activemq-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Christian Schneider <>
Subject Re: question for users of NFS master/slave setups
Date Tue, 01 Mar 2016 16:32:04 GMT
I have seen a similar scenario that I described in a mail on the user list

Do you have any idea why the failover is slower than a simple restart of 
the master?


On 28.02.2016 02:59, Sunil Vishwanath wrote:
> Hi Jim,
>    I am currently using in production two AMQ servers (ver 5.12) with NFSv4.  The failover
from one to the other happens seamlessly without any issues.  As Art stated, NFSv4 is a must
for failover to work properly.  I have the following recommendation to achieve a quick failover:
> .
> 1) It will be a good idea to have 64GB RAM memory in the AMQ server with lots of disk
space to handle high message volume.  If you have less memory than ideal and if there are
too many messages it can cause heavy swapping activity and thus a slowdown in performance.
 You should try to correct the situation
> 2) Make sure to have some kind of monitoring to keep an eye on KahaDB log file count.
 Ideally it should stay low somewhere in the range of 10 files. The count may climb in case
of sudden increase in incoming messages and slower consumption or no consumption.  It is a
good idea to pay attention to the file count and try to resolve the issue if any to reduce
chances of slowdown or AMQ crash.
>   If there are lots of log files (400 plus and each file size 32MB) in the KahaDB directory,
it will take some time to load all the log files after failover.  Once the files are loaded,
AMQ will be ready to process incoming and outgoing messages.
> Let me know if you have any questions.
> Sunil
>> On Feb 27, 2016, at 10:05 AM, James A. Robinson <> wrote:
>> Hi,
>> Thank you for the reply. Do you happen to know what the mount settings
>> actually are in your setup?
>> Yes, we are using nfs4 for this.
>> Jim
>> On Thu, Feb 25, 2016 at 20:21 artnaseef <> wrote:
>>> I've used it successfully more than once without any specific tuning to
>>> NFS.
>>> With that said, systems groups maintained the filesystem, so I may simply
>>> be
>>> unaware of the same.
>>> Note that you'll need NFSv4 for full H/A; NFSv3 clients hold locks
>>> indefinitely when they drop off the server's network (i.e. the server never
>>> times out the lock).
>>> --
>>> View this message in context:
>>> Sent from the ActiveMQ - User mailing list archive at

Christian Schneider

Open Source Architect

View raw message