You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by Marc Boorshtein <mb...@gmail.com> on 2024/09/13 20:33:46 UTC

Exceeded connections on Kubernetes 1.30

Has anyone else run into an issue where ActiveMQ 5.18.4 starts
creating "“Exceeded
the maximum number of allowed client connections” errors one Kubernetes
1.30?  Same code in 1.29 I've never seen this issue.  Looking at the
connections list, I only see 5-6.  I've seen this in AKS and an onprem
cluster.  I increased the max connecitons to MAX_INTEGER but that doesn't
seem like a fix.  The application clients all maintain persistent
connections.

Thanks
Marc

Re: Exceeded connections on Kubernetes 1.30

Posted by Ken Liao <ke...@gmail.com>.
Hi Marc,

What other errors are in activemq.log?
Are there logs that look like " WARN | Could not accept connection from
tcp://<ip address>: Broken pipe (Broken pipe) |
org.apache.activemq.broker.TransportConnector"
or
"Transport Connection to: tcp://<ip address> failed: Channel was inactive
(no connection attempt made) for too (>30000) long"

Thanks,
Ken

On Fri, Sep 13, 2024 at 1:34 PM Marc Boorshtein <mb...@gmail.com>
wrote:

> Has anyone else run into an issue where ActiveMQ 5.18.4 starts
> creating "“Exceeded
> the maximum number of allowed client connections” errors one Kubernetes
> 1.30?  Same code in 1.29 I've never seen this issue.  Looking at the
> connections list, I only see 5-6.  I've seen this in AKS and an onprem
> cluster.  I increased the max connecitons to MAX_INTEGER but that doesn't
> seem like a fix.  The application clients all maintain persistent
> connections.
>
> Thanks
> Marc
>