Mailing-List: contact users-help@activemq.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@activemq.apache.org
Received-SPF: pass (nike.apache.org: local policy includes SPF record at
 spf.trusted-forwarder.org)
From: "Seibt, Volker" <volker.seibt@buenting.de>
To: "users@activemq.apache.org" <users@activemq.apache.org>
Subject: AW: KahaDB journals growing - no visible entries
Thread-Topic: KahaDB journals growing - no visible entries
Thread-Index: AQHO9llQj96CwHG5Gk6ueULeO93A5JpO3oQAgAAV3YA=
Date: Wed, 11 Dec 2013 13:01:59 +0000
Message-ID: <55C255F51A20E2469CA11960927CE18233EE3BF4@MEXMBX02.buenting.de>
References: <1386756190583-4675358.post@n4.nabble.com>
 <443BCE38E921434394B45FF0813FA5781ECFBF09@MSGEXSV21103.ent.wfb.bank.corp>
In-Reply-To: 
 <443BCE38E921434394B45FF0813FA5781ECFBF09@MSGEXSV21103.ent.wfb.bank.corp>
Accept-Language: de-DE, en-US
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="iso-8859-1"
Content-Language: de-DE

In addition to the queues used by the application there is one queue "Activ=
eMQ.DLQ" which holds 77 messages.

There is also one topic "ActiveMQ.Advisory.Queue" which holds 7 messages. T=
hat's all.

After purging "ActiveMQ.DLQ" and restarting the instance ActiveMQ brings up=
 23 messages in "ActiveMQ.DLQ".

Except viewing the admin web page and purging the queue there was no furthe=
r activity.

Regards,

Volker Seibt

-----Urspr=FCngliche Nachricht-----
Von: barry.barnett@wellsfargo.com [mailto:barry.barnett@wellsfargo.com]=20
Gesendet: Mittwoch, 11. Dezember 2013 13:39
An: users@activemq.apache.org
Betreff: RE: KahaDB journals growing - no visible entries

Is there a DLQ for each application queue?  Are those holding any messages?

Regards,

Barry Barnett
WMQ Enterprise Services & Solutions
Wells Fargo
Cell: 704-564-5501

-----Original Message-----
From: volker.seibt [mailto:volker.seibt@buenting.de]=20
Sent: Wednesday, December 11, 2013 5:03 AM
To: users@activemq.apache.org
Subject: KahaDB journals growing - no visible entries

We use ActiveMQ 5.7 as embedded jms provider with camel 2.11.0.

It's a small application which uses about seven or eight queues with about =
max. 1,000 small messages (tickets from a deposit machine) per queue / day.

We use persistent queues via KahaDB and did not change any KahaDB parameter=
s.

With this the number of KahaDB journals (db-xxx.log) is growing every day.
>From 2013/07 until now we got 439 files with the default size of 32mb, whic=
h cumulates to about 12 gb.

I copied this database to a standale ActiveMQ instance to analyze it:

On startup an automatic recovery is startet which "recovers" 5,400,000 (!) =
entries. All queues used by our application are shown with zero entries exc=
ept the ActiveMQ.DLQ (dead letter queue) with 77 entries.

There are also about four or three topics shown with maximum 7 entries.

While running this database on an ActiveMQ instance for 10 minutes it grows=
 by three new files (about 95 mb).

First questions: where do the 5.4 million of entries hide? How can I get ri=
d of them without without a complete deletion of the database? How can such=
 a state occure, and more important, how can we avoid this?


--
View this message in context: http://activemq.2283324.n4.nabble.com/KahaDB-=
journals-growing-no-visible-entries-tp4675358.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.