Hi Ravindra,
Thanks for your suggestion.
Hi all,
Seems like the following is the root cause for this issue:
When an endpoint is unavailable, we noticed that,
whatever is given after the send mediator (in the insequence) doesn't get
executed.
As a result, the BAM mediator, which is placed after
Hi Lakshman,
AFAIK this looks like a configuration issue. You have given a name to
endpoints which were defined inline. Can you please change the
configuration as given in article [1] and simplify this. Then try that out
again.
[1]
Hi All,
As per the offline discussion with Lakmali, The reason for this issue is
when one end point is down, It will fail to go to the bam mediator(which
is responsible for publishing event to destination statistics table in BAM
cassandra database)which is added after send mediator in synapse
Hi Lakshman,
First you should make sure all the required statistics are published to BAM
with the data publisher. In order to do it you can,
1. Add a debug log to the data publisher publishing method to log all the
messages passed through it.
2. Then you can use Cassandra Explorer or another tool
Hi Gihan,
I sent 1 requests with the preconditions you have mentioned. only one
record is missed in cassendra database for every time I sent 1 requests
via APIM.
Thanks
On Wed, Oct 22, 2014 at 9:28 AM, Lakshman Udayakantha lakshm...@wso2.com
wrote:
Hi Gihan,
I will do that test and
Hi Lakshman,
That's is a good observation. So I believe this event should be either
first one or the last one. Can you please do a another small test?
Shutdown the BAM server.
Delete the Cassandra data folder completely (You can find that in
repository/databases/cassandra)
Truncate the APIM
Hi Gihan,
I will try that way. thank you so much for the support you have given so far
Thanks
On Wed, Oct 22, 2014 at 6:49 PM, Gihan Anuruddha gi...@wso2.com wrote:
Hi Lakshman,
That's is a good observation. So I believe this event should be either
first one or the last one. Can you please
Hi Lakshman,
This is very useful results, Thanks for sharing. We will do the same test
for ES as well. Is this test run for single BAM instance, or multiple?
(load balanced environment)
Thanks
On Wed, Oct 22, 2014 at 6:37 PM, Lakshman Udayakantha lakshm...@wso2.com
wrote:
Hi Gihan,
I sent
Hi Manoj,
I tested this with single BAM instance. As Gihan mentioned ,Still need to
check there is a problem in APIM side. I will update this thread if problem
is in APIM
Thanks
On Wed, Oct 22, 2014 at 7:49 PM, Manoj Gunawardena man...@wso2.com wrote:
Hi Lakshman,
This is very useful
Hi All,
I am working on https://wso2.org/jira/browse/APIMANAGER-2064.
By looking at the Cassandra database on BAM, I could figure out some
requests publish to Cassandra database and some do not when I invoke the
api. That is the reason for the missing requests in statistics dashboard.
I tried
Hi Lakshman,
It should be in Cassandra if publishEvent method executes. How do you
validate the data loss in Cassandra side? BAM embedded Cassandra explorer
is a not good tool for that. Can you please verify that with tool like CQL
or cassandra-cli?
Regards,
Gihan
On Tue, Oct 21, 2014 at 9:18
Hi,
This can be an issue for other products as well. ES, AS and EMM also
publishing data to BAM. In ES we are using load balancing data publisher. I
also notice if we call stop method, sometimes data not get publish to BAM.
We changed this to initiate data publisher at the first event and call
Hi Gihan,
These data in cassandra database are summarised by a hive script and insert
into a relational database. I checked that relational database table also.
Number of records in cassandra database and relational database table are
same. Then it should not be a problem with cassandra explorer.
Hi Lakshman,
How many records did you send to BAM and how many events are lost?
Regards,
Gihan
On Wed, Oct 22, 2014 at 8:18 AM, Lakshman Udayakantha lakshm...@wso2.com
wrote:
Hi Gihan,
These data in cassandra database are summarised by a hive script and
insert into a relational database. I
HI Gihan ,
They are lost randomly.
Thanks
On Wed, Oct 22, 2014 at 9:00 AM, Gihan Anuruddha gi...@wso2.com wrote:
Hi Lakshman,
How many records did you send to BAM and how many events are lost?
Regards,
Gihan
On Wed, Oct 22, 2014 at 8:18 AM, Lakshman Udayakantha lakshm...@wso2.com
Hi Lakshman,
You mean data is therw in cassandra and cassandra explorer doesn't show
these data? Did you compare by using cassandra CLI tool?
On Wednesday, October 22, 2014, Lakshman Udayakantha lakshm...@wso2.com
wrote:
HI Gihan ,
They are lost randomly.
Thanks
On Wed, Oct 22, 2014 at
Hi Inosh,
data in cassandra database are summized to a relational database. records
in cassandra database and this relational table is same. so I think there
is no problem with cassandra explorer.
Thanks
On Wed, Oct 22, 2014 at 9:14 AM, Inosh Goonewardena in...@wso2.com wrote:
Hi Lakshman,
Hi,
On Wednesday, October 22, 2014, Lakshman Udayakantha lakshm...@wso2.com
wrote:
Hi Inosh,
data in cassandra database are summized to a relational database. records
in cassandra database and this relational table is same. so I think there
is no problem with cassandra explorer.
Ok
Hi Lakshman,
Can you please do this test. I believe currently you are using BAM 2.4.1.
Please you this 2.5.0 RC [1]. Clean your RDBMS as well. Then using a tool
like JMeter send 10,000 requests to APIM and see you are getting 10,000
events in BAM. Please run these tests without any debugging
Hi Gihan,
I will do that test and get back to you.
Thanks
On Wed, Oct 22, 2014 at 9:23 AM, Inosh Goonewardena in...@wso2.com wrote:
Hi,
On Wednesday, October 22, 2014, Lakshman Udayakantha lakshm...@wso2.com
wrote:
Hi Inosh,
data in cassandra database are summized to a relational
20 matches
Mail list logo
