[Engine-devel] Error looking up infinispan/ovirt-engine

Alon Bar-Lev alonbl at redhat.com
Mon May 27 08:01:25 UTC 2013



----- Original Message -----
> From: "Juan Hernandez" <jhernand at redhat.com>
> To: "Alon Bar-Lev" <alonbl at redhat.com>
> Cc: "Moti Asayag" <masayag at redhat.com>, "Yair Zaslavsky" <yzaslavs at redhat.com>, "engine-devel"
> <engine-devel at ovirt.org>
> Sent: Monday, May 27, 2013 10:58:43 AM
> Subject: Re: [Engine-devel] Error looking up infinispan/ovirt-engine
> 
> On 05/26/2013 12:09 PM, Alon Bar-Lev wrote:
> >
> >
> > ----- Original Message -----
> >> From: "Moti Asayag" <masayag at redhat.com>
> >> To: "Alon Bar-Lev" <alonbl at redhat.com>
> >> Cc: "Yair Zaslavsky" <yzaslavs at redhat.com>, "engine-devel"
> >> <engine-devel at ovirt.org>
> >> Sent: Sunday, May 26, 2013 12:19:14 PM
> >> Subject: Re: [Engine-devel] Error looking up infinispan/ovirt-engine
> >>
> >> On 05/26/2013 11:52 AM, Alon Bar-Lev wrote:
> >>>
> >>>
> >>> ----- Original Message -----
> >>>> From: "Moti Asayag" <masayag at redhat.com>
> >>>> To: "Alon Bar-Lev" <alonbl at redhat.com>
> >>>> Cc: "Yair Zaslavsky" <yzaslavs at redhat.com>, "engine-devel"
> >>>> <engine-devel at ovirt.org>
> >>>> Sent: Sunday, May 26, 2013 11:35:40 AM
> >>>> Subject: Re: [Engine-devel] Error looking up infinispan/ovirt-engine
> >>>>
> >>>> On 05/03/2013 07:23 AM, Alon Bar-Lev wrote:
> >>>>>
> >>>>> Hi,
> >>>>>
> >>>>> Thanks... no I did not follow.
> >>>>>
> >>>>> If -P setup is required then the Makefile should have been changed,
> >>>>> else
> >>>>> our production and rpms are broken now! Well... looking at the pom.xml,
> >>>>> it
> >>>>> looks like the jboss standalone is configured with -P setup is used,
> >>>>> which
> >>>>> is yet another hack introduced to developers... As they do not run our
> >>>>> service, and suffer from lack of proper environment and configuration
> >>>>> settings.
> >>>>>
> >>>>> My work[1] is exactly to address this one, to eliminate the need for
> >>>>> hacking the product and use our production startup sequence and
> >>>>> installation. At ./packaging/services/ovirt-engine.xml.in we already
> >>>>> have
> >>>>> the infinispan settings.
> >>>>>
> >>>>> So still need your help.
> >>>>> Thanks,
> >>>>> Alon
> >>>>
> >>>> I've just started facing the same issue.
> >>>> Also noticed this in the server.log:
> >>>>
> >>>> JBAS014775:    New missing/unsatisfied dependencies:
> >>>>        service jboss.mbean.server (missing) dependents: [service
> >>>> jboss.infinispan.ovirt-engine.config]
> >>>>
> >>>> It seems that the generate ovirt-engine.xml under [1] replaces the
> >>>> infinispan:1.1 version as noted on [2] with infinispan:1.2.
> >>>
> >>> I really don't understand the above statement? who/which component have
> >>> 1.2?
> >>>
> >>
> >> the template ovirt-engine.xml.in contains:
> >>
> >> <subsystem xmlns="urn:jboss:domain:infinispan:** 1.1 **"
> >> default-cache-container="ovirt-engine">
> >>
> >> vs
> >>
> >> the generated var/tmp/ovirt-engine/config/ovirt-engine.xml which contains:
> >>
> >> <subsystem xmlns="urn:jboss:domain:infinispan:** 1.2 **"
> >> default-cache-container="ovirt-engine">
> >>
> >> So it used to work with 1.1 (as defined in the template), but the
> >> generated ovirt-engine.xml replaces the version to 1.2 which cause it
> >> not to work.
> >>
> >> So my question is how the engine-service.py decides to generate the
> >> ovirt-engine.xml with "1.2" version, while the template has "1.1" for
> >> infinispan subsystem element ?
> >
> > Hmmmm!!!!! Good catch! I have never thought that it is possible.
> >
> > The template is generated ok with 1.1, after startup jboss re-write its own
> > configuration and change this to 1.2!!!!
> >
> > Juan, is that normal jboss behavior? Editing its own configuration file?
> >
> 
> Yes, it is normal.

Hi,

What do you suggest to do if application requires specific component version and jboss enforce a different version?

If I understand correctly two versions of jboss can enforce a different version of components.

Thanks,
Alon



More information about the Devel mailing list