[
https://issues.apache.org/jira/browse/PROTON-1899?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jeremy updated PROTON-1899:
---------------------------
Description:
In Qpid JMS, the default value of
[failover.maxReconnectAttempts|https://qpid.apache.org/releases/qpid-jms-0.11.1/docs/index.html#failover-configuration-options]
is -1, meaning it will retry indefinitely.
In Qpid Proton, the default is also to retry indefinitely, however, the default
value of
[reconnect_options.max_attempts|https://qpid.apache.org/releases/qpid-proton-0.22.0/proton/cpp/api/classproton_1_1reconnect__options.html#a7e6b1a75735da02638374a317a232428]
is 0.
I understand that conforming to JMS is perhaps not priority.
However, for me, the problem arises when I want to have failover without
necessarily retrying.
With 0 being the no limit to retry and being the default value, there is no way
(I can forsee) I can tell proton not to retry, but to failover. What I would
expect with failover and no retry is the following behavior:
* Proton tries to connect to the connect URL
* If it fails, it loops over the failover url's once, until it either succeeds
to connect or it fails
With the current way things are done, I am obliged to retry at least once,
repeating the above scenario twice. I tested this by putting no failover URLs,
and specifying a max_attempts of 1. Proton tries to connect twice to the
connect URL.
Perhaps there is a way around this with the current API, however, I couldn't
find it.
was:
In Qpid JMS, the default value of
[failover.maxReconnectAttempts|https://qpid.apache.org/releases/qpid-jms-0.11.1/docs/index.html#failover-configuration-options]
is -1, meaning it will retry indefinitely.
In Qpid Proton, the default is also to retry indefinitely, however, the default
value of
[reconnect_options.max_attempts|https://qpid.apache.org/releases/qpid-proton-0.22.0/proton/cpp/api/classproton_1_1reconnect__options.html#a7e6b1a75735da02638374a317a232428]
is 0.
I understand that conforming to JMS is perhaps not priority.
However, for me, the problem arises when I want to have failover without
necessarily retrying.
With 0 being the no limit to retry and being the default value, there is no way
(I can forsee) I can tell proton not to retry, but to failover. What I would
expect with failover and no retry is the following behavior:
* Proton tries to connect to the connect URL
* If it fails, it loops over the failover url's once, until it either succeeds
to connect or it fails
With the current way things are done, I am obliged to retry at least once,
repeating the above scenario twice. I tested this by putting no failover URLs,
and specifying a max_attempts of 1. Proton tries to connect twice to the
connect URL.
Perhaps there is a way around this with the current API, however, couldn't find
it.
> Qpid proton reconnect_options.max_attempts is not inline with Qpid JMS
> failover.maxReconnectAttempts
> ----------------------------------------------------------------------------------------------------
>
> Key: PROTON-1899
> URL: https://issues.apache.org/jira/browse/PROTON-1899
> Project: Qpid Proton
> Issue Type: Improvement
> Components: cpp-binding
> Affects Versions: proton-c-0.22.0
> Reporter: Jeremy
> Priority: Major
>
> In Qpid JMS, the default value of
> [failover.maxReconnectAttempts|https://qpid.apache.org/releases/qpid-jms-0.11.1/docs/index.html#failover-configuration-options]
> is -1, meaning it will retry indefinitely.
> In Qpid Proton, the default is also to retry indefinitely, however, the
> default value of
> [reconnect_options.max_attempts|https://qpid.apache.org/releases/qpid-proton-0.22.0/proton/cpp/api/classproton_1_1reconnect__options.html#a7e6b1a75735da02638374a317a232428]
> is 0.
> I understand that conforming to JMS is perhaps not priority.
> However, for me, the problem arises when I want to have failover without
> necessarily retrying.
> With 0 being the no limit to retry and being the default value, there is no
> way (I can forsee) I can tell proton not to retry, but to failover. What I
> would expect with failover and no retry is the following behavior:
> * Proton tries to connect to the connect URL
> * If it fails, it loops over the failover url's once, until it either
> succeeds to connect or it fails
> With the current way things are done, I am obliged to retry at least once,
> repeating the above scenario twice. I tested this by putting no failover
> URLs, and specifying a max_attempts of 1. Proton tries to connect twice to
> the connect URL.
> Perhaps there is a way around this with the current API, however, I couldn't
> find it.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
