[infinispan-dev] Hot Rod decoding TRACE logging gone?

Galder Zamarreno galder at redhat.com
Wed Aug 29 07:39:51 EDT 2018


I'd say log it wherever it's less costly.

Originally I wanted this while I was adding new protocol changes to the
Node.js client and the server was hanging. This was due to not sending some
bytes, which couldn't figure out from the logs. I ended up looking at what
the Java client was doing and solved it. At this stage, Intrinsics would
probably have been more interested.

Later on I had an issue with a command and there I was more interested in
finding the value of the remaining parameters at a later stage. So I guess
CacheRequestProcessor would maybe be a better location?

Cheers
Galder

On Mon, Aug 20, 2018 at 3:10 PM Radim Vansa <rvansa at redhat.com> wrote:

> (Routing back to -dev list)
>
> Yes. This could be logged in the most readable format in the operation
> itself, rather than just the bytes as this gets parsed.
> CacheRequestProcessor is probably the best place to do so.
> Or are you interested in seeing that as it comes from the wire? In that
> case I'd suggest the Intrinsics.
>
> Radim
>
> On 08/20/2018 02:43 PM, Galder Zamarreno wrote:
> > Header is just one part of the operation. Individual operation
> > parameters, like version in replaceWithVersion are not logged
> >
> > On Tue, Aug 14, 2018 at 6:22 PM Radim Vansa <rvansa at redhat.com
> > <mailto:rvansa at redhat.com>> wrote:
> >
> >     The header is logged [1] once it gets parsed, but you don't get a log
> >     for each byte being read. You could add logs into Intrinsics [2]
> >     if you
> >     find that useful. You can also see the requests themselves in the
> >     access
> >     log.
> >
> >     Or do you have any particular points where you'd like to get
> >     something
> >     logged on the protocol side? We could add log for entering each state
> >     (log the current path in the grammar tree), though I'd prefer to
> >     activate that just in debug builds since it will increase the code
> >     size
> >     significantly.
> >
> >     Radim
> >
> >     [1]
> >
> https://github.com/infinispan/infinispan/blob/master/server/hotrod/src/main/resources/hotrod.gr#L67
> >     [2]
> >
> https://github.com/infinispan/infinispan/blob/master/server/hotrod/src/main/java/org/infinispan/server/hotrod/Intrinsics.java
> >
> >     On 08/14/2018 05:05 PM, Galder Zamarreno wrote:
> >     > Seems like we've lost all Hot Rod server decoder logging somehow?
> >     >
> >     > There's no trace or anything of any requests coming in :( 🤦‍♂️
> >     >
> >     >
> >     > _______________________________________________
> >     > infinispan-dev mailing list
> >     > infinispan-dev at lists.jboss.org
> >     <mailto:infinispan-dev at lists.jboss.org>
> >     > https://lists.jboss.org/mailman/listinfo/infinispan-dev
> >
> >
> >     --
> >     Radim Vansa <rvansa at redhat.com <mailto:rvansa at redhat.com>>
> >     JBoss Performance Team
> >
>
> --
> Radim Vansa <rvansa at redhat.com>
> JBoss Performance Team
>
> _______________________________________________
> infinispan-dev mailing list
> infinispan-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/infinispan-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/infinispan-dev/attachments/20180829/f892b14d/attachment.html 


More information about the infinispan-dev mailing list