You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tomcat.apache.org by Mark Thomas <ma...@apache.org> on 2023/06/01 15:23:47 UTC
Re: PersistentValve removal ?
On 30/05/2023 22:44, Rémy Maucherat wrote:
> On Tue, May 30, 2023 at 6:46 PM Mark Thomas <ma...@apache.org> wrote:
<snip/>
>> I can't think of a use case that justifies keeping PersistentValve. BZ
>> 66513 talks about non-sticky LB with containers. I'll ask some follow-up
>> questions to see if there is something we are missing but my sense is
>> that this was always broken but the users using it either never observed
>> breakage or it happened rarely enough it wasn't followed up.
>
> +1
> It sounds reasonable to redocument stating the limitations better,
> deprecate, then remove in 11.
The OP for BZ 66513 has indicated that the Valve is useful for them when
running on containers in the cloud. I guess they must have a client that
follows the no more than one concurrent request rule. Given that they
explicitly asked us to keep the Valve, I'm happy to do so. I have a
patch I need to apply that adds some debug logging they also asked for.
I'll resolve the issue once that patch has been applied.
Mark
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tomcat.apache.org
For additional commands, e-mail: dev-help@tomcat.apache.org
Re: PersistentValve removal ?
Posted by Christopher Schultz <ch...@christopherschultz.net>.
Mark,
On 6/1/23 11:23, Mark Thomas wrote:
> On 30/05/2023 22:44, Rémy Maucherat wrote:
>> On Tue, May 30, 2023 at 6:46 PM Mark Thomas <ma...@apache.org> wrote:
>
> <snip/>
>
>>> I can't think of a use case that justifies keeping PersistentValve. BZ
>>> 66513 talks about non-sticky LB with containers. I'll ask some follow-up
>>> questions to see if there is something we are missing but my sense is
>>> that this was always broken but the users using it either never observed
>>> breakage or it happened rarely enough it wasn't followed up.
>>
>> +1
>> It sounds reasonable to redocument stating the limitations better,
>> deprecate, then remove in 11.
>
> The OP for BZ 66513 has indicated that the Valve is useful for them when
> running on containers in the cloud. I guess they must have a client that
> follows the no more than one concurrent request rule. Given that they
> explicitly asked us to keep the Valve, I'm happy to do so. I have a
> patch I need to apply that adds some debug logging they also asked for.
> I'll resolve the issue once that patch has been applied.
I will also re-work my patch for the Store such that it can avoid
duplicate-key violations in cases where users want "insert or update"
semantics (which will honestly probably be everybody). The problem is
that there is no one single way to do this that works across all RDBMS
systems.
-chris
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tomcat.apache.org
For additional commands, e-mail: dev-help@tomcat.apache.org