[Pacemaker] how does the exportfs resource agent work?

Andrew Martin amartin at xes-inc.com
Mon Jan 16 15:53:43 CET 2012


Hi Dennis, 


Have you also added /var/lib/nfs to the shared DRBD resource? This is an important step to ensure that data about currently-open files and mount information is transferred to the other node during failover. See the end of Step 4: 
http://www.howtoforge.com/highly-available-nfs-server-using-drbd-and-heartbeat-on-debian-5.0-lenny 


Thanks, 


Andrew 

----- Original Message -----

From: "emmanuel segura" <emi2fast at gmail.com> 
To: "The Pacemaker cluster resource manager" <pacemaker at oss.clusterlabs.org> 
Sent: Monday, January 16, 2012 6:06:54 AM 
Subject: Re: [Pacemaker] how does the exportfs resource agent work? 

you should check how you mount the nfs cluster share from your client 

for example 

mount -o hard,rw,rsize=32768,wsize=32768,suid,proto=tcp,vers=3 -t nfs your_virtual_ip:/your_cluster_fs_share /mountpoint 

man nfs 

===================================================== 
timeo=n The value in tenths of a second before sending the first retransmission after an RPC timeout. The 
default value is 7 tenths of a second. After the first timeout, the timeout is doubled after each suc- 
cessive timeout until a maximum timeout of 60 seconds is reached or the enough retransmissions have 
occured to cause a major timeout. Then, if the filesystem is hard mounted, each new timeout cascade 
restarts at twice the initial value of the previous cascade, again doubling at each retransmission. 
The maximum timeout is always 60 seconds. Better overall performance may be achieved by increasing the 
timeout when mounting on a busy network, to a slow server, or through several routers or gateways. 

retrans=n The number of minor timeouts and retransmissions that must occur before a major timeout occurs. The 
default is 3 timeouts. When a major timeout occurs, the file operation is either aborted or a "server 
not responding" message is printed on the console. 
======================================================= 


2012/1/16 Dennis Jacobfeuerborn < dennisml at conversis.de > 


What am I supposed to look for? 

Regards, 
Dennis 


On 01/16/2012 12:13 PM, emmanuel segura wrote: 

<blockquote>

I think man nfs it's can help you 

Try to look your nfs client options 

2012/1/15 Dennis Jacobfeuerborn < dennisml at conversis.de 
<mailto: dennisml at conversis.de > > 



Hi, 
I'm trying to build a HA nfs system based on drbd and apart from the 
nfs export everything is working fine. The problem is that when I force 
a failover things seem to work fine yet when I fail back to the 
original system the clients freeze for a very long time. 

/mnt/tmp is the mountpoint on the client and I'm using the following to 
test access: 
for i in `seq 1 2000`; do echo $i; ls /mnt/tmp; sleep 1; done 

on a failover the output look like this: 

... 
47 
testfile testfile2 
48 
testfile testfile2 
49 
testfile testfile2 
50 
testfile testfile2 
51 
testfile testfile2 
52 
ls: cannot open directory /mnt/tmp: Permission denied 
53 
ls: cannot open directory /mnt/tmp: Permission denied 
54 
<<< freeze of several minutes >>> 
testfile testfile2 
55 
testfile testfile2 
56 
testfile testfile2 
... 

The first question I have is how can I prevent the "Permission denied" 
errors? If these occur on e.g. a mountpoint for MySQL for example this 
will no doubt lead to problems with the database and that means the 
storage isn't really redundant. 

The second question is how do I reduce the failover time? I tried 
adding timeo=30 to the client mount options but that doesn't seem to help. 

This is what my cib looks like: 

node storage1.dev 
node storage2.dev 
primitive p_drbd_nfs ocf:linbit:drbd \ 
params drbd_resource="nfs" \ 
op monitor interval="15" role="Master" \ 
op monitor interval="30" role="Slave" 
primitive p_exportfs_data ocf:heartbeat:exportfs \ 
params fsid="1" directory="/mnt/data/export" 
options="rw,no_root_squash" clientspec="*" \ 
op monitor interval="30s" 
primitive p_fs_data ocf:heartbeat:Filesystem \ 
params device="/dev/drbd/by-res/nfs" directory="/mnt/data" 
fstype="ext3" \ 
op monitor interval="10s" 
primitive p_ip_nfs ocf:heartbeat:IPaddr2 \ 
params ip="192.168.2.190" cidr_netmask="24" \ 
op monitor interval="30s" 
group g_nfs p_fs_data p_exportfs_data p_ip_nfs 
ms ms_drbd_nfs p_drbd_nfs \ 
meta master-max="1" master-node-max="1" clone-max="2" 
clone-node-max="1" notify="true" 
colocation c_nfs_on_drbd inf: g_nfs ms_drbd_nfs:Master 
order o_drbd_before_nfs inf: ms_drbd_nfs:promote g_nfs:start 
property $id="cib-bootstrap-options" \ 
dc-version="1.0.12-unknown" \ 
cluster-infrastructure="__ openais" \ 

expected-quorum-votes="2" \ 
stonith-enabled="false" \ 
no-quorum-policy="ignore" 
rsc_defaults $id="rsc-options" \ 
resource-stickiness="200" 

Regards, 
Dennis 

______________________________ ___________________ 
Pacemaker mailing list: Pacemaker at oss.clusterlabs.org 
<mailto: Pacemaker at oss. clusterlabs.org > 
http://oss.clusterlabs.org/__ mailman/listinfo/pacemaker 

< http://oss.clusterlabs.org/ mailman/listinfo/pacemaker > 

Project Home: http://www.clusterlabs.org 
Getting started: 
http://www.clusterlabs.org/__ doc/Cluster_from_Scratch.pdf 

< http://www.clusterlabs.org/ doc/Cluster_from_Scratch.pdf > 
Bugs: http://bugs.clusterlabs.org 




-- 
esta es mi vida e me la vivo hasta que dios quiera 


______________________________ _________________ 
Pacemaker mailing list: Pacemaker at oss.clusterlabs.org 
http://oss.clusterlabs.org/ mailman/listinfo/pacemaker 

Project Home: http://www.clusterlabs.org 
Getting started: http://www.clusterlabs.org/ doc/Cluster_from_Scratch.pdf 
Bugs: http://bugs.clusterlabs.org 






______________________________ _________________ 
Pacemaker mailing list: Pacemaker at oss.clusterlabs.org 
http://oss.clusterlabs.org/ mailman/listinfo/pacemaker 

Project Home: http://www.clusterlabs.org 
Getting started: http://www.clusterlabs.org/ doc/Cluster_from_Scratch.pdf 
Bugs: http://bugs.clusterlabs.org 

</blockquote>



-- 
esta es mi vida e me la vivo hasta que dios quiera 

_______________________________________________ 
Pacemaker mailing list: Pacemaker at oss.clusterlabs.org 
http://oss.clusterlabs.org/mailman/listinfo/pacemaker 

Project Home: http://www.clusterlabs.org 
Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf 
Bugs: http://bugs.clusterlabs.org 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oss.clusterlabs.org/pipermail/pacemaker/attachments/20120116/b39b89ca/attachment-0001.html>


More information about the Pacemaker mailing list