In the logfile I also see:<br>
<br>
[329014.592452] INFO: task nfsd:2252 blocked for more than 120 seconds.<br>
[329014.592820] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables<br>
this message.<br>
[329014.593273] nfsd D 0000000000000007 0 2252 2<br>
0x00000000<br>
[329014.593278] ffff88060a847c40 0000000000000046 ffff88060a847bf8<br>
0000000300000001<br>
[329014.593284] ffff88060a847fd8 ffff88060a847fd8 ffff88060a847fd8<br>
0000000000013780<br>
[329014.593290] ffff8806091416f0 ffff8806085bc4d0 ffff88060a847c50<br>
ffff88061870c800<br>
[329014.593295] Call Trace:<br>
[329014.593303] [<ffffffff8165a55f>] schedule+0x3f/0x60<br>
[329014.593309] [<ffffffff81265085>] jbd2_log_wait_commit+0xb5/0x130<br>
[329014.593315] [<ffffffff8108aec0>] ? add_wait_queue+0x60/0x60<br>
[329014.593321] [<ffffffff812111b8>] ext4_sync_file+0x208/0x2d0<br>
[329014.593328] [<ffffffff811a62dd>] vfs_fsync_range+0x1d/0x40<br>
[329014.593339] [<ffffffffa0227e51>] nfsd_commit+0xb1/0xd0 [nfsd]<br>
[329014.593349] [<ffffffffa022f28d>] nfsd3_proc_commit+0x9d/0x100 [nfsd]<br>
[329014.593356] [<ffffffffa0222a4b>] nfsd_dispatch+0xeb/0x230 [nfsd]<br>
[329014.593373] [<ffffffffa00e9d95>] svc_process_common+0x345/0x690<br>
[sunrpc]<br>
[329014.593379] [<ffffffff8105f990>] ? try_to_wake_up+0x200/0x200<br>
[329014.593391] [<ffffffffa00ea1e2>] svc_process+0x102/0x150 [sunrpc]<br>
[329014.593397] [<ffffffffa02221ad>] nfsd+0xbd/0x160 [nfsd]<br>
[329014.593403] [<ffffffffa02220f0>] ? nfsd_startup+0xf0/0xf0 [nfsd]<br>
[329014.593407] [<ffffffff8108a42c>] kthread+0x8c/0xa0<br>
[329014.593412] [<ffffffff81666bf4>] kernel_thread_helper+0x4/0x10<br>
[329014.593416] [<ffffffff8108a3a0>] ? flush_kthread_worker+0xa0/0xa0<br>
[329014.593420] [<ffffffff81666bf0>] ? gs_change+0x13/0x13<br>
<br>
<br>
Has anyone an idea what could cause such problems? I have no idea for<br>
further analysis.<br>
<br>
Is ext4 unsuitable for such a setup? Or is the linux nfs3 implementation<br>
broken? Are buffers too large such that one has too wait too long for a<br>
flush?<br>
<br>
Thanks<br>
Christoph Bartoschek<br>
<br>
<br>
<br>
_______________________________________________<br>
Pacemaker mailing list: <a href="mailto:Pacemaker@oss.clusterlabs.org">Pacemaker@oss.clusterlabs.org</a><br>
<a href="http://oss.clusterlabs.org/mailman/listinfo/pacemaker" target="_blank">http://oss.clusterlabs.org/mailman/listinfo/pacemaker</a><br>
<br>
Project Home: <a href="http://www.clusterlabs.org" target="_blank">http://www.clusterlabs.org</a><br>
Getting started: <a href="http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf" target="_blank">http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf</a><br>
Bugs: <a href="http://bugs.clusterlabs.org" target="_blank">http://bugs.clusterlabs.org</a><br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br>esta es mi vida e me la vivo hasta que dios quiera<br>
</div>